JP7358981B2 - Information processing program, information processing method, and information processing device - Google Patents

Description

The present invention relates to an information processing program, an information processing method, and an information processing device.

Traditionally, inquiries are received from service users, and the service operator handles the inquiries by referring to accumulated knowledge and sequentially carrying out multiple tasks according to the work flow corresponding to the received inquiry. There is a system that does this. Knowledge is, for example, reference information when performing a task.

As a prior art technique, for example, a weight is assigned to each step constituting a workflow according to the difficulty of the step, and the progress status of the workflow is presented using the assigned weight.

Japanese Patent Application Publication No. 2019-036279

However, in the conventional technology, when tags used when searching for desired data from a data group are attached to each piece of data in the data group, the workload of the worker increases. . For example, if we assign appropriate tags to a huge amount of knowledge one by one from the perspective of searching for knowledge useful for carrying out a task, the workload for the service administrator who assigns the tags will increase. I invite you.

In one aspect, the invention aims to provide appropriate tags to data.

According to one embodiment, vector information representing characteristics of a character string obtained by combining the location, name, and content of data is associated with a tag attached to the data and used to classify the data. Using the attached training data, a learning model that outputs a tag corresponding to the input vector information is generated, and when input of data to be newly assigned a tag is received, the location and name of the target data are generated. Information processing that adds a tag output from the learning model to the target data in response to inputting vector information representing the characteristics of a character string obtained by combining and the content into the generated learning model. A program, an information processing method, and an information processing device are proposed.

According to one aspect, it becomes possible to attach appropriate tags to data.

FIG. 1 is an explanatory diagram showing a configuration example of an information processing system 100 in an embodiment. FIG. 2 is an explanatory diagram (part 1) showing an example of operation of the information processing system 100. FIG. 3 is an explanatory diagram (part 2) showing an example of operation of the information processing system 100. FIG. 4 is an explanatory diagram (part 3) showing an example of operation of the information processing system 100. FIG. 5 is an explanatory diagram showing a configuration example of the information processing apparatus 101. FIG. 6 is a block diagram showing an example of the hardware configuration of the information processing apparatus 101. FIG. 7 is an explanatory diagram showing an example of the stored contents of the flow data 501. FIG. 8 is an explanatory diagram showing an example of the stored contents of the ticket data 502. FIG. 9 is an explanatory diagram showing an example of the stored contents of the knowledge data 503. FIG. 10 is an explanatory diagram showing an example of the storage contents of the utilization history data 504. FIG. 11 is a block diagram showing an example of the hardware configuration of the user device 102. As shown in FIG. FIG. 12 is a block diagram showing an example of the functional configuration of the information processing apparatus 101. FIG. 13 is an explanatory diagram (part 1) showing an example of the operation of displaying knowledge related to a task. FIG. 14 is an explanatory diagram (Part 2) showing an example of the operation of displaying knowledge related to a task. FIG. 15 is an explanatory diagram (part 3) showing an example of the operation of displaying knowledge related to a task. FIG. 16 is an explanatory diagram (part 1) showing an example of the operation of adding a tag to knowledge. FIG. 17 is an explanatory diagram (part 2) showing an example of the operation of adding a tag to knowledge. FIG. 18 is an explanatory diagram (part 3) showing an example of the operation of adding a tag to knowledge. FIG. 19 is an explanatory diagram (Part 4) showing an example of the operation of adding a tag to knowledge. FIG. 20 is an explanatory diagram (part 1) showing an example of the operation of presenting the utilization status of knowledge to the service administrator. FIG. 21 is an explanatory diagram (part 2) showing an example of the operation of presenting the utilization status of knowledge to the service administrator. FIG. 22 is an explanatory diagram (part 3) showing an example of the operation of presenting the utilization status of knowledge to the service administrator. FIG. 23 is an explanatory diagram (part 4) showing an example of the operation of presenting the usage status of knowledge to the service administrator. FIG. 24 is a flowchart showing an example of a ticket processing procedure. FIG. 25 is a flowchart (Part 1) showing an example of a task processing procedure. FIG. 26 is a flowchart (part 2) showing an example of a task processing procedure. FIG. 27 is a flowchart illustrating an example of a knowledge search processing procedure. FIG. 28 is a flowchart illustrating an example of a related knowledge display processing procedure. FIG. 29 is a flowchart illustrating an example of a classification model learning processing procedure. FIG. 30 is a flowchart illustrating an example of a knowledge classification processing procedure. FIG. 31 is a flowchart (Part 1) illustrating an example of a dashboard display processing procedure. FIG. 32 is a flowchart (Part 2) illustrating an example of the dashboard display processing procedure.

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of an information processing program, an information processing method, and an information processing apparatus according to the present invention will be described in detail below with reference to the drawings.

(An example of an information processing method according to an embodiment)
First, an example of the information processing method according to the embodiment will be described using FIGS. 1 to 5. FIG. 1 is an explanatory diagram showing a configuration example of an information processing system 100 in an embodiment. In FIG. 1, an information processing system 100 includes an information processing device 101, a user device 102, an operator device 103, and an administrator device 104.

In the information processing system 100, an information processing device 101, a user device 102, an operator device 103, and an administrator device 104 are connected via a wired or wireless network 110. The network 110 is, for example, a LAN (Local Area Network), a WAN (Wide Area Network), the Internet, or the like.

The information processing system 100 is a system for receiving inquiries from service users and dealing with the received inquiries. A service user is, for example, a user who uses a predetermined service. In the following explanation, a service user may be simply referred to as a "user." The predetermined service is, for example, a business service. The information processing system 100 is, for example, a system for realizing a support center, a service desk, a help desk, etc. for a predetermined service. The inquiry may be, for example, a notification of trouble occurrence automatically issued from another system.

The information processing device 101 is a computer that controls the information processing system 100. The information processing device 101 receives, for example, an inquiry from a service user. For example, the information processing device 101 transmits the received inquiry to the administrator device 104. Thereby, the information processing apparatus 101 allows a service administrator to refer to the inquiry, for example. The service administrator is, for example, an administrator who belongs to the operation side of a predetermined service and manages the service operators. In the following description, the service administrator may be simply referred to as "administrator." The service operator is, for example, an operator who belongs to the operation side of a predetermined service and deals with inquiries. In the following explanation, the service operator may be simply referred to as "operator."

For example, the information processing device 101 receives from the administrator device 104 a work flow for dealing with the inquiry, which is set by the service administrator who referred to the inquiry. A work flow includes, for example, one or more phases. A phase includes, for example, one or more tasks. A task is, for example, work. Specifically, the work flow may be a series of work procedures for responding to inquiries that change display values on a screen. In this case, the work flow sequentially includes, for example, a task of confirming the screen design, a task of confirming a table to be referred to when displaying the screen, and a task of changing table values. For example, the information processing device 101 transmits a work flow to the operator device 103. Thereby, the information processing apparatus 101 allows a service operator to refer to the work flow, perform tasks included in the phases forming the work flow, and respond to inquiries, for example.

The information processing device 101 stores knowledge, for example. Knowledge is, for example, reference information when performing a task. Specifically, the knowledge includes documents created in advance, FAQs, tickets issued in the past, and the like. The document is, for example, maintenance material or a design document at the time of development. A ticket is, for example, management information for managing inquiries. The information processing device 101 transmits knowledge to the operator device 103, for example. Thereby, the information processing apparatus 101 allows a service operator to refer to and utilize knowledge when performing a task. Browsing refers to viewing or confirming the contents of knowledge. Exploitation is the use of knowledge in performing tasks. The information processing device 101 is, for example, a server, a PC (Personal Computer), or the like.

The user device 102 is a computer used by a service user. The user device 102 generates an inquiry according to an operation input by a service user, and transmits it to the information processing device 101, for example. The user device 102 is, for example, a PC, a tablet terminal, a smartphone, or the like.

The operator device 103 is a computer used by a service operator. The operator device 103 receives, for example, a work flow from the information processing device 101, and outputs it so that the service operator can refer to it. The operator device 103 receives, for example, knowledge from the information processing device 101 and outputs it so that the service operator can refer to it. Thereby, the operator device 103 can enable the service operator to perform the tasks included in the phase of forming the work flow while referring to or utilizing knowledge. The operator device 103 is, for example, a PC, a tablet terminal, a smartphone, or the like.

The administrator device 104 is a computer used by a service administrator. The administrator device 104 receives, for example, an inquiry from the information processing device 101, and outputs it so that the service administrator can refer to it. The administrator device 104 sets a work flow for dealing with the inquiry, for example, according to the operation input of the service administrator, and transmits the set work flow to the operator device 103. The administrator device 104 is, for example, a PC, a tablet terminal, a smartphone, or the like.

Although the case where the information processing device 101 and the operator device 103 are different devices has been described here, the present invention is not limited to this. For example, the information processing device 101 may also operate as the operator device 103. Further, although the case where the information processing device 101 and the administrator device 104 are different devices has been described here, the present invention is not limited to this. For example, the information processing device 101 may also operate as the administrator device 104.

Here, conventionally, there is a problem that the work burden on the service operator increases. For example, each time a service operator executes a task, he or she searches for useful knowledge that corresponds to the current task from among a vast amount of miscellaneous knowledge.

At this time, the more knowledge there is, the greater the amount of work required to search for useful knowledge from a huge amount of miscellaneous knowledge, leading to an increase in the work burden placed on the service operator. Specifically, knowledge includes documents created in advance, FAQs, tickets issued in the past, and the like.Even if we limit the types of documents, there are more than 120,000 pieces of knowledge, which can exceed 30 GB. For this reason, the amount of work required for a service operator to search for useful knowledge tends to easily increase.

Furthermore, for example, there may be a case in which the service operator is unable to determine which knowledge is useful among a huge amount of miscellaneous knowledge. For example, if a service operator has relatively little work experience, it is difficult for the service operator to judge which knowledge is useful among a huge amount of miscellaneous knowledge. Specifically, even if the service operator confirms the content of the knowledge, it is not possible to determine whether it is useful or not, or it takes a relatively large amount of effort to determine whether it is useful or not. Tend. In this case, the service operator is likely to carry out the task without referring to the knowledge, leading to an increased workload on the service operator.

In addition, in this case, the service operator is expected to search for useful knowledge from a huge amount of miscellaneous knowledge while consulting with other service operators who have relatively more business experience, This may lead to an increase in the workload placed on the personnel. In addition, in this case, the service operator is likely to carry out the task while consulting with other service operators who have relatively more work experience, without referring to knowledge, and the This may also lead to an increase in the workload.

In addition, appropriate clues have not been set up for service operators to search for useful knowledge from a huge amount of miscellaneous knowledge, which increases the amount of work required to search for useful knowledge. This may lead to an increase in the workload placed on service operators. For example, there may be a case where knowledge is not given an appropriate tag that can be used to determine what kind of task the knowledge is useful for, based on the content of the knowledge.

Furthermore, if service operators with relatively little work experience find it difficult to carry out their tasks, they may be more likely to lose motivation to work or leave their jobs. As a result, work becomes increasingly individualized, which may lead to an increased workload on service operators who have relatively more work experience. Therefore, it is difficult to maintain or improve QCD (Quality, Cost, and Delivery). On the other hand, service managers tend to want to review work flows and knowledge search methods to make it easier to find knowledge and perform tasks.

However, in the past, it was difficult for service administrators to review work flows and knowledge search methods, making it difficult to reduce the workload on service operators, and increasing the workload on service administrators. The problem is that it can lead to For example, the service administrator will have to manually investigate which knowledge was used in the inquiries that have already been handled and review the work flow and knowledge search method, which increases the workload on the service administrator. This may lead to

Furthermore, conventionally, there is a problem in that the time required for a service user to obtain a response from a service operator to an inquiry increases. For example, service users may not be able to receive an answer to their inquiries from the service operator when the service operator, who has relatively more work experience, is on vacation, resulting in an increase in the time required to receive an answer. . In addition, the service user may not be able to receive a response to the inquiry from the service operator at a time when there is no available service operator, resulting in an increase in the time required to obtain the response. Furthermore, it is difficult for service users to refer to vast and miscellaneous knowledge by themselves, for example. In addition, since service users have difficulty in obtaining answers to their inquiries from service operators, their own business operations are adversely affected.

For these reasons, it is thought that there is a desire to make it easier for service operators to find appropriate knowledge when performing tasks. It is also believed that there is a desire to make it easier for service administrators to review work flows and knowledge search methods. It is also believed that there is a desire for service users to be able to quickly obtain answers to their inquiries from service operators. Here, using FIGS. 2 to 4, a description will be given of specifically how the information processing system 100 according to the embodiment is preferably operated in order to satisfy the various demands described above.

