JP7611779B2 - Information Retrieval System - Google Patents

Description

The present invention relates to an information search system.

So-called in-house portal sites that utilize websites and the like are widely used for information searches within companies. For such in-house portal sites, the maintenance work (adding and updating content, etc.) of the administrator who manages the site and its contents plays an important role. In addition, technology has been proposed that allows users to efficiently search for desired information. For example, Patent Document 1 discloses displaying the hierarchical structure of content within an in-house portal on a web browser.

JP 2008-242803 A

However, as the amount of content increases and the years of operation pass, users may find it difficult to search for the information they are looking for. In response to these situations, the administrator of the in-house portal site updates pages on the site and adds and updates content, but sometimes this maintenance work cannot keep up.

The present invention was made in consideration of these circumstances, and aims to provide a technology that allows users to efficiently search for desired information.

In order to achieve the above object, one aspect of the present invention is
a obtaining means for obtaining the text of a user's question;
An identification means for identifying a predetermined purpose based on the text;
A search means for searching for desired information corresponding to the predetermined purpose;
a recording means for recording desired information corresponding to the predetermined purpose based on a search log related to the predetermined purpose;
an output means for outputting the search results;
It is an information retrieval system having the above structure.

The present invention allows users to efficiently search for desired information.

FIG. 13 is a diagram showing an example of information retrieval and recording. FIG. 1 is a diagram showing an outline of the configuration of an information search system. FIG. 2 is a block diagram showing a hardware configuration of a server. FIG. 2 is a functional block diagram illustrating an example of a functional configuration of a server. 13 is a flowchart showing the operation of an information search process. FIG. 13 is a diagram showing a display example during information search. 13 is a flowchart showing an operation of an information registration process. FIG. 13 is a diagram showing a display example when registering information. 13 is a flowchart showing an operation of an information notification process. FIG. 13 is a diagram showing a display example when notifying information.

(Embodiment)
＜Overview＞
Hereinafter, the present embodiment will be described with reference to the drawings. Fig. 1 is a diagram showing an example of information search and recording according to the present embodiment. In the present embodiment, an example of information search using a chat function on a terminal such as a smartphone will be described.

Figure 1 shows an example of messages exchanged between server 1 (information search device) and users A to C by operating terminal 2. For example, assume that user A inputs a question such as "Where is the xxx document?" (message CC1 from user A). In this case, when server 1 receives message CC1, it identifies a specific purpose from the question and searches for desired information corresponding to the specific purpose. For example, in the above case, the specific purpose is to search for the "location" of the "xxx document". Then, server 1 searches for the "location" of the "xxx document" from a storage unit such as the in-house portal CP, cloud storage CS, or desired information DB41. Details of each storage unit will be described later. If server 1 can obtain the "location of the xxx document" from the in-house portal CP, it responds to user A with information on the storage location of the "xxx document" (for example, URL: Uniform Resource Locator). In the example of FIG. 1, a response such as "It's here: http://portal/xxx.ppt" (message CC2 from the server) is displayed on terminal 2. In this way, when user A is able to search for the desired information, server 1 displays the desired information on terminal 2.

For example, when user A inputs a question such as "Where is the remote work application form?" (message CC3 from user A), server 1 searches for the desired information corresponding to the specified predetermined purpose in the same manner as described above. For example, in the above case, the predetermined purpose is to search for the "location" of the "remote work application form". Here, if server 1 is unable to obtain the "location" of the "remote work application form" from any of the memory units, it responds to user A to that effect. In the example of FIG. 1, a response such as "I don't know" (message CC4 from the server) is displayed on terminal 2. Here, in this embodiment, if the desired information is not obtained, the user who asked the question or other users can provide the desired information to server 1. For example, in the example of FIG. 1, when user B sends message CC5 (from user B) stating "It's here. The RW application form, http://storage.com/yyy.xls", server 1 records in desired information DB 41 that the location of the "RW application form" is "http://storage.com/yyy.xls". Then, once recording is complete, server 1 replies with message CC6 (from the server) stating "I remembered." In this way, even if server 1 is unable to search for desired information (information to achieve a specific purpose), it can record the desired information based on interactions with the user (search log).

