JP7664724B2 - Search suggestion management device, search suggestion management system, search suggestion management method, and program - Google Patents

Description

The present invention relates to a search suggestion management device, a search suggestion management system, a search suggestion management method, and a program.

There are various technologies that present search query suggestions to users to assist them in entering search queries.

Patent Document 1 discloses a search suggestion technology. Paragraph [0095] states, "The search suggestion device of this embodiment can appropriately suggest (offer) search query candidates according to the attributes of the user who inputs the search query (including access authority, etc.). In other words, search query candidates are not unnecessarily displayed for documents to which access is not permitted based on the user attributes."

JP 2015-106354 A

When presenting search query candidates, if search query candidates that meet the user's wishes can be presented, it will contribute to improving search efficiency. However, there are situations where skill is required to come up with an appropriate search query. In particular, in knowledge sharing situations where technical information is shared, it is desirable to make effective use of the search techniques of skilled people.

The search suggestion technology described in Patent Document 1 presents search query candidates according to the user's access authority, and does not envision utilizing the search techniques of skilled individuals.

The present invention has been made in consideration of the above points, and aims to provide a technology that makes effective use of search skills to provide search suggestions.

This application includes multiple means for solving the above problems, examples of which are as follows:

In order to solve the above problem, a search suggestion management device according to one embodiment of the present invention is characterized by comprising: a search log acquisition unit that acquires a user's search log; a score calculation unit that calculates a score based on the user's search behavior for a search query candidate including one or more search queries input by the user; a suggestion index generation unit that generates suggestion index information that associates the search query candidate, the user, and the score; a tutoring target identification unit that identifies a tutoring target for another user to study when searching; and a suggestion extraction unit that, upon acquiring a search query input by the other user, extracts the search query candidate from the suggestion index information using the tutoring target user, the input search query, and the score.

The search suggestion management device may include a search success/failure determination unit that refers to the search log and determines whether the user's search behavior is successful, and the score calculation unit may calculate the score for the search query candidate related to the successful search behavior to be higher than the scores of the other search query candidates.

The search log may include information output as a search result and an operation log for the information, and the search success/failure determination unit may determine that the search action leading to obtaining the search result was successful if the operation log satisfies a predetermined search success condition.

The score calculation unit may be characterized in that it refers to the search log, generates search behavior information in which one search behavior is associated with one or more input operations, and when calculating the score for the search query candidate, if there is another input operation associated with the same search behavior as the input operation related to the search query candidate, it calculates the score according to the other input operation.

The suggestion extraction unit may be characterized in that it associates a ranking based on the score with the extracted search query candidates and outputs them.

The search log acquisition unit may acquire the search log including at least one of a search target period or a search target category, and when at least one of the search target period or the search target category is input, the score calculation unit calculates the score for a combination of the search query candidate and at least one of the input search target period or the search target category, and when the search query candidate extracted using the score is associated with the search target period or the search target category, the suggestion extraction unit may output the associated search target period or the search target category together with the search query candidate.

The search suggestion management device may include a deleted information identification unit that identifies deleted information that has been deleted from the information to be searched, and the suggestion extraction unit may be characterized in that, when the operation log related to the deleted information satisfies the search success condition, the search suggestion extraction unit suppresses extraction of the search query candidate that searches the deleted information.

In order to solve the above-mentioned problems, a search suggestion management system according to another aspect of the present invention is a search suggestion management system having a terminal device, a search server device, and a search suggestion management device, which are connected to each other so as to be able to communicate with each other, wherein the terminal device has a search request unit that receives a search query input from a user and transmits a search request to the search server device, and an output unit that receives and displays search query candidates from the search suggestion management device, the search server device has a search unit that searches for information to be searched using the search query related to the search request, and the search suggestion management device has a search log acquisition unit that acquires a search log of the user, and a search log including one or more search queries input by the user. The terminal device includes a score calculation unit that calculates a score for a search query candidate based on the search behavior of the user, a suggestion index generation unit that generates suggestion index information that associates the search query candidate, the user, and the score, a target identification unit that identifies a target under which the other user will study when searching, and a suggestion extraction unit that, upon receiving a search query input from the other user, extracts the search query candidate from the suggestion index information using the target user, the input search query, and the score, and the output unit of the terminal device displays the search query candidate extracted by the suggestion extraction unit of the search suggestion management device.