2 to 4 are explanatory diagrams showing operational examples of the information processing system 100. For example, as shown in FIG. 2, when a service operator receives an inquiry from a service user, it is preferable that the service operator view the ticket screen 201 and be able to check the progress of the work flow and the tasks to be performed. . The ticket screen 201 displays, for example, a work flow set in advance by a service administrator. Specifically, the ticket screen 201 displays information on each phase forming a work flow and information on tasks included in the phases.

Furthermore, when the service operator clicks on task information, it is preferable to enable the service operator to view a search screen 202 that automatically presents knowledge that is determined to be useful when implementing the task, and to use the knowledge to Preferably, the task is actionable. According to this, it is determined that the request for making it easier for a service operator to find appropriate knowledge when performing a task is likely to be satisfied.

Furthermore, it is preferable that the service operator be able to record the details of how the inquiry was handled, the time required to perform the task, and the like. Furthermore, it is preferable that the service manager be able to confirm the contents of the inquiry, the time required to deal with the inquiry, and the like. According to this, it is determined that the request for making it easier for a service administrator to review work flows and knowledge search methods can be easily satisfied. Next, the description will move on to FIG. 3.

For example, as shown in FIG. 3, it is preferable that the service administrator be able to refer to a list of documents output from the information processing apparatus 101 and to be able to add tags to the documents. Further, it is preferable that the service administrator be able to input a tagged document into the information processing apparatus 101. The information processing apparatus 101 reflects the input manually tagged documents in the tagging list 300. According to this, it is determined that the request for making it easier for a service operator to find appropriate knowledge when performing a task is likely to be satisfied. Here, in order to realize the operations of the service user, service operator, and service administrator illustrated in FIGS. 2 and 3, the information processing system 100 performs a series of operations as illustrated in FIG. is preferred. Next, the description will move on to FIG. 4.

As shown in FIG. 4, the information processing system 100 preferably displays a ticket screen 201 on the operator device 103 in response to an inquiry from a service user or a trouble notification from an ICT system. Furthermore, when the service operator executes a task while referring to the ticket screen 201, the information processing system 100 searches for useful knowledge from the knowledge data 401, and displays the search screen 202 containing the searched knowledge to the operator. It is preferable to display it on the device 103. According to this, the information processing system 100 can make it easier for the service operator to find appropriate knowledge when performing a task.

Here, the information processing system 100 preferably accumulates utilization history data 402 indicating that the service operator referred to or utilized knowledge when performing a task. Further, the information processing system 100 updates the tags associated with the knowledge in the knowledge data, or adds a new tag to the knowledge to which no tag has been added in the knowledge data, based on the accumulated usage history data 402. It is preferable to do so. According to this, the information processing system 100 can make it easier for the service operator to find appropriate knowledge when performing a task.

Further, the information processing system 100 preferably causes the administrator device 104 to display a dashboard screen 410 based on the accumulated usage history data 402. The dashboard screen 410 displays the number of times knowledge has been referenced, the number of times knowledge has been utilized, the rate at which knowledge has been utilized, the time required to perform a task, the time required to deal with an inquiry, etc. . According to this, the information processing system 100 can visualize the usage status of knowledge and enable the service manager to understand it. The usage status of knowledge includes, for example, a situation where knowledge is not referenced and therefore is not utilized, a situation where knowledge is referred to but not utilized, and a situation where knowledge is referred to and utilized. Therefore, the information processing system 100 can make it easier for the service administrator to review the work flow and knowledge search method.

Next, using FIG. 5, a configuration example of the information processing apparatus 101 in the information processing system 100 according to the embodiment that implements the operation examples shown in FIGS. 2 to 4 will be described. FIG. 5 is an explanatory diagram showing a configuration example of the information processing apparatus 101. As shown in FIG. 5, the information processing device 101 has flow data 501, ticket data 502, knowledge data 503, usage history data 504, and a tagging list 505. The information processing device 101 also includes a search model 506 and a classification model 507.

The information processing device 101 causes the operator device 103 to display a ticket screen 511 based on the flow data 501 and ticket data 502. The information processing device 101 displays a search screen 512. The information processing device 101 acquires the service operator's operation input on the search screen 512 from the operator device 103, and uses the search model 506 based on the knowledge data 503 to determine whether the service operator's operation input is useful when executing the task. Search for knowledge. The information processing device 101 causes the operator device 103 to display a search screen 512 that includes the searched knowledge.

The information processing device 101 acquires the service operator's operation input on the search screen 512 from the operator device 103 and updates the utilization history data 504. The information processing device 101 generates a classification model 507 based on the usage history data 504 and the tagging list 505. The information processing apparatus 101 automatically adds tags to knowledge using the generated classification model 507. The information processing device 101 causes the administrator device 104 to display a dashboard screen 513 based on the usage history data 504. The dashboard screen 513 shows the usage status of knowledge.

As described above, the information processing apparatus 101 allows the service operator to respond to inquiries according to the work flow. The information processing device 101 allows a service operator to refer to knowledge that is determined to be useful when executing a task, making it easier to execute the task, and reducing the workload on the service operator. It is possible to reduce the Therefore, when carrying out a task, the service operator can easily find appropriate knowledge from a huge amount of miscellaneous knowledge, making it easier to carry out the task, and reducing the workload.

In addition, even if a service operator has relatively little work experience, it will be easier to find appropriate knowledge from a huge amount of miscellaneous knowledge when carrying out a task, and it will be easier for service operators to find appropriate knowledge from a vast amount of miscellaneous knowledge when carrying out a task. Tasks can be carried out without consulting other people. Therefore, the information processing device 101 can reduce the workload of other service operators. Further, the information processing device 101 can make it easier for the service operator to perform tasks, can suppress a decline in work motivation, and can make it easier to maintain or improve QCD.

Further, the information processing device 101 can make it easier for the service operator to utilize appropriate knowledge, and can improve the probability that the service administrator will not have to review the work flow and knowledge search method. . Therefore, the information processing apparatus 101 can reduce the workload placed on the service administrator. In addition, the information processing device 101 can make it easier for service operators to carry out tasks based on appropriate knowledge, and reduce the time required for service users to receive answers to their inquiries from service operators. It is possible to reduce the

Furthermore, the information processing apparatus 101 can automatically add tags to knowledge, and can accurately search for knowledge that is determined to be useful when performing a task, and make it available for reference by a service operator. Therefore, when carrying out a task, the service operator can easily find appropriate knowledge from a huge amount of miscellaneous knowledge, making it easier to carry out the task, and reducing the workload. Further, the information processing apparatus 101 can make it easier for the service administrator to review the work flow and knowledge search method using the dashboard screen 513. Therefore, the information processing apparatus 101 can reduce the workload placed on the service administrator. In addition, the service administrator can review the work flow and knowledge search method to reduce the work burden on the service operator.

(Example of hardware configuration of information processing device 101)
Next, an example of the hardware configuration of the information processing device 101 will be described using FIG. 6.

FIG. 6 is a block diagram showing an example of the hardware configuration of the information processing apparatus 101. In FIG. 6, the information processing device 101 includes a CPU (Central Processing Unit) 601, a memory 602, a network I/F (Interface) 603, a recording medium I/F 604, and a recording medium 605. Further, each component is connected to each other by a bus 600.

Here, the CPU 601 is in charge of overall control of the information processing apparatus 101. The memory 602 includes, for example, a ROM (Read Only Memory), a RAM (Random Access Memory), and a flash ROM. Specifically, for example, a flash ROM or ROM stores various programs, and a RAM is used as a work area for the CPU 601. The program stored in the memory 602 is loaded into the CPU 601 and causes the CPU 601 to execute the coded processing.

Network I/F 603 is connected to network 110 through a communication line, and connected to other computers via network 110. The network I/F 603 serves as an internal interface with the network 110, and controls data input/output from other computers. The network I/F 603 is, for example, a modem or a LAN adapter.

The recording medium I/F 604 controls reading/writing of data to/from the recording medium 605 under the control of the CPU 601 . The recording medium I/F 604 is, for example, a disk drive, an SSD (Solid State Drive), a USB (Universal Serial Bus) port, or the like. The recording medium 605 is a nonvolatile memory that stores data written under the control of the recording medium I/F 604. The recording medium 605 is, for example, a disk, a semiconductor memory, a USB memory, or the like. The recording medium 605 may be removable from the information processing apparatus 101.

In addition to the components described above, the information processing device 101 may include, for example, a keyboard, a mouse, a display, a printer, a scanner, a microphone, a speaker, and the like. Further, the information processing apparatus 101 may have a plurality of recording medium I/Fs 604 and recording media 605. Further, the information processing apparatus 101 does not need to have the recording medium I/F 604 or the recording medium 605.

(Stored content of flow data 501)
Next, an example of the stored contents of the flow data 501 will be described using FIG. 7. The flow data 501 is realized, for example, by a storage area such as the memory 602 or the recording medium 605 of the information processing apparatus 101 shown in FIG.

FIG. 7 is an explanatory diagram showing an example of the stored contents of the flow data 501. As shown in FIG. 7, the flow data 501 has fields of flow name, phase ID, phase name, phase description, task ID, task name, and task description. In the flow data 501, task data is stored as a record 700-a by setting information in each field for each task. a is any integer.

A flow name that identifies a work flow for dealing with an inquiry is set in the flow name field. A phase ID that identifies one of the phases forming the work flow is set in the phase ID field. A phase name that identifies one of the phases forming the work flow is set in the phase name field. In the phase description field, a phase description indicating the work content of one of the phases forming the work flow is set.

A task ID that identifies any task included in any phase forming the work flow is set in the task ID field. A task name that identifies any task included in any phase forming the work flow is set in the task name field. In the task description field, a task description indicating the work content of any task included in any phase forming the work flow is set.

(Stored contents of ticket data 502)
Next, an example of the stored contents of the ticket data 502 will be described using FIG. 8. The ticket data 502 is realized, for example, by a storage area such as the memory 602 or the recording medium 605 of the information processing apparatus 101 shown in FIG.

FIG. 8 is an explanatory diagram showing an example of the stored contents of the ticket data 502. As shown in FIG. 8, the ticket data 502 includes a ticket ID, a question text, an answer text, the number of uses (tickets), the number of uses (FAQ), the number of uses (documents), response time, and tasks. It has a field with corresponding time for each. The response time field for each task is, for example, a task 1 response time field, a task 2 response time field, and the like. The ticket data 502 is stored as a record 1100-b by setting information in each field for each inquiry. b is an arbitrary integer.

In the ticket ID field, a ticket ID that identifies the ticket, which is management information for managing inquiries, is set. A question text that is an inquiry is set in the question text field. In the answer text field, a reply text to the inquiry is set.

The number of uses (tickets) field is set to the number of times past tickets, which serve as knowledge, have been used when dealing with an inquiry. The number of uses (FAQ) field is set to the number of times the FAQ, which is knowledge, is used when dealing with an inquiry. The number of uses (documents) field is set to the number of times documents serving as knowledge are used when dealing with an inquiry.

In the response time field, the time required to respond to the inquiry is set. In the response time field for each task, the time required to perform the task is set. Specifically, the time required to complete the i-th task is set in the task i response time field. i is any integer.

The ticket data 502 may further include fields for the number of references (ticket), the number of references (FAQ), and the number of references (document). The number of references (tickets) field is set to the number of times past tickets that serve as knowledge are referenced when dealing with an inquiry. The number of references (FAQ) field is set to the number of times the FAQ serving as knowledge is referred to when dealing with an inquiry. The number of references (documents) field is set to the number of times a document serving as knowledge is referred to when dealing with an inquiry.

(Memory content of knowledge data 503)
Next, an example of the stored contents of the knowledge data 503 will be explained using FIG. 9. The knowledge data 503 is realized, for example, by a storage area such as the memory 602 or the recording medium 605 of the information processing apparatus 101 shown in FIG.

FIG. 9 is an explanatory diagram showing an example of the stored contents of the knowledge data 503. As shown in FIG. 9, the knowledge data 503 has fields of knowledge ID, knowledge title, knowledge content, knowledge path, and knowledge tag. The knowledge data 503 is stored as a record 900-c by setting information in each field for each knowledge. c is any integer.

A knowledge ID for identifying knowledge is set in the knowledge ID field. The title given to the knowledge is set in the knowledge title field. The knowledge content is set in the knowledge content field. In the example of FIG. 9, description of the contents of knowledge is omitted.