For example, when user C inputs a question such as "Who is the person in charge of the zzz case?" (message CC7 from user C), the server 1 searches for desired information corresponding to the specified predetermined purpose in the same manner as described above. For example, in the above case, the predetermined purpose is to find the "person in charge" of the "zzz case". If the server 1 cannot obtain the "person in charge" of the "zzz case" from any of the storage units, it responds to that effect to user C. In the example of FIG. 1, a response such as "I don't know" (message CC8 from the server) is displayed on the terminal 2. If user B then sends a message CC9 (from user B) saying "It's Yamada-san from the Sales Department 1", the server 1 records in the desired information DB 41 that the "person in charge" of the "zzz case" is "Yamada-san from the Sales Department 1". In this way, even if the server 1 cannot search for the information the user is looking for, if another user knows it, by recording it in the desired information DB 41, the server 1 can respond the next time a similar question is input. The system configuration according to this embodiment will be described below.

<System Configuration>
2 is a diagram showing an outline of the system configuration of the information search system according to this embodiment. The information search system according to this embodiment is configured by connecting a server 1 that performs processing, and a terminal 2 and a terminal 3 that are operated by users, via a predetermined network N such as the Internet.

Server 1 executes various processes in cooperation with the operations of terminals 2 and 3. Terminals 2 and 3 accept input from the user and display the results of processing by server 1.

<Hardware Configuration>
3 is a block diagram showing a hardware configuration of the server 1 according to this embodiment. The server 1 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a bus 14, an input/output interface 15, an output unit 16, an input unit 17, a storage unit 18, a communication unit 19, and a drive 20.

The CPU 11 executes various processes according to a program recorded in the ROM 12 or a program loaded from the storage unit 18 to the RAM 13. The RAM 13 also stores data and the like required for the CPU 11 to execute various processes as appropriate. The CPU 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input/output interface 15 is also connected to this bus 14.

The input/output interface 15 is connected to an output unit 16, an input unit 17, a memory unit 18, a communication unit 19, and a drive 20. The output unit 16 is composed of a display, a speaker, etc., and outputs various information as images and sounds. The input unit 17 is composed of a keyboard, a mouse, etc., and inputs various information. The memory unit 18 is composed of a hard disk, a DRAM (Dynamic Random Access Memory), etc., and stores various data. The communication unit 19 communicates with other devices via a network N including the Internet.

Removable media 21, which may be a magnetic disk, optical disk, magneto-optical disk, or semiconductor memory, is appropriately attached to the drive 20. Programs read from the removable media 21 by the drive 20 are installed in the storage unit 18 as necessary. The removable media 21 can also store various data stored in the storage unit 18 in the same way as the storage unit 18.

Although not shown, terminal 2 and terminal 3 have the hardware configuration shown in FIG. 3.

<Functional configuration>
4 is a functional block diagram showing an example of a functional configuration of the server 1 according to the present embodiment. In the CPU 11 of the server 1, an acquisition unit 31, a specification unit 32, a search unit 33, an output unit 34, a recording unit 35, a notification unit 36, and a score calculation unit 37 function during operation.

The acquisition unit 31 acquires the text of the user's question via the terminal 2. In this embodiment, an example will be described in which the acquisition unit 31 acquires character information input by the user as text, but for example, voice information uttered by the user may also be acquired as text. In this case, the acquired voice information is converted into text by various methods within the terminal or server. The acquisition unit 31 may also acquire text from multiple users via the terminal 2, or may acquire text from multiple users operating each terminal via multiple terminals. Note that the subject acquired by the acquisition unit 31 is not limited to the user's question, and may acquire, for example, an instruction to record desired information (an instruction to record a storage location) from the user.

The identification unit 32 identifies the predetermined purpose based on the text acquired by the acquisition unit 31. For example, the identification unit 32 extracts a "target" and a "content" from the above text by performing various natural language processing, and identifies the predetermined purpose. For example, when the text "Where is the xxx material?" is acquired, the identification unit 32 extracts "xxx material" as the "target" and "location" as the "content". Then, based on these "target" and "content", the identification unit 32 identifies that the predetermined purpose is to search for the "location" of the "xxx material". Note that the server 1 does not need to include the identification unit 32, and in this case, the identification of the predetermined purpose based on the text can be performed using an external system. Note that the predetermined purpose may be identified based on the acquired text using a classifier generated by machine learning.