For example, if the knowledge is a past ticket, the content of the knowledge is the content of the inquiry included in the ticket. For example, if the knowledge is FAQ, the content of the knowledge is the content of the question text. For example, if the knowledge is a document, the content of the knowledge is the content of the document. The document is, for example, maintenance material or a design document at the time of development.

Path information indicating the location of knowledge is set in the knowledge path field. The tag given to the knowledge is set in the knowledge tag field. If the knowledge has not yet been tagged, the tag field of the knowledge is set to (none). The knowledge tag is assigned manually or automatically, for example.

(Memory content of usage history data 504)
Next, an example of the storage contents of the utilization history data 504 will be explained using FIG. 10. The usage history data 504 is realized, for example, by a storage area such as the memory 602 or the recording medium 605 of the information processing apparatus 101 shown in FIG.

FIG. 10 is an explanatory diagram showing an example of the storage contents of the utilization history data 504. As shown in FIG. 10, the usage history data 504 includes a ticket ID, a question text, a flow name, a phase ID, a phase name, a task ID, a task name, a task description, a knowledge ID, It has fields for display order, session ID, and usage method. The usage history data 504 is stored as a record 1000-d by setting information in each field for each ticket. d is an arbitrary integer.

A ticket ID for identifying the ticket is set in the ticket ID field. A ticket is management information for managing inquiries. A question text, which is an inquiry managed by a ticket, is set in the question text field.

In the flow name field, a flow name that identifies the work flow used when dealing with the inquiry is set. A phase ID that identifies one of the phases forming the work flow is set in the phase ID field. A phase name that identifies one of the phases forming the work flow is set in the phase name field.

A task ID that identifies any task included in any phase forming the work flow is set in the task ID field. A task name that identifies any task included in any phase forming the work flow is set in the task name field. In the task description field, a task description indicating the work content of any task included in any phase forming the work flow is set.

A knowledge ID for identifying knowledge is set in the knowledge ID field. In the display order field, the display order in which knowledge is displayed on the search screen 512 when dealing with an inquiry managed by a ticket is set. In the session ID field, a session ID generated in response to displaying the search screen 512 when dealing with an inquiry managed by a ticket is set. The session ID is information for grouping knowledge that has been searched, referenced, and utilized on the same search screen 512.

Flag information indicating whether the knowledge has been referred to or utilized is set in the utilization method field. For example, in the case where the knowledge is referred to but not utilized, Clicked is set in the utilization method field. Specifically, Clicked is set in the utilization method field when link information displayed in association with knowledge is clicked. For example, if the knowledge is referenced and utilized, "Utilized" is set in the utilization method field. Specifically, "Utilized" is set in the utilization method field when a "utilization button" displayed in association with knowledge is clicked.

Various types of data such as flow data 501, ticket data 502, knowledge data 503, and utilization history data 504 may not have any of the above-mentioned fields, or may have other fields other than the above-mentioned fields. It may also have fields.

(Example of hardware configuration of user device 102)
Next, an example of the hardware configuration of the user device 102 included in the information processing system 100 shown in FIG. 1 will be described using FIG. 11.

FIG. 11 is a block diagram showing an example of the hardware configuration of the user device 102. As shown in FIG. In FIG. 11, the user device 102 includes a CPU 1101, a memory 1102, a network I/F 1103, a recording medium I/F 1104, a recording medium 1105, a display 1106, and an input device 1107. Further, each component is connected to each other by a bus 1100.

Here, the CPU 1101 is in charge of overall control of the user device 102. Memory 1102 includes, for example, ROM, RAM, flash ROM, and the like. Specifically, for example, a flash ROM or ROM stores various programs, and a RAM is used as a work area for the CPU 1101. The program stored in the memory 1102 is loaded into the CPU 1101 and causes the CPU 1101 to execute the coded processing.

Network I/F 1103 is connected to network 110 through a communication line, and connected to other computers via network 110. The network I/F 1103 serves as an internal interface with the network 110, and controls data input/output from other computers. The network I/F 1103 is, for example, a modem or a LAN adapter.

The recording medium I/F 1104 controls reading/writing of data to/from the recording medium 1105 under the control of the CPU 1101 . The recording medium I/F 1104 is, for example, a disk drive, an SSD, a USB port, or the like. The recording medium 1105 is a nonvolatile memory that stores data written under the control of the recording medium I/F 1104. The recording medium 1105 is, for example, a disk, a semiconductor memory, a USB memory, or the like. The recording medium 1105 may be removable from the user device 102.

Display 1106 displays data such as cursors, icons, or toolboxes, as well as documents, images, and functional information. The display 1106 is, for example, a CRT (Cathode Ray Tube), a liquid crystal display, an organic EL (Electroluminescence) display, or the like. The input device 1107 has keys for inputting characters, numbers, various instructions, etc., and inputs data. The input device 1107 may be a keyboard, a mouse, or the like, or may be a touch panel type input pad, a numeric keypad, or the like.

In addition to the components described above, the user device 102 may include, for example, a printer, a scanner, a microphone, a speaker, and the like. Further, the user device 102 may have a plurality of recording medium I/Fs 1104 and recording media 1105. Further, the user device 102 does not need to have the recording medium I/F 1104 or the recording medium 1105.

(Example of hardware configuration of operator device 103)
The hardware configuration example of the operator device 103 included in the information processing system 100 shown in FIG. 1 is the same as the hardware configuration example of the user device 102 shown in FIG. 11, and therefore a description thereof will be omitted.

(Example of hardware configuration of administrator device 104)
The hardware configuration example of the administrator device 104 included in the information processing system 100 shown in FIG. 1 is the same as the hardware configuration example of the user device 102 shown in FIG. 11, and therefore a description thereof will be omitted.

(Functional configuration example of information processing device 101)
Next, an example of the functional configuration of the information processing device 101 will be described using FIG. 12.

FIG. 12 is a block diagram showing an example of the functional configuration of the information processing apparatus 101. The information processing device 101 includes a storage unit 1200, an acquisition unit 1201, a learning unit 1202, an assignment unit 1203, a search unit 1204, a recording unit 1205, an analysis unit 1206, and an output unit 1207.

The storage unit 1200 is realized by, for example, a storage area such as the memory 602 or the recording medium 605 shown in FIG. 6. Although a case will be described below in which the storage unit 1200 is included in the information processing apparatus 101, the present invention is not limited to this. For example, there may be a case where the storage unit 1200 is included in a device different from the information processing device 101, and the storage contents of the storage unit 1200 can be referenced from the information processing device 101.

The acquisition unit 1201 to output unit 1207 function as an example of a control unit. Specifically, the acquisition unit 1201 to the output unit 1207 execute the program by causing the CPU 601 to execute a program stored in a storage area such as the memory 602 or the recording medium 605 shown in FIG. This function is realized by The processing results of each functional unit are stored in a storage area such as the memory 602 or the recording medium 605 shown in FIG. 6, for example.

The storage unit 1200 stores various information that is referenced or updated in the processing of each functional unit. The storage unit 1200 stores, for example, inquiries from service users.

The storage unit 1200 stores, for example, knowledge. Knowledge is, for example, reference information used by a service operator when the service operator performs a task in response to an inquiry from a service user. A service operator is a worker who performs a task. The knowledge is referenced or utilized by a service operator, for example. Specifically, the knowledge includes documents created in advance, FAQs, tickets, and the like. A ticket is, for example, management information for managing inquiries.

Specifically, the storage unit 1200 stores the contents of knowledge. The storage unit 1200 stores, for example, knowledge and tags attached to the knowledge in association with each other. A tag is, for example, information used to classify knowledge. The tag is attached manually or by the attaching unit 1203, for example. The storage unit 1200 stores, for example, the location and name of knowledge. The location and name of knowledge are expressed by the full path of the knowledge.

The storage unit 1200 stores reference results, for example. The reference record indicates, for example, that knowledge has been referenced. Specifically, the reference record refers to knowledge that is searched when the service operator executes a task in response to an inquiry from a service user, and is referenced by the service operator when the service operator executes the task. will be stored if the Specifically, the storage unit 1200 stores reference results in association with tasks.

The storage unit 1200 stores, for example, utilization results. The utilization record indicates, for example, that knowledge has been referenced and utilized. Utilization results specifically refer to the knowledge searched when a service operator executes a task in response to an inquiry from a service user, and the knowledge that is utilized by the service operator when the service operator executes the task. will be stored if the Specifically, the storage unit 1200 stores utilization results in association with tasks.

The acquisition unit 1201 acquires various information used in processing of each functional unit. The acquisition unit 1201 stores the acquired various information in the storage unit 1200 or outputs it to each functional unit. Further, the acquisition unit 1201 may output various information stored in the storage unit 1200 to each functional unit. The acquisition unit 1201 acquires various information based on, for example, operation input by a service operator or a service administrator. The acquisition unit 1201 may receive various information from a device different from the information processing device 101, for example.

The acquisition unit 1201 acquires an inquiry. The acquisition unit 1201 receives, for example, an inquiry from the user device 102. The acquisition unit 1201 may acquire an inquiry based on an operation input by a service operator or a service administrator.

The acquisition unit 1201 acquires a tag to be added to knowledge. The acquisition unit 1201 receives, for example, a tag to be added to knowledge from the operator device 103 or the administrator device 104. The acquisition unit 1201 may acquire a tag to be added to knowledge based on an operational input from a service operator or a service administrator.

The acquisition unit 1201 receives input of knowledge to which a new tag is to be added. The acquisition unit 1201 receives, for example, knowledge to which a new tag is to be added from the operator device 103 or the administrator device 104. The acquisition unit 1201 may acquire knowledge to which a new tag is to be added based on an operation input from a service operator or a service administrator.

The acquisition unit 1201 may receive a start trigger that starts processing of any functional unit. The start trigger is, for example, a predetermined operation input by a service operator or a service administrator. The start trigger may be, for example, receiving predetermined information from another computer. The start trigger may be, for example, that any functional unit outputs predetermined information.

The acquisition unit 1201 receives, for example, a regularly issued signal as a start trigger for starting processing between the learning unit 1202 and the providing unit 1203. The acquisition unit 1201 receives, for example, the acquisition of an inquiry as a trigger for starting processing between the search unit 1204 and the recording unit 1205. The acquisition unit 1201 receives, for example, a periodically issued signal as a trigger for starting the processing of the analysis unit 1206.

The learning unit 1202 generates a learning model. The learning model is, for example, a model that outputs a tag corresponding to input vector information. Specifically, the learning model is a classification model 1610, which will be described later. Vector information is, for example, information obtained from knowledge. The learning unit 1202 performs learning using, for example, learning data that associates vector information representing the characteristics of a character string obtained by combining the location, name, and content of knowledge with a tag attached to knowledge. Generate the model. For example, the location, name, and content of the knowledge may be combined as they are, or the location, name, and content of the knowledge may be processed and then combined.

Specifically, the learning unit 1202 arranges bit information indicating whether each of a plurality of predetermined words appears in a character string obtained by combining the knowledge location, name, and content. , generate vector information. The learning unit 1202 then generates a learning model using learning data in which the generated vector information is associated with the tag given to the knowledge. Thereby, the learning unit 1202 can generate a learning model that can estimate tags that are determined to be appropriate to be added to knowledge.

For example, the learning unit 1202 associates vector information representing the characteristics of a character string obtained by combining the location, name, and content of knowledge used when performing a task with a tag attached to knowledge. A learning model may be generated using the attached training data. Specifically, the learning unit 1202 uses vector information representing the characteristics of a character string obtained by combining the location, name, and content of the knowledge utilized when performing the task, and the tag attached to the knowledge. A learning model may be generated using learning data that is associated with. As a result, the learning unit 1202 can use learning data that includes tags that have been actually used and are determined to make knowledge searchable appropriately, and generate a learning model that can accurately estimate tags. be able to.

The learning unit 1202, for example, combines vector information representing the characteristics of a character string obtained by combining the location, name, and content of knowledge to which a tag specified by a service operator or service administrator is attached, and the tag. A learning model may be generated using learning data associated with . As a result, the learning unit 1202 can use learning data that includes tags that are determined to be appropriately searchable because they are specified by the service operator or service administrator, and can estimate tags with high accuracy. A learning model can be generated. Although the case where the learning unit 1202 generates vector information has been described here, the present invention is not limited to this. For example, the learning unit 1202 may receive vector information from another computer. Further, for example, the learning unit 1202 may acquire vector information based on an operation input by an operator of the information processing apparatus 101.

The adding unit 1203 adds a tag to the target knowledge. For example, the assigning unit 1203 assigns the tag obtained as a result of inputting vector information representing the characteristics of a character string obtained by combining the location, name, and content of the target knowledge into the generated learning model. Grant to knowledge.

For example, the adding unit 1203 arranges bit information indicating whether each word of a plurality of words appears in a character string obtained by combining the location, name, and content of the target knowledge, and generates vector information. generate. Then, the adding unit 1203 adds a tag obtained as a result of inputting the generated vector information to the generated learning model to the target knowledge. Thereby, the adding unit 1203 can add to the knowledge a tag that is determined to be appropriate to add to the knowledge.

The assigning unit 1203 assigns, to knowledge, a tag assigned to a task performed using knowledge among a plurality of tagged tasks. For example, the assigning unit 1203 assigns, to knowledge, a tag assigned to a task performed by referring to or utilizing knowledge, among a plurality of tasks assigned tags. Thereby, the adding unit 1203 can automatically add tags to knowledge even before generating a learning model, and can reduce the workload. Further, the adding unit 1203 can add tags to the knowledge used when performing a task, and can generate learning data, making it possible to generate a learning model that can estimate tags with high accuracy. can.

The adding unit 1203 may add a tag specified by the service operator or service administrator to knowledge specified by the service operator or service administrator. As a result, the assigning unit 1203 can generate learning data that includes tags that are determined to be appropriately searchable because the service operator or service administrator has specified them, and can accurately assign tags. It is possible to generate a learning model that can be estimated.

The search unit 1204 generates narrowing conditions. The narrowing condition is, for example, one or more tags. Specifically, the narrowing conditions are generated from the name of the task to be performed in response to the inquiry. Specifically, the search unit 1204 generates narrowing conditions from the task name. More specifically, the search unit 1204 generates one or more tags as narrowing conditions based on the noun extracted from the name of the task. The tag is, for example, a noun. Thereby, the search unit 1204 can generate narrowing conditions for improving the accuracy of searching for knowledge used when performing a task.

The search unit 1204 searches for knowledge based on the service user's operation input on the search screen 512. The search screen 512 is displayed on the user device 102 by, for example, a display unit. The search screen 512 is, for example, a screen for searching for knowledge used when performing a task. The search unit 1204 searches for knowledge related to the first task from among a plurality of knowledge based on the first narrowing condition, for example. The first narrowing condition is generated based on the first task, for example. Specifically, the first narrowing condition is generated from the name of the first task. Specifically, the search unit 1204 searches for knowledge to which a tag serving as the first narrowing condition is attached from among the plurality of knowledge, as knowledge related to the first task. Thereby, the search unit 1204 can enable the service operator to refer to the knowledge used when performing the task.

The search unit 1204 searches for knowledge related to the first task from among the plurality of knowledge based on, for example, the inquiry, the work content of the first task, and the first narrowing condition. Specifically, the search unit 1204 sets a combination of the inquiry and the work content of the first task in the search sentence. Then, the search unit 1204 searches for knowledge that corresponds to the set search statement among the knowledge that is tagged with the first narrowing condition from among the plurality of knowledge, as knowledge related to the first task. do. Thereby, the search unit 1204 can enable the service operator to refer to the knowledge used when performing the task.

For example, the search unit 1204 determines whether any knowledge related to the searched first task was used by the service operator when executing the first task. Here, when the search unit 1204 determines that any of the knowledge has been used by the service operator, the search unit 1204 sets a second narrowing condition that is stored in association with any of the knowledge and under which any of the knowledge is searched. Identify. The second narrowing condition is, for example, a tag stored in association with any piece of knowledge. Then, the search unit 1204 searches for knowledge related to the second task from among the plurality of knowledge based on the specified second narrowing condition. Specifically, the search unit 1204 searches for knowledge related to the second task, which is tagged with a tag included in the second narrowing condition, from among a plurality of tagged pieces of knowledge. Thereby, the search unit 1204 can enable the service operator to refer to the knowledge used when performing the task.

For example, the search unit 1204 determines whether any knowledge related to the searched first task was used by the service operator when executing the first task. Here, when the search unit 1204 determines that any of the knowledge has been used by the service operator, the search unit 1204 sets a second narrowing condition that is stored in association with any of the knowledge and under which any of the knowledge is searched. conditions that are included in the first narrowing-down condition and not included in the first narrowing-down condition. The second narrowing condition is, for example, a tag stored in association with any piece of knowledge. Then, the search unit 1204 searches for knowledge related to the second task from among the plurality of knowledge based on the specified conditions. Specifically, the search unit 1204 selects a second piece of knowledge to which a tag is included in the second narrowing condition and not included in the first narrowing condition, from among a plurality of tagged pieces of knowledge. Search for knowledge related to a task. Thereby, the search unit 1204 can enable the service operator to refer to the knowledge used when performing the task.

The search unit 1204 is configured to search based on, for example, conditions included in the second narrowing conditions but not included in the first narrowing conditions, the inquiry, the work content of the second task, and the third narrowing conditions. , searches for knowledge related to the second task from among the plurality of knowledge. The third narrowing condition is generated based on the second task, for example. Specifically, the third narrowing condition is generated from the name of the second task. Specifically, the search unit 1204 sets a combination of the inquiry and the work content of the second task in the search sentence. Then, the search unit 1204 selects the knowledge that corresponds to the set search statement from among the knowledge that is tagged with a tag that is included in the second narrowing condition but not included in the first narrowing condition. is searched as knowledge related to the second task. Thereby, the search unit 1204 can enable the service operator to refer to the knowledge used when performing the task.

The recording unit 1205 acquires a different search identifier each time the search screen 512 is displayed. The search identifier is an identifier for grouping one or more pieces of knowledge searched on the same search screen 512 and selected by the service operator. Acquisition is, for example, generation. The acquisition timing is, for example, when the search screen 512 is displayed. The acquisition timing may be, for example, when the search screen 512 is displayed and the search identifier is used.

The recording unit 1205 obtains, for example, by generating a different search identifier each time the search screen 512 is displayed. For example, the recording unit 1205 may change the search identifier when transitioning from the search screen 512 to a screen for purposes other than search, and when subsequently displaying the search screen 512 anew, the changed search identifier may be changed. A search identifier may also be acquired. For example, the recording unit 1205 may change the search identifier when closing the search screen 512, and then acquire the changed search identifier when displaying the search screen 512 anew. It's okay. Thereby, the recording unit 1205 can group one or more pieces of knowledge that are likely to be used at the same timing and for the same purpose.

The recording unit 1205 stores the search information acquired when displaying the search screen 512 for one or more pieces of knowledge searched and selected on the search screen 512 for searching knowledge used when performing the third task. Assign an identifier. Thereby, the recording unit 1205 can group one or more pieces of knowledge that are likely to be used at the same timing and for the same purpose.

The recording unit 1205 further assigns a task identifier for identifying the third task to one or more pieces of knowledge. Thereby, the recording unit 1205 can group one or more pieces of knowledge that are likely to be used at the same timing and for the same purpose.

After implementing the third task, if any knowledge is found on the search screen 512 for searching for knowledge to be used when implementing the fourth task, the recording unit 1205 further specifies other knowledge. do. The third task and the fourth task may be the same task. For example, the recording unit 1205 identifies other knowledge that has been given the same search identifier as any of the knowledge. Then, the recording unit 1205 sets the identified other knowledge in addition to any of the knowledges as a search result. Thereby, the recording unit 1205 can output the searched knowledge and other knowledge that is likely to be used at the same timing and for the same purpose as the search result. The recording unit 1205 can make it easier for the service operator to refer to useful knowledge when performing a task.

For example, the recording unit 1205 identifies other knowledge to which the same search identifier as at least one of the one or more search identifiers is attached to any one of the knowledges. Then, the recording unit 1205 sets the identified other knowledge in addition to any of the knowledges as a search result. Thereby, the recording unit 1205 can output the searched knowledge and other knowledge that is likely to be used at the same timing and for the same purpose as the search result. The recording unit 1205 can make it easier for the service operator to refer to useful knowledge when performing a task.

For example, the recording unit 1205 records other knowledge that has been assigned the same search identifier as the search identifier assigned to any knowledge, and that has been assigned the same task identifier as the task identifier assigned to any of the knowledge. Identify. Then, the recording unit 1205 sets the identified other knowledge in addition to any of the knowledges as a search result. Thereby, the recording unit 1205 can output the searched knowledge and other knowledge that is likely to be used at the same timing and for the same purpose as the search result. The recording unit 1205 can make it easier for the service operator to refer to useful knowledge when performing a task.

For example, the recording unit 1205 identifies other knowledge that has at least one of the same assigned search identifiers and that has at least one of the assigned task identifiers that is the same as any of the pieces of knowledge. do. For example, the other knowledge is assigned the same search identifier as at least one of the one or more search identifiers assigned to any of the knowledges. Further, the other knowledge is assigned, for example, the same task identifier as at least one of the one or more task identifiers assigned to any of the knowledge. Then, the recording unit 1205 sets the identified other knowledge in addition to any of the knowledges as a search result. Thereby, the recording unit 1205 can output the searched knowledge and other knowledge that is likely to be used at the same timing and for the same purpose as the search result. The recording unit 1205 can make it easier for the service operator to refer to useful knowledge when performing a task.

When the service operator refers to the knowledge searched when the service operator executes the task in response to the inquiry, the analysis unit 1206 associates the reference history indicating that the knowledge was referenced with the task. and stored in the storage unit 1200. Thereby, the analysis unit 1206 can accumulate reference records in order to enable the service administrator to consider whether or not the situation is such that the knowledge can be appropriately referred to or utilized.

When the service operator utilizes the knowledge retrieved when the service operator executes a task in response to an inquiry, the analysis unit 1206 records the utilization record showing that the knowledge was referenced and utilized. , are stored in the storage unit 1200 in association with the tasks. For example, when link information displayed in association with knowledge is used, the analysis unit 1206 determines that the knowledge has been referred to by the service operator. The link information is, for example, an address. Thereby, the analysis unit 1206 can accumulate utilization results in order to enable the service administrator to consider whether or not the knowledge can be appropriately referred to or utilized.

The analysis unit 1206 stores, in the storage unit 1200, the display order of knowledge in association with the task when displaying the knowledge retrieved when the service operator executes the task in response to the inquiry. For example, when the analysis unit 1206 receives an operation indicating that the knowledge has been utilized in the reception area displayed in association with the knowledge, the analysis unit 1206 determines that the knowledge has been utilized by the service operator when performing the task. The reception area is, for example, an area that accepts an operation indicating that knowledge has been utilized. Thereby, the analysis unit 1206 can accumulate display rankings in order to enable the service administrator to consider whether or not the situation is such that the knowledge can be appropriately referred to or utilized.

The analysis unit 1206 sets, in the analysis result to be displayed, information indicating the percentage of handled queries belonging to each of a plurality of classifications based on predetermined criteria, in which knowledge is utilized. The plurality of classifications are, for example, a plurality of periods divided into certain units. The period is, for example, a day, a week, a month, or the like.

For example, the analysis unit 1206 sets, in the analysis result to be displayed, information indicating the ratio of knowledge of each of the plurality of types being utilized among the handled inquiries belonging to each category. For example, the analysis unit 1206 sets, in the analysis results to be displayed, information indicating the percentage of knowledge being utilized and the percentage of knowledge not being utilized among the handled inquiries belonging to each category. Thereby, the analysis unit 1206 can set the analysis results to be displayed in order to enable the service administrator to consider whether or not the situation is such that the knowledge can be appropriately referred to or utilized.

For example, the analysis unit 1206 further sets statistical values of the time required for inquiries that have been handled using knowledge and statistical values of the time required for inquiries that have been handled without using knowledge in the analysis results to be displayed. do. For example, the analysis unit 1206 further calculates statistical values of the time required for inquiries that have been handled using knowledge of each of the plurality of types, and statistical values of the time required for inquiries that have been handled without using knowledge. Set the analysis results to be displayed. Thereby, the analysis unit 1206 can set the analysis results to be displayed in order to enable the service administrator to consider whether or not the situation is such that the knowledge can be appropriately referred to or utilized.

When one of the plurality of classifications is selected, the analysis unit 1206 determines that it is requested to display information regarding a specific task in response to a handled inquiry belonging to one of the selected classifications. .

When requested to display information regarding a specific task, the analysis unit 1206 further sets analysis results to be displayed. For example, the analysis unit 1206 calculates the number and utilization of knowledge referenced when performing a specific task, based on the reference record regarding the specific task and the utilization record regarding the specific task, which are stored in the storage unit 1200. Set the number of knowledge to be displayed in the analysis results. Thereby, the analysis unit 1206 can set the analysis results to be displayed in order to enable the service administrator to consider whether or not the situation is such that the knowledge can be appropriately referred to or utilized.