In this embodiment, the predetermined purpose is information that indicates the intention of the user's question, which is identified based on the user's question. For example, in this embodiment, as described above, the predetermined purpose is identified by extracting the "target" and "content" from the user's question.

"Target" is a word indicating the subject of the search obtained from the text of the user's question, for example, "documents for XX," "application form for XX," "case for XX," "deadline for XX," etc.

"Content" refers to words that indicate the content of the search (genre, attributes) obtained from the text of the user's question, such as words indicating the person in charge, deadline, location, document data, procedure, or all of these.

The search unit 33 searches for desired information corresponding to the specified purpose identified by the identification unit 32. For example, as described above, if the specified purpose is to search for the "location" of "xxx documents", the search unit 33 searches for the "location" of "xxx documents" in the desired information DB 41, the internal company information DB 51, the external storage 52, etc., which will be described later. Note that the server 1 does not need to be equipped with the search unit 33, and in this case, the search for the desired information can be performed using an external system. Note that a classifier generated by machine learning may be used to search for (extract) the desired information corresponding to the specified purpose.

The output unit 34 outputs the search results by the search unit 33 to the terminal 2. For example, if the desired information is acquired, the output unit 34 outputs the desired information to the terminal 2. Also, for example, if the desired information is not acquired, the output unit 34 outputs a message to that effect to the terminal 2.

The recording unit 35 records the desired information based on a search log related to a specific purpose. For example, when the search unit 33 is unable to find the desired information and an instruction to record the desired information is received from the user, the recording unit 35 links information related to the specific purpose and the desired information based on the interaction with the user (search log) and records them in the desired information DB 41.
The method of issuing a recording instruction is not particularly limited, and the recording instruction may be issued by a message such as "It's here," or by a message such as "Please register XX."

When content related to a keyword (interest information) recorded in interest information DB 42 is recorded or updated, notification unit 36 issues a predetermined notification to users who are interested in that keyword. Here, it is assumed that users have previously recorded keywords of interest in interest information DB 42 by linking them to the user. For example, if "xxx materials" is recorded in interest information DB 42 as a keyword in which a user is interested, and "xxx materials" is saved or updated by another user, notification unit 36 issues a notification such as "xxx materials have been uploaded (or updated)" to users who are interested in "xxx materials."

The score calculation unit 37 calculates a score (level) indicating the degree of growth of the information search system for each predetermined organization to which multiple users belong. Here, an organization is a group to which multiple users belong. Examples of organizations include a company, a department, and an organization. The score indicating the degree of growth of the information search system is a score calculated based on, for example, the number of times recorded by the recording unit 35, the information capacity accumulated in the desired information DB 41, the number of times the desired information DB 41 was searched by the search unit 33, or a combination of these. The score calculation unit 37 records the calculated score in the growth degree DB 53. In addition to calculating the score, the score calculation unit may perform labeling such as "expert level" and "standard" and record it in the growth degree DB 53. In addition, the score calculation unit 37 may record information indicating the usage status of the information search system in the growth degree DB 53, and may calculate the above score based on the information indicating the usage status.

In one area of the memory unit 18 shown in FIG. 4, a desired information DB 41 and an interest information DB 42 are provided. In addition, as memory units external to the server 1, an internal information DB 51, an external storage 52, and a growth degree DB 53 are provided.

As described above, the desired information DB 41 is a storage unit that stores "targets," "contents," and "desired information" extracted based on a specific purpose. In this embodiment, the desired information DB 41 is described as being provided in an area of the storage unit 18 of the server 1, but it may also be provided in an area of an external storage device of the server 1.

As described above, the interest information DB 42 is a storage unit that stores, for each user, keywords that the user is interested in. In this embodiment, an example is described in which the interest information DB 42 is provided in an area of the storage unit 18 of the server 1, but it may also be provided in an area of an external storage device of the server 1.

The internal company information DB 51 is a storage unit that stores information to be displayed on an internal portal site, etc. In this embodiment, an example is described in which the internal company information DB 51 is provided in an area of an external storage device different from the server 1, but it may also be provided in an area of the storage unit 18 of the server 1.

The external storage 52 is any external storage device such as cloud storage. Note that multiple external storages 52 may be provided.

The growth degree DB53 is a storage unit in which the growth degree of the information search system is stored for each organization. The growth degree DB53 records the scores and the usage status of the information search system from each server managed by multiple organizations.

<Processing content: Information search processing>
Fig. 5 is a flowchart showing an example of information search processing according to this embodiment. In Fig. 5, an example will be described in which the server 1 and the terminal 2 communicate with each other via an arbitrary network N such as the Internet.

In step S1, terminal 2 accepts a question from user A. In this embodiment, an example will be described in which user A inputs text in a chat format, causing terminal 2 to accept the question. Note that the method of accepting a question is not limited to text input, and any method may be used. For example, user A may input a question by voice, causing terminal 2 to accept the question.

In step S2, the text of user A's question is sent from terminal 2 to server 1.

In step S3, the acquisition unit 31 acquires the text of user A's question.

In step S4, the identification unit 32 identifies the predetermined purpose based on the text acquired by the acquisition unit 31. For example, the identification unit 32 identifies the predetermined purpose by performing various natural language processing on the acquired text.

In step S5, the search unit 33 searches for desired information corresponding to the specified purpose identified by the identification unit 32. For example, the search unit 33 searches for the "location" of the desired information (e.g., xxx documents) in the desired information DB 41, the in-house information DB 51, the external storage 52, etc.

In step S6, the output unit 34 outputs the results of the search performed by the search unit 33 to the terminal 2. If the desired information is acquired, the output unit 34 outputs the desired information to the terminal 2. If the desired information is not acquired, the output unit 34 outputs a message to that effect to the terminal 2. Here, an example will be described in which the desired information (e.g., a URL indicating the storage location of the xxx material) is acquired as a search result.

In step S7, terminal 2 displays the search results output by server 1 on the output unit (e.g., a display) of terminal 2.

In step S8, terminal 2 accepts an instruction from user A to access the search results (e.g., the above URL). The method of accepting an instruction to access the search results is not particularly limited. For example, when the search results are displayed in the form of a hyperlink, user A may select the search result, thereby accepting an instruction to access. Also, for example, the user may input the search results as a chat message, thereby accepting an instruction to access. Here, an example will be described in which the user inputs the URL of the search results in a chat message.

In step S9, the terminal 2 sends the above-mentioned access instruction message to the server 1. In this case, the message is the URL of the above-mentioned search result.

In step S10, when the acquisition unit 31 acquires the above-mentioned message, the search unit 33 acquires electronic data corresponding to the URL of the above-mentioned search result. Here, it is assumed that predetermined document data is acquired as the electronic data corresponding to the URL of the search result.

In step S11, the output unit 34 transmits the electronic data acquired by the acquisition unit 31 to the terminal 2.

In step S12, the terminal 2 displays the electronic data acquired by the server 1 on an output unit (e.g., a display) of the terminal 2. The terminal 2 may open the electronic data acquired by the server 1 using a specified application.

Figure 6 is a diagram showing a display example during information search according to this embodiment. Figure 6 (A) is a diagram showing a display example of an application running on terminal 2 according to this embodiment. Information search VA (Virtual Assistant) is an agent that virtually interacts with the user. As shown in Figure 6 (A), when user A inputs "Look for xxx materials," as shown in Figure 6 (B), server 1 outputs a message such as "It's here. http://portal/xxx.ppt" to terminal 2 as information (e.g., URL) indicating the location of "xxx materials." Then, as shown in Figure 6 (C), when user A inputs "http://portal/xxx.ppt" (or selects the URL output by server 1), as shown in Figure 6 (D), electronic data corresponding to the input URL is obtained and output to terminal 2. In the application according to this embodiment, when user A responds to the above message by sending a message saying "That's right. Thank you," server 1 responds with "I understand," as shown in FIG. 6(E). In this way, by having the user input whether or not the information output by server 1 matches the information the user desires, and having server 1 manage this, the accuracy of the information search process in server 1 can be improved.

<Processing content: Information registration processing>
Fig. 7 is a flowchart showing an example of information search processing according to this embodiment. In Fig. 5, an example will be described in which the server 1 and the terminal 2 communicate with each other via an arbitrary network N such as the Internet. Note that in the above-mentioned application, multiple users can view and input messages.

Steps S21 to S27 are similar to steps S1 to S7 described above, and therefore will not be described here. Note that in the example of FIG. 7, a case will be described in which the desired information cannot be acquired in steps S25 and S26. Specifically, for example, the search results are displayed on terminal 2 in step S27, but the search results are not the information that user A desired.