When requested to display information regarding a specific task, the analysis unit 1206 further sets analysis results to be displayed. For example, the analysis unit 1206 sets the display order of knowledge referenced or utilized when performing a specific task to the analysis result to be displayed, based on the display order regarding the specific task stored in the storage unit 1200. do. Thereby, the analysis unit 1206 can set the analysis results to be displayed in order to enable the service administrator to consider whether or not the situation is such that the knowledge can be appropriately referred to or utilized.

The output unit 1207 outputs the processing results of any of the functional units. The output format is, for example, displaying on a display, printing out to a printer, transmitting to an external device via network I/F 603, or storing in a storage area such as memory 602 or recording medium 605. Thereby, the output unit 1207 can notify a service operator or a service administrator of the processing results of any of the functional units, thereby improving the usability of the information processing apparatus 101.

The output unit 1207 outputs search results, for example. Specifically, the output unit 1207 outputs information on the searched knowledge. The knowledge information includes, for example, link information to the knowledge, the name of the knowledge, the contents of the knowledge, and the like. The output destination is, for example, the operator device 103. Thereby, the output unit 1207 can enable the service operator to understand the searched knowledge information.

Specifically, the output unit 1207 displays the searched knowledge in association with link information to the knowledge. The output destination is, for example, the operator device 103. Thereby, the output unit 1207 can enable the service operator to understand the searched knowledge information.

Specifically, the output unit 1207 displays the searched knowledge in association with a reception area that accepts an operation indicating that the knowledge has been utilized. The output destination is, for example, the operator device 103. Thereby, the output unit 1207 can enable the service operator to understand the searched knowledge information.

Specifically, the output unit 1207 outputs the searched knowledge and other identified knowledge that is given the same search identifier as the searched knowledge as search results. Specifically, the output unit 1207 outputs the searched knowledge and other identified knowledge that is given the same search identifier and the same task identifier as the searched knowledge, as search results.

The output unit 1207 outputs, for example, an analysis result. Specifically, the output unit 1207 outputs information indicating the percentage of handled queries that belong to each of a plurality of classifications based on predetermined criteria set in the analysis results to be displayed, in which knowledge is utilized. indicate. Thereby, the analysis unit 1206 can enable the service administrator to examine whether or not the situation is such that the knowledge can be appropriately referred to or utilized.

Specifically, the output unit 1207 displays information indicating the ratio of knowledge of each of the plurality of types being utilized among the handled inquiries belonging to each category set in the analysis results to be displayed. do. Thereby, the analysis unit 1206 can enable the service administrator to examine whether or not the situation is such that the knowledge can be appropriately referred to or utilized.

Specifically, the output unit 1207 outputs information indicating the proportion of handled queries belonging to each category set in the analysis results to be displayed, the proportion of which knowledge has been utilized, and the proportion of which knowledge has not been utilized. Display. Thereby, the analysis unit 1206 can enable the service administrator to examine whether or not the situation is such that the knowledge can be appropriately referred to or utilized.

Specifically, the output unit 1207 further outputs the statistical value of the time required for inquiries that have been handled using knowledge and the time required for inquiries that have been handled without using knowledge, which are set in the analysis results to be displayed. Display statistics and. Thereby, the analysis unit 1206 can enable the service administrator to examine whether or not the situation is such that the knowledge can be appropriately referred to or utilized.

Specifically, the output unit 1207 further outputs statistical values of the time required for inquiries that have been handled using the knowledge of each of the plurality of types set in the analysis results to be displayed, and the information that uses the knowledge. Displays statistics on the time required for queries that have already been addressed. Thereby, the analysis unit 1206 can enable the service administrator to examine whether or not the situation is such that the knowledge can be appropriately referred to or utilized.

Specifically, when the output unit 1207 is requested to display information regarding a specific task, the output unit 1207 outputs the number of knowledge referenced when performing the specific task, which is set in the analysis result to be output. Display the number of knowledge utilized in association with a specific task. Thereby, the analysis unit 1206 can enable the service administrator to examine whether or not the situation is such that the knowledge can be appropriately referred to or utilized.

When the output unit 1207 is requested to display information regarding a specific task, the output unit 1207 specifies the display order of knowledge that is referenced or utilized when performing the specific task, which is set in the analysis results to be output. displayed in association with the task. Thereby, the analysis unit 1206 can enable the service administrator to examine whether or not the situation is such that the knowledge can be appropriately referred to or utilized.

Although a case has been described here in which the processing target of the learning unit 1202 and the analysis unit 1206 is knowledge, the present invention is not limited to this. For example, the processing target of the learning unit 1202 and the analysis unit 1206 may be data other than knowledge. Further, here, the information processing apparatus 101 includes a storage unit 1200, an acquisition unit 1201, a learning unit 1202, an assignment unit 1203, a search unit 1204, a recording unit 1205, an analysis unit 1206, and an output unit 1207. Although the case including the above has been described, the case is not limited to this. For example, the information processing device 101 may not include any functional unit and may be able to communicate with an external computer that includes the functional unit.

(Example of operation of information processing system 100)
Next, an example of the operation of the information processing system 100 will be described using FIGS. 13 to 21. First, using FIGS. 13 to 15, in the information processing system 100, the information processing device 101 controls the operator device 103, and the operator device 103 displays a search screen 512 indicating the task to be performed by the service operator. An example of operation for displaying related knowledge will be described.

FIGS. 13 to 15 are explanatory diagrams showing an example of an operation for displaying knowledge related to a task. In FIG. 13, the information processing device 101 receives an inquiry. In the example of FIG. 13, the information processing apparatus 101 accepts the inquiry "Please confirm ○○." Upon receiving the inquiry, the information processing apparatus 101 transmits the inquiry to the administrator apparatus 104.

The information processing device 101 communicates with the administrator device 104 and sets a work flow for handling the inquiry based on the flow data 501. The information processing device 101 updates the ticket data 502 and the usage history data 504 based on the inquiry and the set work flow. The information processing device 101 causes the display 1106 of the operator device 103 to display a ticket screen 511 showing the set work flow based on the flow data 501.

The information processing apparatus 101 detects that information for any task has been clicked on the ticket screen 511. In the following explanation, the task corresponding to the clicked information may be referred to as "target task." In the example of FIG. 13, the information processing apparatus 101 detects that "Check data by executing SQL" has been clicked.

When the information processing apparatus 101 detects that information on any task has been clicked, it acquires the task name and task description for the target task based on the flow data 501 or the usage history data 504. The information processing apparatus 101 generates a search sentence based on the inquiry and the acquired task description. In the example of FIG. 13, the information processing apparatus 101 generates the search sentence "Please check ○○. Please check the contents of the DB data."

The information processing apparatus 101 extracts nouns from the acquired task names and sets narrowing conditions. In the example of FIG. 13, the information processing apparatus 101 sets the narrowing condition "SQL, data confirmation". The information processing apparatus 101 searches for knowledge related to the target task based on the generated search sentence, the set narrowing conditions, and the knowledge data 503. The information processing device 101 acquires the title of the searched knowledge based on the knowledge data 503.

When the information processing device 101 searches for knowledge, the information processing device 101 displays a search screen 512 on the display 1106 of the operator device 103. When the information processing apparatus 101 displays the search screen 512, it generates a unique session ID for each search screen 512. The information processing apparatus 101 displays link information corresponding to the knowledge and a utilization button corresponding to the knowledge on the search screen 512 based on the title of the acquired knowledge. The information processing apparatus 101 detects that a utilization button corresponding to any knowledge has been clicked on the search screen 512. In the following explanation, the knowledge corresponding to the clicked utilization button may be referred to as "target knowledge." In the example of FIG. 13, the information processing apparatus 101 detects that the utilization button for "data confirmation method.doc" and the utilization button for "DB login procedure.xls" have been clicked.

When the information processing apparatus 101 detects that the utilization button has been clicked, it updates the utilization history data 504 for the target knowledge. The information processing apparatus 101 updates the utilization history data 504 by associating the target knowledge, the task ID that identifies the target task, and the generated session ID, for example. Thereby, the information processing device 101 enables the service administrator to grasp the usage status of knowledge in any work flow based on the usage history data 504 when reviewing the work flow and knowledge search method. I can do it.

In addition, the information processing device 101 refers to the tag corresponding to the knowledge utilized when performing the current task based on the utilization history data 504, and the information processing device 101 refers to the tag corresponding to the knowledge utilized when implementing the current task, and the information processing device This makes it easier to search for knowledge. Regarding the case where the information processing device 101 refers to the tag corresponding to the knowledge used when performing the current task, and makes it easier for the service operator to search for useful knowledge when performing subsequent tasks. , will be described later using FIG. 14, for example.

Furthermore, based on the utilization history data 504, the information processing apparatus 101 can group one or more pieces of knowledge that have been utilized on the same search screen 512 in order to perform the same task. Therefore, the information processing apparatus 101 can refer to one or more grouped pieces of knowledge from now on and can easily present useful knowledge when the service operator performs a task. A case where the information processing apparatus 101 refers to one or more grouped pieces of knowledge and displays the pieces of knowledge will be described later with reference to FIG. 15, for example. Next, the description will move on to FIG. 14.

In FIG. 14, the information processing device 101 causes the operator device 103 to display a ticket screen 511, similar to FIG. The service operator clicks "Confirm screen information" to execute the task. When "Confirm screen information" is clicked on the ticket screen 511, the information processing device 101 causes the operator device 103 to display a search screen 512, similarly to FIG. The information processing apparatus 101 searches for knowledge related to the target task, as in FIG. 13 . Similar to FIG. 13, the information processing apparatus 101 displays the title of the searched knowledge on the search screen 512. Here, it is assumed that the service operator utilizes "○○ screen.xls" and clicks the utilization button. Similar to FIG. 13, the information processing apparatus 101 updates the utilization history data 504 to enable identification of utilized knowledge. The information processing device 101 then displays the ticket screen 511 on the operator device 103 again.

The service operator clicks "Confirm DB Information" to perform the next task. When “Confirm DB Information” is clicked on the ticket screen 511, the information processing device 101 causes the operator device 103 to display a search screen 512. Unlike FIG. 13, the information processing device 101 identifies the knowledge that was utilized when performing the previous task based on the knowledge data 503 and the usage history data 504, and identifies the tag given to the identified knowledge. Add to the filtering conditions. In the example of FIG. 14, the information processing apparatus 101 specifies the tags "human resources management work" and "screen information" that are added to "○○ screen. information" and add the tag "Human resources management operations" to the current narrowing conditions. The information processing apparatus 101 searches for knowledge related to the current target task based on the added narrowing conditions. Similar to FIG. 13, the information processing apparatus 101 displays the title of the searched knowledge on the search screen 512.

As a result, the information processing device 101 searches for knowledge that is useful when executing the current task, taking into account the flow of tasks in the work flow and based on the attributes of the knowledge that was utilized when executing the previous task. can be made possible. Therefore, the information processing apparatus 101 can enable the service operator to refer to useful knowledge when performing a task. Next, moving on to the description of FIG. 15, a case will be described in which the information processing apparatus 101 refers to one or more grouped pieces of knowledge and displays the knowledge.

In FIG. 15, the information processing device 101 causes the operator device 103 to display a ticket screen 511, similar to FIG. When "Confirm screen information" is clicked on the ticket screen 511, the information processing device 101 causes the operator device 103 to display a search screen 512, similarly to FIG. The information processing apparatus 101 searches for knowledge related to the target task, as in FIG. 13 . In the example of FIG. 15, it is assumed that the information processing apparatus 101 searches for "■■processing.md". Similar to FIG. 13, the information processing apparatus 101 displays the title of the searched knowledge on the search screen 512.

At this time, if there is a plurality of grouped knowledges including the searched knowledge, the information processing device 101 causes the search screen 512 to display the title of each of the grouped knowledges. . In the example of FIG. 15, the information processing apparatus 101 associates the task ID of "confirm screen information" with "■■ process.md" based on the usage history data 504, and .md" is associated with the same session ID as "○○ screen.xls". Then, the information processing device 101 displays “○○ screen.xls” on the search screen 512.

Here, even though the information processing apparatus 101 is associated with the same session ID as "■■processing.md", the task ID of "confirm user manual" is different from "■■processing.md". Do not display the associated knowledge. In the example of FIG. 15, the information processing apparatus 101 does not display "user manual.pdf" or the like.

Thereby, the information processing apparatus 101 can allow the service operator to refer to other knowledge that was used in the past by another service operator at the same timing as the searched knowledge. In addition, the information processing device 101 uses other knowledge that was used in addition to the searched knowledge when another service operator executed the current task in the past for the same purpose as the searched knowledge. If it is determined that it is possible, it can be made available for reference by the service operator. Therefore, the information processing apparatus 101 can enable the service operator to refer to useful knowledge when performing a task.