In step S28, the terminal 2 accepts input of information from user B, who is different from user A. Here, user A checks whether the information input by user B matches the information desired by user A. For example, user A checks whether the information input by user B matches the information desired by user A based on the display of the information input by user B (document file name, URL, etc.). In addition, user A checks whether the information input by user B matches the information desired by user A by, for example, accessing the information input by user B (opening a document, accessing a URL, etc.). Then, when the information input by user B matches the information desired by user A, user A issues an instruction to register the information input by user B. An example of a registration instruction is when user A inputs a message such as "Register". When user A issues a registration instruction, the process of the next step S29 is executed.

In step S29, an instruction to register the desired information is received from user A. Here, an example is described in which the information entered by user B in step S28 matches the desired information of user A. Note that an instruction to register the desired information described above may also be received from user B.

In step S30, terminal 2 sends the above registration instruction message to server 1.

In step S31, when the recording unit 35 receives the above-mentioned registration instruction message, it links information about the specified purpose with the desired information and records them in the desired information DB 41.

In step S32, the output unit 34 sends a message to the terminal 2 indicating that the desired information has been recorded in the desired information DB 41.

8 is a diagram showing a display example when registering information according to this embodiment. FIG. 8(A) is the same as FIG. 6(D), and shows a display example of the state in which electronic data corresponding to the URL input by user A is acquired and output to terminal 2. A diagram showing a display example of an application operating on terminal 2 according to this embodiment. FIG. 8(B) shows a display example in the case where user A inputs "This is not it" when the search result output to terminal 2 is not the information desired by user A. Here, if user B, who is different from user A, knows "xxx material", user B can present "xxx material" to user A. For example, as shown in FIG. 8(C), user B can send a message saying "How about this material?" and attach a material. Then, as shown in FIG. 8(D), if the material attached by user B is the desired information, user A can instruct server 1 to register (record) the attached material by sending a message saying "That's right. Thank you. VA, register this." Once registration is complete, the server 1 sends a message to the terminal 2 saying "You have been registered," as shown in FIG. 8 (E).

<Processing content: Information notification processing>
Fig. 9 is a flowchart showing an example of the information notification process according to the present embodiment. In Fig. 7, the server 1, the terminal 2, and the terminal 3 communicate with each other via an arbitrary network N such as the Internet.

Steps S41 to S44 are similar to steps S29 to S32 described above, so we will not explain them here.

In step S45, if the desired information registered in step S43 (step S31) matches or is similar to a keyword in the interest information DB 42, the notification unit 36 extracts a user corresponding to a keyword that matches or is similar to the desired information. Here, it is assumed that user D is extracted as a user corresponding to the above keyword.

In step S46, the notification unit 36 notifies terminal 3, which is user D's terminal, that information in which user D is interested has been registered.

In step S47, terminal 3 displays the notification output by server 1 on an output unit (e.g., a display) of terminal 3.

In step S48, terminal 3 receives an instruction to obtain the information notified by server 1.

In step S49, terminal 3 sends a message to server 1 to obtain the above-mentioned notification information.

In step S50, when the acquisition unit 31 acquires the above-mentioned message, the search unit 33 acquires electronic data (information of interest) corresponding to the above-mentioned notification information.

In step S51, the output unit 34 transmits the electronic data acquired by the acquisition unit 31 to the terminal 3. In other words, the server 1 provides information in which the user D is interested to the user D.

Figure 10 shows an example of a display when notifying information according to this embodiment. Terminal 2 shows an example of a display when a message indicating that registration has been completed is sent from server 1 in response to a registration instruction from user A. Terminal 3 shows an example of a display when user D, who corresponds to a keyword that matches or is similar to the registered information, is notified of a message indicating that information of interest to him has been registered ("xxx materials have been uploaded"). Here, when user D inputs a message indicating that he wishes to obtain the notified information ("I would like xxx materials"), server 1 provides user D with information that is of high interest to user D along with the message "xxx materials please."

<Advantageous Effects of the Present Embodiment>
According to the above embodiment, a user can input information to be searched for into a chat, and if the information is already registered in the server, the information can be output. In other words, the user can efficiently search for the desired information without being aware of the location of the desired information. In addition, the user can search for the desired information without being aware of the system in which the desired information is stored.