In addition, the information processing device 101 may include other knowledge that was utilized in addition to the searched knowledge when another service operator executed a task different from the current task in the past into the current task. It is possible to decide that it is not useful and not allow the service operator to refer to it. Therefore, the information processing apparatus 101 allows the service operator to refer to knowledge that is useful when performing a task.

Next, an example of the operation in which the information processing apparatus 101 adds a tag to knowledge in the information processing system 100 will be described using FIGS. 16 to 19.

FIGS. 16 to 19 are explanatory diagrams showing examples of operations for adding tags to knowledge. In FIG. 16, at the time of installation, the information processing apparatus 101 combines consecutive words included in the task name given to each of a plurality of tasks to generate a tag list. The tag list includes, for example, the tags "screen information", "DB information", "SQL", and "data information".

The information processing apparatus 101 extracts tags included in knowledge from the tag list, and temporarily adds the extracted tags to knowledge. In the example of FIG. 16, the information processing apparatus 101 stores a plurality of knowledge 1600, one or more knowledge 1601 to which the tags "screen information" and "DB information" are attached, and one or more knowledge 1601 to which the tags "SQL" and "data information" are attached. The information is classified into one or more pieces of knowledge 1602. A case in which the information processing apparatus 101 temporarily attaches a tag to knowledge will be described later with reference to FIG. 17, for example.

During operation, the information processing apparatus 101 generates a classification model 1610 based on utilization history data 504 that records knowledge utilized on the search screen 512. During operation, the information processing apparatus 101 may generate the classification model 1610 based on the tagging list 505 that records knowledge to which tags specified by the service operator or service administrator are added. The classification model 1610 is, for example, a model that outputs a tag corresponding to input vector information. Vector information is, for example, information obtained from knowledge. A case in which the information processing apparatus 101 generates the classification model 1610 will be described later using, for example, FIG. 18.

The information processing apparatus 101 uses the classification model 1610 to automatically tag untagged knowledge, or re-tags already tagged knowledge. A case in which the information processing apparatus 101 uses the classification model 1610 to add tags to knowledge will be described later with reference to FIG. 19, for example.

Thereby, the information processing apparatus 101 can assign appropriate tags to knowledge, and can easily present appropriate knowledge to the service operator. Further, even if the tag assigned at the time of installation is inappropriate, the information processing device 101 can re-assign an appropriate tag. Furthermore, the information processing apparatus 101 can reduce the work burden of adding tags on the service administrator. Next, the explanation will move on to FIG. 17.

In FIG. 17, at the time of introduction, the information processing apparatus 101 combines consecutive words included in the task name given to each of a plurality of tasks to generate a tag list. In the example of FIG. 17, the information processing apparatus 101 extracts the tag "screen information" from the task name "screen information confirmation". Furthermore, the information processing apparatus 101 extracts the tag "DB information" from the task name "DB information confirmation". The information processing apparatus 101 also extracts the tag "SQL, data confirmation" from the task name "SQL execution to confirm data". The information processing apparatus 101 generates a tag list 1700 including the extracted tags.

The information processing apparatus 101 temporarily assigns a tag to knowledge based on the tag list 1700. For example, if a tag in the tag list 1700 is included in a character string that combines the full path and content of knowledge, the information processing apparatus 101 temporarily assigns the tag to the knowledge. The information processing apparatus 101 updates the knowledge data 503 based on the result of adding the tag. Thereby, the information processing apparatus 101 can eliminate the need for the service administrator to set tags, and can reduce the workload of assigning tags.

Furthermore, when the service operator or service manager explicitly specifies a tag to be attached to knowledge, the information processing apparatus 101 records the knowledge and the tag in association with each other in the tagging list 505. Next, the description will move on to FIG. 18.

In FIG. 18, during operation, the information processing apparatus 101 identifies a tag attached to at least one piece of knowledge as a "correct answer" based on a tagging list 505, usage history data 504, or knowledge data 503. . For example, the information processing apparatus 101 identifies, as a "correct answer", a tag that is assigned to knowledge that has actually been used and is determined to be useful, based on the usage history data 504 and the knowledge data 503. Further, the information processing apparatus 101 identifies, for example, a tag explicitly designated by the service operator or service administrator and determined to be useful as a "correct answer" based on the tagging list 505.

Based on the knowledge data 503, the information processing device 101 acquires the full path and content of the knowledge for each knowledge to which the specified tag is attached, and vectorizes a character string that combines the full path and content of the acquired knowledge. do. For example, the information processing device 101 generates vector information by arranging bit information indicating whether each of a plurality of predetermined words appears in a character string. The information processing apparatus 101 generates a combination of the identified tag and the generated vector information as teacher data.

The information processing device 101 generates a classification model 1610 based on the teacher data. As the generation method, for example, error back propagation can be adopted. The classification model 1610 can be a tree structure, a mathematical model, a neural network, or the like. Thereby, the information processing apparatus 101 can automatically estimate tags that are preferably added to knowledge using the classification model 1610. Next, the explanation will move on to FIG. 19.

In FIG. 19, the information processing apparatus 101 uses a classification model 1610 to automatically tag untagged knowledge, or re-tags tagged knowledge. For example, the information processing apparatus 101 acquires the full path and contents of the knowledge to be added based on the knowledge data 503, and vectorizes a character string that combines the acquired full path and contents of the knowledge. For example, the information processing device 101 generates vector information by arranging bit information indicating whether each of a plurality of predetermined words appears in a character string. Then, the information processing apparatus 101 acquires the tag by inputting the generated vector information to the classification model 1610. The information processing apparatus 101 adds the acquired tag to the target knowledge.

Thereby, the information processing apparatus 101 can assign appropriate tags to knowledge, and can easily present appropriate knowledge to the service operator. Further, even if the tag assigned at the time of installation is inappropriate, the information processing device 101 can re-assign an appropriate tag. Furthermore, the information processing apparatus 101 can reduce the work burden of adding tags on the service administrator.

Next, an example of the operation in which the information processing device 101 in the information processing system 100 presents the utilization status of knowledge to the service manager will be described using FIGS. 20 to 23.

FIGS. 20 to 23 are explanatory diagrams showing an example of the operation of presenting the usage status of knowledge to the service administrator. In FIG. 20, the information processing device 101 causes the administrator device 104 to display a dashboard screen 513. Dashboard screen 513 includes display columns 2001 to 2006, for example. A display field 2001 displays a list of folders. The display column 2001 displays, for example, a folder in which a ticket is saved in a selectable manner. The display field 2001 displays, for example, a "Link" button that transitions to a maintenance screen for maintaining knowledge used when dealing with inquiries managed by tickets saved in a folder. A display column 2002 displays the number of tickets in the selected folder.

The display column 2003 displays, for each type of knowledge, a knowledge utilization rate that indicates the proportion of tickets that utilize knowledge of that type among all tickets. A display column 2003 displays the percentage of tickets that do not utilize knowledge among all tickets. The percentage of tickets that do not utilize knowledge is "none" in the figure. The display column 2004 displays, for each type of knowledge, the average response time of inquiries managed by tickets that utilize knowledge of that type. A display column 2004 displays the average response time of inquiries managed by tickets that do not utilize knowledge. The average response time for tickets that do not utilize knowledge is "none" in the figure.

The display column 2005 displays, for each month and each type of knowledge, a knowledge utilization rate that indicates the ratio of tickets that utilize knowledge of the type among the tickets in the selected folder. The display column 2005 displays the percentage of tickets that do not utilize knowledge among the tickets in the selected folder, for each month. The percentage of tickets that do not utilize knowledge is "none" in the figure. The display column 2006 displays the average response time of inquiries managed by tickets that utilize knowledge of the relevant type in the selected folder for each month and for each type of knowledge. The display column 2006 displays, by month, the average response time of inquiries that are in the selected folder and are managed by tickets that do not utilize knowledge. The average response time for tickets that do not utilize knowledge is "none" in the figure.

Now, moving on to the description of FIG. 21, an operation example in which the information processing apparatus 101 calculates various knowledge utilization rates to be displayed in the display columns 2003 and 2005 will be described. As shown in FIG. 21, the information processing device 101 identifies the knowledge utilized in the inquiry managed by the ticket based on the usage history data 504, and identifies the type of the identified knowledge based on the knowledge data 503. do. The information processing device 101 sets the number of uses corresponding to the ticket type in the ticket data 502 based on the identified ticket type and the identified knowledge.

At this time, the information processing device 101 may identify the knowledge referenced in the inquiry managed by the ticket based on the usage history data 504 and identify the type of the identified knowledge based on the knowledge data 503. . The information processing apparatus 101 may then set the number of references corresponding to the ticket type in the ticket data 502 based on the identified ticket type and the identified knowledge.

Thereafter, the information processing device 101 calculates, for each type of knowledge, a knowledge utilization rate indicating the proportion of tickets that utilize knowledge of that type among all tickets, based on the ticket data 502 in which the number of utilizations has been set. Furthermore, the information processing device 101 displays the percentage of tickets for which knowledge is not used among all tickets, based on the ticket data 502 in which the number of uses is set.

In addition, based on the ticket data 502 in which the number of uses has been set, the information processing device 101 calculates, for each month and for each type of knowledge, the proportion of tickets that use knowledge of the relevant type among the tickets in the selected folder. Calculate the knowledge utilization rate shown. Furthermore, the information processing device 101 displays the percentage of tickets that do not utilize knowledge among the tickets in the selected folder for each month, based on the ticket data 502 in which the number of utilization is set. Thereby, the information processing apparatus 101 can display various knowledge utilization rates in the display columns 2003 and 2005.

Now, moving on to the description of FIG. 22, an operation example in which the information processing apparatus 101 calculates various average response times to be displayed in the display columns 2004 and 2006 will be described. As shown in FIG. 22, the information processing device 101 identifies the knowledge utilized in the inquiry managed by the ticket based on the usage history data 504, and identifies the type of the identified knowledge based on the knowledge data 503. do. Then, based on the response time of the ticket data 502, the information processing device 101 calculates, for each type of knowledge identified, the average response time of inquiries managed by tickets that utilize the knowledge of that type. Furthermore, the information processing apparatus 101 calculates the average response time of inquiries managed by tickets that do not utilize knowledge, based on the response time of the ticket data 502.

Furthermore, based on the response time of the ticket data 502, the information processing device 101 stores inquiries managed by tickets that are in the selected folder and that utilize knowledge of the relevant type for each month and for each type of knowledge identified. Calculate the average response time. Furthermore, based on the response time of the ticket data 502, the information processing device 101 displays, by month, the average response time of inquiries that are in the selected folder and are managed by tickets that do not utilize knowledge. Thereby, the information processing apparatus 101 can display various average response times in the display columns 2004 and 2006.

Returning to the explanation of FIG. 20, when the service administrator refers to the dashboard screen 513, the service administrator checks the knowledge utilization rate and verifies the usefulness of the knowledge or the validity of the search method. Further, by referring to the dashboard screen 513, the service manager checks the average response time, verifies the effect of utilizing knowledge, and verifies the usefulness of knowledge or the validity of the search method.

Thereby, the service administrator can intuitively understand the usage status of knowledge through the dashboard screen 513. Then, the service administrator can effectively review the content of knowledge, work flow, knowledge search method, etc. based on the verified results. The service administrator may also click on the "Link" button to organize the knowledge. Next, moving on to the explanation of FIG. 23, the case where the "Link" button is clicked will be explained.

As shown in FIG. 23, when the "Link" button is clicked, the information processing apparatus 101 causes the administrator apparatus 104 to display a maintenance screen 2300 for the folder corresponding to the "Link" button. The maintenance screen 2300 is a screen for maintaining knowledge used when dealing with inquiries managed by tickets saved in a folder.

The maintenance screen 2300 includes a display field 2301, a display field 2302, and a display field 2303. A display column 2301 displays the average response time for each work flow. Similar to FIG. 22, the information processing apparatus 101 calculates the average response time for each work flow and displays it in the display column 2301.

A display column 2302 displays a list of tasks. For example, the display column 2302 displays, for each task, the number of tickets on which the task was executed. For example, the display column 2302 displays, for each task, the number of references to knowledge and the number of uses of knowledge when various service operators implement the task. The display column 2303 displays usage history data 504.

The service administrator refers to the maintenance screen 2300 and reviews the content of knowledge, work flow, knowledge search method, and the like. For example, the service manager refers to the display field 2301 and narrows down the work flows that have a relatively long average response time and are likely to have problems. The service administrator refers to the display column 2302 to check the number of references and the number of uses for the narrowed down work flow, and if the number of uses is less than the number of references by more than a certain level, the service administrator checks the content of knowledge and the knowledge search method. It is determined that there is a problem. The service administrator then considers ways to expand the knowledge that is easy to utilize and improve the content of the knowledge. Additionally, the service administrator may review the knowledge search method to make it easier to search for knowledge that is not only referenced but also utilized.

In addition, the service administrator refers to the display field 2303 to check the display order of the narrowed-down work flow, and if the display order of the knowledge to be utilized is below a certain level, the service administrator can check the content of the knowledge and the knowledge search method. determine that there is a problem. The service administrator then considers improvements to the content of the knowledge. Further, the service administrator may review the knowledge search method so that the display ranking of the knowledge that is utilized becomes higher.

In this way, the service administrator can reduce the workload compared to manually checking each ticket and reviewing the knowledge content, work flow, knowledge search method, etc. I can do it. The service administrator can use the dashboard screen 513 and the maintenance screen 2300 to determine how it is preferable to improve the content of knowledge, the work flow, and the knowledge search method.

Although a case has been described here in which the information processing apparatus 101 counts the number of references and the number of uses of knowledge at the timing of displaying the dashboard screen 513, the present invention is not limited to this. For example, the information processing apparatus 101 may increment the number of references or utilization of knowledge each time knowledge is referenced or utilized.

(Ticket processing procedure)
Next, an example of a ticket processing procedure executed by the information processing apparatus 101 will be described using FIG. 24. Ticket processing is realized by, for example, the CPU 601 shown in FIG. 6, a storage area such as a memory 602 or a recording medium 605, and a network I/F 603.

FIG. 24 is a flowchart showing an example of a ticket processing procedure. In FIG. 24, the information processing apparatus 101 acquires ticket data 502 and flow data 501 (step S2401). Next, the information processing device 101 displays a ticket details screen (step S2402).

The information processing apparatus 101 then determines whether to complete the ticket processing (step S2403). Here, if the ticket processing is completed (step S2403: Yes), the information processing apparatus 101 moves to the processing of step S2407. On the other hand, if the ticket processing is not completed (step S2403: No), the information processing apparatus 101 moves to the process of step S2404.

In step S2404, the information processing apparatus 101 accepts the selection of a task (step S2404). Next, the information processing apparatus 101 executes task processing, which will be described later with reference to FIGS. 25 and 26, for the selected task (step S2405). The information processing device 101 then updates the ticket data 502 (step S2406). The information processing apparatus 101 returns to the process of step S2403.

In step S2407, the information processing apparatus 101 updates the ticket data 502 (step S2407). The information processing device 101 then ends the ticket processing.

(Task processing procedure)
Next, an example of a task processing procedure executed by the information processing apparatus 101 will be described using FIGS. 25 and 26. Task processing is realized by, for example, the CPU 601 shown in FIG. 6, a storage area such as a memory 602 or a recording medium 605, and a network I/F 603.

25 and 26 are flowcharts illustrating an example of a task processing procedure. In FIG. 25, the information processing apparatus 101 generates a session ID (step S2501).

Next, the information processing apparatus 101 executes knowledge search processing, which will be described later using FIG. 27 (step S2502). The information processing apparatus 101 then displays the search screen 512 including the searched knowledge (step S2503).

Next, the information processing apparatus 101 determines whether knowledge has been referenced (step S2504). Here, if it is referenced (step S2504: Yes), the information processing apparatus 101 moves to the process of step S2505. On the other hand, if it has not been referenced (step S2504: No), the information processing apparatus 101 moves to the process of step S2506.

In step S2505, the information processing apparatus 101 updates the utilization history data 504 (step S2505). For example, if there is no record corresponding to the referenced knowledge in the utilization history data 504, the information processing apparatus 101 adds the record and sets various information. If there is a record corresponding to the referenced knowledge, the information processing apparatus 101 updates the record. Specifically, the information processing apparatus 101 sets a task ID in the task ID field of the record corresponding to the referenced knowledge in the utilization history data 504. Specifically, the information processing apparatus 101 sets Clicked in the utilization method field of the record corresponding to the referenced knowledge in the utilization history data 504. At this time, the information processing apparatus 101 may further specifically update the reference number field of the record corresponding to the referenced knowledge in the ticket data 502. The information processing apparatus 101 then proceeds to the process of step S2506.

In step S2506, the information processing apparatus 101 determines whether the knowledge has been utilized (step S2506). Here, if it is being utilized (step S2506: Yes), the information processing apparatus 101 moves to the process of step S2507. On the other hand, if it is not being utilized (step S2506: No), the information processing apparatus 101 moves to the process of step S2601 in FIG. 26.

In step S2507, the information processing apparatus 101 updates the utilization history data 504 (step S2507). For example, if there is no record corresponding to the utilized knowledge in the utilization history data 504, the information processing device 101 adds the record and sets various information. The information processing device 101 updates the record if there is a record corresponding to the utilized knowledge. Specifically, the information processing apparatus 101 sets Utilized in the utilization method field of the record corresponding to the utilized knowledge in the utilization history data 504. Specifically, the information processing apparatus 101 may set the generated session ID in the session ID field of the record corresponding to the utilized knowledge in the utilization history data 504. At this time, the information processing apparatus 101 may further specifically update the field of the number of utilizations of the record corresponding to the utilized knowledge in the ticket data 502. The information processing apparatus 101 then proceeds to the process of step S2601 in FIG. 26.

In FIG. 26, the information processing apparatus 101 determines whether to close the search screen 512 (step S2601). Here, in the case of closing (step S2601: Yes), the information processing apparatus 101 moves to the process of step S2602. On the other hand, if it is not closed (step S2601: No), the information processing apparatus 101 moves to the process of step S2603.

In step S2602, the information processing apparatus 101 closes the search screen 512 and then determines whether to perform a search again (step S2602). Here, if a re-search is to be performed (step S2602: Yes), the information processing apparatus 101 returns to the process of step S2501 in FIG. 25. On the other hand, if re-search is not to be performed (step S2602: No), the information processing apparatus 101 moves to the process of step S2604.

In step S2603, the information processing apparatus 101 determines whether to perform a re-search without closing the search screen 512 (step S2603). Here, if a re-search is to be performed (step S2603: Yes), the information processing apparatus 101 returns to the process of step S2502 in FIG. 25. On the other hand, if re-search is not to be performed (step S2603: No), the information processing apparatus 101 moves to the process of step S2604.

In step S2604, the information processing apparatus 101 determines whether the task has been completed (step S2604). Here, if the task is not completed (step S2604: No), the information processing apparatus 101 returns to the process of step S2503 in FIG. 25. On the other hand, if the task has been completed (step S2604: Yes), the information processing apparatus 101 ends the task processing.

Here, a case has been described in which the information processing apparatus 101 sets the generated session ID in the session ID field of the record corresponding to the utilized knowledge in the utilization history data 504 in step S2507, but this is not limited to this. do not have. For example, the information processing apparatus 101 may set the generated session ID in the session ID field of the record corresponding to the referenced knowledge in the usage history data 504 in step S2505. Furthermore, for example, the information processing apparatus 101 may generate a new record in the usage history data 504 at the time when knowledge is searched, and set the generated session ID.

(Knowledge search processing procedure)
Next, an example of a knowledge search processing procedure executed by the information processing apparatus 101 will be described using FIG. 27. The knowledge search process is realized by, for example, the CPU 601 shown in FIG. 6, a storage area such as a memory 602 or a recording medium 605, and a network I/F 603.

FIG. 27 is a flowchart illustrating an example of a knowledge search processing procedure. In FIG. 27, the information processing apparatus 101 sets narrowing conditions (step S2701). The narrowing condition is, for example, a condition that includes a noun included in the task name as a tag.

Next, the information processing apparatus 101 determines whether a processed task exists (step S2702). Here, if there is a processed task (step S2702: Yes), the information processing apparatus 101 moves to the process of step S2703. On the other hand, if there is no processed task (step S2702: No), the information processing apparatus 101 moves to the process of step S2706.

In step S2703, the information processing apparatus 101 extracts tags from the utilized knowledge (step S2703). Next, the information processing apparatus 101 determines whether an additional tag exists (step S2704). Additional tags are tags used in subsequent searches.

Here, if an additional tag exists (step S2704: Yes), the information processing apparatus 101 moves to the process of step S2705. On the other hand, if there is no additional tag (step S2704: No), the information processing apparatus 101 moves to the process of step S2706.

In step S2705, the information processing apparatus 101 adds an additional tag to the narrowing down conditions (step S2705). The information processing apparatus 101 then proceeds to the process of step S2706.

In step S2706, the information processing apparatus 101 searches for knowledge (step S2706). The information processing device 101 then ends the knowledge search process.

(Related knowledge display processing procedure)
Next, an example of a related knowledge display processing procedure executed by the information processing apparatus 101 will be described using FIG. 28. The related knowledge display process is realized by, for example, the CPU 601 shown in FIG. 6, a storage area such as a memory 602 or a recording medium 605, and a network I/F 603.

FIG. 28 is a flowchart illustrating an example of a related knowledge display processing procedure. In FIG. 28, the information processing apparatus 101 determines whether the "reference history" button has been pressed (step S2801). Here, if the "reference history" button is not pressed (step S2801: No), the information processing apparatus 101 ends the related knowledge display process. On the other hand, if the "reference history" button is pressed (step S2801: Yes), the information processing apparatus 101 moves to the process of step S2802.

In step S2802, the information processing apparatus 101 acquires the task ID and session ID of the displayed knowledge (step S2802). Next, the information processing apparatus 101 searches for related knowledge based on the acquired task ID and session ID (step S2803). For example, the information processing apparatus 101 searches for related knowledge in which the acquired task ID and session ID are associated, based on the knowledge data 503.

Then, the information processing apparatus 101 determines whether related knowledge exists (step S2804). Here, if the related knowledge does not exist (step S2804: No), the information processing apparatus 101 ends the related knowledge display process. On the other hand, if related knowledge exists (step S2804: Yes), the information processing apparatus 101 moves to the process of step S2805.

In step S2805, the information processing apparatus 101 displays related knowledge (step S2805). The information processing device 101 then ends the related knowledge display process.

(Classification model learning processing procedure)
Next, an example of a classification model learning processing procedure executed by the information processing apparatus 101 will be described using FIG. 29. The classification model learning process is realized by, for example, the CPU 601 shown in FIG. 6, a storage area such as a memory 602 or a recording medium 605, and a network I/F 603.

FIG. 29 is a flowchart illustrating an example of a classification model learning processing procedure. In FIG. 29, the information processing apparatus 101 acquires a combination of knowledge and tag that becomes "correct" data (step S2901).

Next, the information processing apparatus 101 acquires the full path and content of the acquired knowledge (step S2902). The information processing apparatus 101 then extracts words from the acquired full path and content (step S2903).

Next, the information processing apparatus 101 vectorizes the extracted words (step S2904). The information processing apparatus 101 then vectorizes the acquired tag (step S2905).

Next, the information processing apparatus 101 generates training data based on the word vector and the tag vector (step S2906). The information processing apparatus 101 then generates the classification model 507 based on the generated training data (step S2907). After that, the information processing device 101 ends the classification model learning process.

(Knowledge classification processing procedure)
Next, an example of a knowledge classification processing procedure executed by the information processing apparatus 101 will be described using FIG. 30. The knowledge classification process is realized by, for example, the CPU 601 shown in FIG. 6, a storage area such as a memory 602 or a recording medium 605, and a network I/F 603.

FIG. 30 is a flowchart illustrating an example of a knowledge classification processing procedure. In FIG. 30, the information processing apparatus 101 acquires knowledge to which a tag is to be added (step S3001). The target knowledge is, for example, other than knowledge that has been explicitly manually tagged.

Next, the information processing apparatus 101 acquires the full path and content of the acquired knowledge (step S3002). The information processing apparatus 101 then extracts words from the acquired full path and content (step S3003).

Next, the information processing apparatus 101 determines whether the classification model 507 has been generated (step S3004). Here, if it has been generated (step S3004: Yes), the information processing apparatus 101 moves to the process of step S3005. On the other hand, if it has not been generated (step S3004: No), the information processing apparatus 101 moves to the process of step S3008.

In step S3005, the information processing apparatus 101 vectorizes the extracted words (step S3005). Next, the information processing apparatus 101 executes tag inference processing based on the classification model 507 (step S3006). The information processing apparatus 101 executes tag inference processing by inputting vector information obtained through vectorization into the classification model 507, for example. The information processing apparatus 101 then obtains the inference result (step S3007). The inference result is a tag. After that, the information processing apparatus 101 moves to the process of step S3010.