In addition, if the desired information is not registered on the server, the user who asked the question or another user can tell the server the information they are looking for (issue a registration instruction), and from the next time onwards, the latest information can be output. This makes it possible to increase the amount of information stored in the information search system. It also makes it possible to provide an information search environment that grows and evolves autonomously.

In addition, as described above, the server can record desired information based on a search log for a specific purpose, reducing the operational burden on administrators of existing systems (e.g., in-house portals).

In addition, by entering documents and other items that users wish to share with other users into the chat, the documents can be automatically stored in the appropriate location. Also, by registering keywords related to certain information in advance on the server, users who need certain information can receive notifications from the server when information related to the keywords is stored, saved, or updated.

In addition, by calculating a score indicating the degree of growth of the information search system for each specified organization, it is possible to compare usage with other companies and other departments, etc., and confirm the degree to which the information search system has grown, which can lead to improving the company's strength. In addition, by calculating and comparing the above scores and adding gamification elements, it is possible to promote business while empowering companies in a playful way.

Although one embodiment of the present invention has been described above, the present invention is not limited to the above-mentioned embodiment, and modifications and improvements that can achieve the object of the present invention are included in the present invention.

(Modification)
In the above embodiment, an example was described in which the desired information was recorded based on a search log related to a predetermined purpose when the server was unable to search for the desired information, but the desired information may be recorded regardless of whether the desired information was unable to be searched. For example, when the desired material was searched for and a new additional material was created in addition to the desired material, the server may record the desired information (additional material) based on a search log related to a predetermined purpose by attaching the material with a message saying "Please save the material for XX." Also, for example, when the desired material was searched for and a change was made to the desired material, the server may record (update) the desired information based on a search log related to a predetermined purpose by attaching the material with a message saying "Please update the material for XX."

In the above embodiment, an example of identifying a specific purpose by extracting the "target" and "content" from the user's question has been described, but there are cases where the "target" and "content" are not uniquely identified. For example, when a question such as "I would like to apply for xxx" is input, the server can identify the "target" as "xxx application", but cannot uniquely identify the "content". In such a case, for example, the server may output the location of the "xxx application" application form (or the application form data itself), such as "here is the xxx application form". In addition, for example, the server may ask a question asking about the "content", such as "What content would you like for the xxx application?" In addition, for example, the server may output all information related to the "xxx application". Here, all information includes, for example, the "location of the application form", "location of the application method or application procedure", "application deadline", "application window (person in charge)", etc. in the "xxx application".

In the above embodiment, an example has been described in which there is only one search result for desired information corresponding to a specific purpose. Specifically, for example, the search unit acquires desired information corresponding to a specific purpose and the accuracy of the desired information as a search result. Then, the search unit outputs the desired information with the highest accuracy as a search result. In this way, even if there are multiple search results for desired information corresponding to a specific purpose, the search unit can output one search result.
Here, the accuracy is information indicating the degree of match (score) between the desired purpose and the desired information. The accuracy can be obtained, for example, by using the classifier described above.
The detection unit may also adjust the acquired accuracy. For example, the search unit may adjust the accuracy to be higher for documents that are searched for frequently. The search unit may also adjust the accuracy based on a user's instruction (for example, pressing a button, posting to a chat, etc.). For example, the search unit may adjust the accuracy to be higher when the user responds to the output search results that "it's correct." The search unit may also adjust the accuracy to be lower when the user responds to the output search results that "it's not what I expected."
Also, for example, when there are multiple search results for desired information corresponding to a specific purpose, the search unit displays the multiple search results (e.g., multiple search results with a specific accuracy or higher) on the user's terminal. Then, when the user selects any information, the search unit may adjust the accuracy of the selected information to be higher (e.g., in the next or subsequent search process).
The server may output to the terminal a question asking which of the multiple detection results should be output. In addition, when there are multiple search results for desired information corresponding to a specific purpose, the server may output all of the multiple detection results to the terminal.

In the above embodiment, an example was described in which desired information based on text information entered on a chat was recorded as a search log for a specific purpose, but the search log is not limited to the above. For example, a system log may be used as the search log. In this case, for example, when multiple files are output to a terminal as search results, the server may measure the viewing time of the user and identify and record (register) the file that was viewed the longest among the viewed files as the desired information.

In the above embodiment, an example was described in which a user inputs a question by inputting characters into a chat-style application, but the method of inputting a question is not particularly limited. For example, a user may input a question by speaking into a terminal.

In the above embodiment, a smartphone or other terminal is used as an example of the terminal, but the terminal is not particularly limited, and any terminal such as a PC or tablet may be used.

In the above embodiment, in FIGS. 7 and 8, an example was described in which desired information provided by another user (user B) different from the user who asked the question (user A) is recorded in association with information related to a specified purpose. However, this is not limited to this, and desired information provided by the user who asked the question may be recorded in association with information related to a specified purpose. For example, it is conceivable that, if the user who asked the question is unable to obtain the desired information, the user may instruct the server to register materials that the user has researched and discovered as desired information.

For example, the above-mentioned series of processes can be executed by hardware or software. In other words, the functional configuration of FIG. 4 is merely an example and is not particularly limited. In other words, it is sufficient that the information processing system is provided with a function that can execute the above-mentioned series of processes as a whole, and the type of functional block used to realize this function is not particularly limited to the example of FIG. 4. The location of the functional block is also not particularly limited to that of FIG. 4 and can be arbitrary. For example, the functional block of the server may be transferred to another terminal, etc. Conversely, the functional block of the terminal may be transferred to the server, etc. Furthermore, one functional block may be configured as hardware alone, software alone, or a combination of both.

When a series of processes is executed by software, the programs that make up the software are installed on a computer or the like from a network or a recording medium. The computer may be a computer that is built into dedicated hardware. The computer may also be a computer that is capable of executing various functions by installing various programs, such as a server, a general-purpose smartphone, or a personal computer.

The recording medium containing such a program may be configured not only as a removable medium (not shown) that is distributed separately from the device body in order to provide the program to the user, but also as a recording medium that is provided to the user in a state where it is already installed in the device body.

In this specification, the steps describing the program to be recorded on the recording medium include not only processes that are performed chronologically according to the order, but also processes that are not necessarily performed chronologically but are performed in parallel or individually. In addition, in this specification, the term "system" refers to an overall device that is composed of multiple devices, multiple means, etc.

1: Server 2: Terminal 3: Terminal 11: CPU 18: Storage unit 19: Communication unit 31: Acquisition unit 32: Identification unit 33: Search unit 34: Output unit 35: Recording unit 36: Notification unit 37: Score calculation unit 41: Desired information DB 42: Interest information DB
51: Internal information DB 52: External storage 53: Growth degree DB

Claims (11)