In step S3008, the information processing apparatus 101 compares the word with the tag list (step S3008). Next, the information processing apparatus 101 obtains tags that match the word as a result of the comparison (step S3009). The information processing apparatus 101 then proceeds to the process of step S3010.

In step S3010, the information processing apparatus 101 adds a tag to knowledge (step S3010). The information processing device 101 then ends the knowledge classification process.

(Dashboard display processing procedure)
Next, an example of a dashboard display processing procedure executed by the information processing apparatus 101 will be described using FIGS. 31 and 32. The dashboard display process is realized by, for example, the CPU 601 shown in FIG. 6, a storage area such as a memory 602 or a recording medium 605, and a network I/F 603.

31 and 32 are flowcharts illustrating an example of a dashboard display processing procedure. In FIG. 31, the information processing apparatus 101 acquires ticket data 502 (step S3101).

Next, the information processing device 101 identifies, for each ticket, whether knowledge is used or not, and the type of knowledge used (step S3102). Then, the information processing apparatus 101 calculates the knowledge utilization rate for each type of knowledge utilized (step S3103).

Next, the information processing apparatus 101 obtains the response time for each ticket (step S3104). The information processing apparatus 101 then calculates the average response time for each type of knowledge utilized (step S3105).

Next, the information processing apparatus 101 displays a dashboard (step S3106). The information processing apparatus 101 then proceeds to the process of step S3201 in FIG. 32.

In FIG. 32, the information processing apparatus 101 determines whether or not to drill down (step S3201). Drilldown is to refine the data summary items. Here, if drill down is not to be performed (step S3201: No), the information processing apparatus 101 moves to the process of step S3208. On the other hand, when performing drill down (step S3201: Yes), the information processing apparatus 101 moves to the process of step S3202.

In step S3202, the information processing apparatus 101 acquires usage history data 504 (step S3202). Next, the information processing apparatus 101 classifies the usage history by task (step S3203). Then, the information processing apparatus 101 calculates the number of knowledge references for each classification (step S3204).

Next, the information processing apparatus 101 calculates the number of knowledge uses for each classification (step S3205). The information processing apparatus 101 then displays knowledge reference/utilization information for each task based on the number of knowledge references for each category and the number of knowledge uses for each category (step S3206).

Next, the information processing apparatus 101 displays detailed information of the utilized knowledge (step S3207). The information processing apparatus 101 then proceeds to the process of step S3208.

In step S3208, the information processing apparatus 101 determines whether to close the dashboard (step S3208). Here, if it is not closed (step S3208: No), the information processing apparatus 101 returns to the process of step S3106 in FIG. 31. On the other hand, when closing (step S3208: Yes), the information processing apparatus 101 ends the dashboard display process.

As described above, the information processing apparatus 101 can generate vector information representing the characteristics of a character string obtained by combining the location, name, and content of data. According to the information processing device 101, learning data that associates generated vector information with tags used for classifying data attached to data is used to output a tag corresponding to input vector information. A model can be generated. According to the information processing apparatus 101, it is possible to receive input of data to which a new tag is to be added. According to the information processing device 101, the tag obtained as a result of inputting vector information representing the characteristics of a character string obtained by combining the location, name, and content of the target data into the generated learning model is used as the target data. can be attached to the data. Thereby, the information processing apparatus 101 can eliminate the need for the service administrator to set tags, and can reduce the workload of assigning tags.

According to the information processing device 101, bit information indicating whether or not each of a plurality of predetermined words appears in a character string obtained by combining the location, name, and content of data is arranged, and a vector is generated. Information can be generated. According to the information processing device 101, bit information indicating whether each word of a plurality of words appears in a character string obtained by combining the location, name, and content of target data is arranged, and a vector is generated. Information can be generated. Thereby, the information processing apparatus 101 can accurately represent the characteristics of a character string and generate vector information useful for generating or searching a learning model.

According to the information processing apparatus 101, a first tag that is attached to a task performed using data among a plurality of tagged tasks can be attached to data. According to the information processing device 101, learning is performed in which vector information representing the characteristics of a character string obtained by combining the location, name, and content of data to which the first tag is attached is associated with the first tag. A learning model can be generated using the data. As a result, the information processing device 101 can use learning data that includes tags that have been actually used and are determined to make knowledge searchable appropriately, and generate a learning model that can accurately estimate tags. can do.

According to the information processing apparatus 101, it is possible to generate vector information representing the characteristics of a character string obtained by combining the location, name, and content of data used when performing a task. According to the information processing apparatus 101, a learning model can be generated using learning data in which generated vector information is associated with a tag attached to the data and used for classifying data. As a result, the information processing device 101 can use learning data that includes tags that have been actually used and are determined to make knowledge searchable appropriately, and generate a learning model that can accurately estimate tags. can do.

According to the information processing apparatus 101, the second tag for which the designation has been received can be attached to the data for which the designation has been received. According to the information processing device 101, learning is performed in which vector information representing the characteristics of a character string obtained by combining the location, name, and content of data to which the second tag is attached is associated with the second tag. A learning model can be generated using the data. As a result, the information processing device 101 can use learning data that includes a tag that is determined to be appropriately searchable for knowledge because it has been specified by the service operator or service administrator, and can estimate the tag with high accuracy. A learning model can be generated.

According to the information processing apparatus 101, the location and name of data can be specified from the full path of the data. According to the information processing apparatus 101, the location and name of target data can be specified from the full path of the target data. Thereby, the information processing apparatus 101 can make the location and name of the data available.

According to the information processing device 101, knowledge can be employed as data. Thereby, the information processing apparatus 101 can add tags useful for classifying knowledge to knowledge. Therefore, the information processing apparatus 101 can enable a service operator to search for useful knowledge when performing a task.

Note that the information processing method described in this embodiment can be realized by executing a program prepared in advance on a computer such as a PC or a workstation. The information processing program described in this embodiment is recorded on a computer-readable recording medium, and executed by being read from the recording medium by the computer. The recording medium includes a hard disk, a flexible disk, a CD (Compact Disc)-ROM, an MO, a DVD (Digital Versatile Disc), and the like. Furthermore, the information processing program described in this embodiment may be distributed via a network such as the Internet.

Regarding the embodiments described above, the following additional notes are further disclosed.

(Additional note 1) Learning data in which vector information representing the characteristics of a character string obtained by combining the location, name, and content of data is associated with a tag attached to the data used to classify the data to generate a learning model that outputs a tag corresponding to the input vector information,
When inputting data to which a new tag is to be added is received, vector information representing the characteristics of a character string obtained by combining the location, name, and content of the target data is input into the generated learning model. attaching a tag output from the learning model to the target data in accordance with the
An information processing program that causes a computer to perform processing.

(Additional note 2) Bit information indicating whether each word of a plurality of predetermined words appears in a character string obtained by combining the location, name, and content of the data is arranged, and the location of the data is Generate vector information representing the characteristics of the character string obtained by combining the position, name, and content,
The location of the target data is determined by arranging bit information indicating whether each word of the plurality of words appears in a character string obtained by combining the location, name, and content of the target data. The information processing program according to appendix 1, characterized in that the computer executes a process of generating vector information representing characteristics of a character string obtained by combining a name and a content.

(Additional note 3) causing the computer to perform a process of assigning to the data a first tag assigned to a task performed using the data among a plurality of tasks assigned tags;
The process of generating the learning model includes:
Using learning data in which the first tag is associated with vector information representing the characteristics of a character string obtained by combining the location, name, and content of the data to which the first tag is attached, The information processing program according to Supplementary Note 1 or 2, wherein the information processing program generates the learning model.

(Additional note 4) The process of generating the learning model is
Vector information representing the characteristics of a character string obtained by combining the location, name, and content of the data used when performing the task, and a tag used to classify the data that is attached to the data. The information processing program according to any one of appendices 1 to 3, characterized in that the learning model is generated using learning data associated with.

(Additional note 5) causing the computer to perform a process of attaching a second tag for which the designation has been received to the data for which the designation has been received;
The process of generating the learning model includes:
Using learning data in which the second tag is associated with vector information representing the characteristics of a character string obtained by combining the location, name, and content of the data to which the second tag is attached, 5. The information processing program according to any one of appendices 1 to 4, which generates the learning model.

(Additional note 6) The location and name of the data are expressed by the full path of the data,
6. The information processing program according to any one of appendices 1 to 5, wherein the location and name of the target data are expressed by a full path of the target data.

(Supplementary Note 7) The information processing program according to any one of Supplementary Notes 1 to 6, wherein the data is knowledge.

(Additional note 8) Learning data in which vector information representing the characteristics of a character string obtained by combining the location, name, and content of data is associated with a tag used to classify the data that is attached to the data. to generate a learning model that outputs a tag corresponding to the input vector information,
When inputting data to which a new tag is to be added is received, vector information representing the characteristics of a character string obtained by combining the location, name, and content of the target data is input into the generated learning model. attaching a tag output from the learning model to the target data in accordance with the
An information processing method characterized in that processing is performed by a computer.

(Additional note 9) Learning data in which vector information representing the characteristics of a character string obtained by combining the location, name, and content of data is associated with a tag attached to the data used to classify the data to generate a learning model that outputs a tag corresponding to the input vector information,
When inputting data to which a new tag is to be added is received, vector information representing the characteristics of a character string obtained by combining the location, name, and content of the target data is input into the generated learning model. attaching a tag output from the learning model to the target data in accordance with the
An information processing device comprising a control section.

100 Information processing system 101 Information processing device 102 User device 103 Operator device 104 Administrator device 110 Network 201,511 Ticket screen 202,512 Search screen 401,503 Knowledge data 402,504 Utilization history data 410,513 Dashboard screen 501 Flow data 502 Ticket data 505 Tagging list 506 Search model 507,1610 Classification model 600,1100 Bus 601,1101 CPU
602,1102 Memory 603,1103 Network I/F
604, 1104 Recording medium I/F
605,1105 Recording medium 1106 Display 1107 Input device 1200 Storage unit 1201 Acquisition unit 1202 Learning unit 1203 Adding unit 1204 Search unit 1205 Recording unit 1206 Analysis unit 1207 Output unit 1600-1602 Knowledge 1700 Tag list 2001-2006 ,2301~2303 Display column 2300 Maintenance screen

Claims (5)

Using learning data that associates vector information representing the characteristics of a character string obtained by combining the location, name, and content of the data with a tag used to classify the data that is attached to the data, Generates a learning model that outputs tags corresponding to the input vector information,
When inputting data to which a new tag is to be added is received, vector information representing the characteristics of a character string obtained by combining the location, name, and content of the target data is input into the generated learning model. attaching a tag output from the learning model to the target data in accordance with the
An information processing program that causes a computer to perform processing. Bit information indicating whether or not each word of a plurality of predetermined words appears in a character string obtained by combining the location, name, and content of the data is arranged, and the location and name of the data are combined. Generate vector information representing the characteristics of the character string obtained by combining the contents,
The location of the target data is determined by arranging bit information indicating whether each word of the plurality of words appears in a character string obtained by combining the location, name, and content of the target data. 2. The information processing program according to claim 1, causing the computer to execute a process of generating vector information representing characteristics of a character string obtained by combining a name and a content. causing the computer to perform a process of assigning to the data a first tag assigned to a task performed using the data among a plurality of tagged tasks;
The process of generating the learning model includes:
Using learning data in which the first tag is associated with vector information representing the characteristics of a character string obtained by combining the location, name, and content of the data to which the first tag is attached, The information processing program according to claim 1 or 2, wherein the information processing program generates the learning model. Using learning data that associates vector information representing the characteristics of a character string obtained by combining the location, name, and content of the data with a tag used to classify the data that is attached to the data, Generates a learning model that outputs tags corresponding to the input vector information,
When inputting data to which a new tag is to be added is received, vector information representing the characteristics of a character string obtained by combining the location, name, and content of the target data is input into the generated learning model. attaching a tag output from the learning model to the target data in accordance with the
An information processing method characterized in that processing is performed by a computer. Using learning data that associates vector information representing the characteristics of a character string obtained by combining the location, name, and content of the data with a tag used to classify the data that is attached to the data, Generates a learning model that outputs tags corresponding to the input vector information,
When inputting data to which a new tag is to be added is received, vector information representing the characteristics of a character string obtained by combining the location, name, and content of the target data is input into the generated learning model. attaching a tag output from the learning model to the target data in accordance with the
An information processing device comprising a control section.

Images