a obtaining means for obtaining the text of a user's question;
An identification means for identifying a predetermined purpose based on the text;
a search means for searching for desired information corresponding to the predetermined purpose and obtaining a degree of accuracy indicating a degree of agreement between the predetermined purpose and the desired information;
a recording means for recording information related to the predetermined purpose and the desired information in association with each other based on a search log related to the predetermined purpose;
an output means for outputting the search results;
a calculation means for calculating a score for each of a plurality of users based on information indicating a usage situation of the information retrieval system ;
having
When there are a plurality of pieces of desired information having a predetermined accuracy or higher as the desired information corresponding to the predetermined purpose, the search means outputs the plurality of pieces of desired information to the output means as search results, and when any of the desired information is selected by a user, the search means increases the accuracy of the selected desired information in the next or subsequent search processes .
Information retrieval system. a obtaining means for obtaining the text of a user's question;
An identification means for identifying a predetermined purpose based on the text;
a search means for searching for desired information corresponding to the predetermined purpose and obtaining a degree of accuracy indicating a degree of agreement between the predetermined purpose and the desired information;
a recording means for recording information related to the predetermined purpose and the desired information in association with each other based on a search log related to the predetermined purpose;
an output means for outputting the search results;
having
the search means causes the output means to output only the most accurate desired information from among the desired information corresponding to the predetermined purpose as a search result;
If the output desired information is the desired information expected by the user, the accuracy of the desired information is increased in the next or subsequent search process , and if the output desired information is not the desired information expected by the user, the accuracy of the desired information is decreased in the next or subsequent search process .
Information retrieval system. the specifying means extracts a search target and a content from the text, and if the content cannot be extracted from the text, asks a user a question about the content;
3. An information retrieval system according to claim 1 or 2. If the search means is unable to search for the desired information, the recording means records the desired information based on a search log related to the predetermined purpose.
4. An information retrieval system according to claim 1. When the search means is unable to search for the desired information, the output means outputs a message indicating that the desired information is not available;
When the acquisition means receives an instruction to record the desired information from the user who has asked the question or another user, the recording means records the information regarding the predetermined purpose and the desired information in association with each other in the storage means.
An information retrieval system according to any one of claims 1 to 4. The acquiring means acquires the text from a plurality of users;
When the acquisition means receives an instruction to record the desired information from the user who asked the question or another user, the recording means links the information regarding the predetermined purpose with the desired information provided by the other user different from the user who asked the question and records them in the storage means.
An information retrieval system according to any one of claims 1 to 5. The method further comprises: recording in advance in a storage means a keyword in which a user is interested for each user; and, when content related to the keyword in which the user is interested is recorded or updated, issuing a predetermined notification to the user corresponding to the keyword in which the user is interested.
An information retrieval system according to any one of claims 1 to 6. The process to be executed by an information retrieval system constructed using a computer is as follows:
an acquisition step for acquiring a text of a user's question;
identifying a predetermined purpose based on the text;
a search step of searching for desired information corresponding to the predetermined purpose and acquiring a degree of accuracy indicating a degree of agreement between the predetermined purpose and the desired information;
an output step for outputting the search results;
a recording step of recording information related to the predetermined purpose and the desired information in association with each other based on a search log related to the predetermined purpose ;
a calculation step of calculating a score for each of a plurality of users based on information indicating a usage situation of the information retrieval system ;
having
In the search step, if there are a plurality of pieces of desired information having a predetermined accuracy or higher as the desired information corresponding to the predetermined purpose, the plurality of pieces of desired information are output as search results in the output step, and when any of the desired information is selected by the user, the accuracy of the selected desired information is increased in the next or subsequent search processes .
A method for controlling an information retrieval system. The process to be executed by an information retrieval system constructed using a computer is as follows:
a obtaining step for obtaining the text of a user's question;
identifying a predetermined purpose based on the text;
a search step of searching for desired information corresponding to the predetermined purpose and acquiring a degree of accuracy indicating a degree of agreement between the predetermined purpose and the desired information;
an output step for outputting the search results;
a recording step of recording information related to the predetermined purpose and the desired information in association with each other based on a search log related to the predetermined purpose ;
having
In the search step, only the most accurate desired information is output as a search result from among the desired information corresponding to the predetermined purpose in the output step;
If the output desired information is the desired information expected by the user, the accuracy of the desired information is increased in the next or subsequent search process , and if the output desired information is not the desired information expected by the user, the accuracy of the desired information is decreased in the next or subsequent search process .
A method for controlling an information retrieval system. an acquisition step for acquiring a text of a user's question;
identifying a predetermined purpose based on the text;
a searching step of searching for desired information corresponding to the predetermined purpose and acquiring a degree of accuracy indicating a degree of agreement between the predetermined purpose and the desired information;
an output step for outputting the search results;
a recording step of recording information related to the predetermined purpose and the desired information in association with each other based on a search log related to the predetermined purpose ;
a calculation step of calculating a score for each of a plurality of users based on information indicating a usage situation of the information retrieval system ;
A computer program for causing a computer to execute the following:
In the search step, if there are a plurality of pieces of desired information having a predetermined accuracy or higher as the desired information corresponding to the predetermined purpose, the plurality of pieces of desired information are output as search results in the output step, and when any of the desired information is selected by the user, the accuracy of the selected desired information is increased in the next or subsequent search processes .
Computer program. a obtaining step for obtaining the text of a user's question;
identifying a predetermined purpose based on the text;
a searching step of searching for desired information corresponding to the predetermined purpose and acquiring a degree of accuracy indicating a degree of agreement between the predetermined purpose and the desired information;
an output step for outputting the search results;
a recording step of recording information related to the predetermined purpose and the desired information in association with each other based on a search log related to the predetermined purpose ;
A computer program for causing a computer to execute the following:
In the search step, only the most accurate desired information is output as a search result from among the desired information corresponding to the predetermined purpose in the output step;
If the output desired information is the desired information expected by the user, the accuracy of the desired information is increased in the next or subsequent search process , and if the output desired information is not the desired information expected by the user, the accuracy of the desired information is decreased in the next or subsequent search process .
Computer program.

Images