In order to solve the above problem, a search suggestion management method according to another aspect of the present invention is a search suggestion management method by a search suggestion management device, and is characterized by comprising: a search log acquisition step of acquiring a user's search log; a score calculation step of calculating a score based on the user's search behavior for a search query candidate including one or more search queries input by the user; a suggestion index generation step of generating suggestion index information that associates the search query candidate, the user, and the score; a mentor identification unit that identifies a mentor under which another user will study when searching; and a suggestion extraction step of, when a search query input by the other user is obtained, extracting the search query candidate from the suggestion index information using the user under the mentor, the input search query, and the score.

In order to solve the above problem, a program according to another aspect of the present invention is a program for causing a processing unit of a computer to execute a search suggestion management method, and is characterized in that the program executes a search log acquisition step for acquiring a user's search log, a score calculation step for calculating a score based on the search behavior of the user for a search query candidate including one or more search queries input by the user, a suggestion index generation step for generating suggestion index information that associates the search query candidate, the user, and the score, a mentor identification unit for identifying a mentor under which another user will study when searching, and a suggestion extraction step for extracting the search query candidate from the suggestion index information using the user under the mentor, the input search query, and the score when a search query input by the other user is obtained.

The present invention provides a technology that makes effective use of search skills to provide search suggestions.

Problems, configurations, and advantages other than those described above will become clear from the description of the embodiments below.

FIG. 1 illustrates an example of an overview of a search suggestion management system. FIG. 2 is a diagram illustrating an example of functional blocks of a search suggestion management system. FIG. 4 is a diagram illustrating an example of a data structure of search history information. FIG. 2 is a diagram illustrating an example of a data structure of search behavior information. FIG. 13 is a diagram illustrating an example of a data structure of suggested index information. FIG. 2 illustrates an example of a hardware configuration of the suggestion management device 100. 13 is a flowchart illustrating an example of a search behavior information generating process. 13 is a flowchart illustrating an example of a suggestion index generating process. 13 is a flowchart illustrating an example of a suggestion providing process. FIG. 13 is a diagram illustrating an example of a data structure of suggested index information in a first modified example. 11A is a diagram showing an example of a search screen in this embodiment, FIG. 11B is an example (part 1) of the search screen in the first modified example, and FIG. 11C is an example (part 2) of the search screen in the first modified example. 13 is an example of a functional block of a search suggestion management system according to a second modified example. 13 is a flowchart illustrating an example of a suggested index generating process according to a second modified example.

Below, an example of an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing an example of an outline of a search suggestion management system 1. The search suggestion management system 1 in this embodiment includes a search suggestion management device 100, a search server device 200, and a terminal device 300, and each device is connected to each other so that they can communicate with each other.

The search suggestion management device 100 is a server computer, a PC, or the like, and presents search query suggestions to a user performing a search. The search server device 200 is a server computer, a PC, or the like, and searches for document index information 221 to be searched in response to a search request from a terminal device 300, and outputs the search results. The terminal device 300 is a PC, smartphone, PDA (Personal Digital Assistant), notebook computer, tablet terminal, or the like, and is operated by a user.

First, when a user inputs a search query into the search window 351 of the terminal device 300, the terminal device 300 transmits the search query to the search suggestion management device 100 (FIG. 1 (1)). The user also inputs, to the terminal device 300, information indicating another user who will be a mentor for the search. As an example, a mentor is someone who serves as a reference for the user when searching. The terminal device 300 transmits the mentor to the search suggestion management device 100 (FIG. 1 (2)).

The search suggestion management device 100 receives the search query sent from the terminal device 300 (FIG. 1 (3)). The search suggestion management device 100 then refers to the suggestion index information 123 described below, and extracts search query candidates to be presented to the terminal device 300 as suggestions using the received search query, other users who will be trained, and a score calculated based on the search behavior of the users who will be trained (FIG. 1 (4)).

Then, the search suggestion management device 100 transmits the extracted suggestions to the terminal device 300 (FIG. 1 (5)). Hereinafter, the process from "(3) Receiving a search query" to "(5) Responding to a suggestion" executed by the search suggestion management device 100 will be described as a suggestion providing process. When the terminal device 300 receives the suggestions, the search query candidates are displayed on the terminal device 300.

The user of the terminal device 300 refers to the displayed search query candidates to determine a search query to execute a search, enters it in the search window 351, and presses the search button. That is, the terminal device 300 accepts the input operation of a search instruction related to the determined search query. In response to the input operation of the search instruction, the terminal device 300 transmits a search request to the search server device 200 (FIG. 1 (6)). The search server device 200 extracts documents corresponding to the search query from the document index information 221, and transmits information identifying the extracted documents as search results to the terminal device 300 (FIG. 1 (7)). The terminal device 300 accepts input operations such as selection operations and viewing operations related to the extracted documents.

The search server device 200 transmits a search log that records the processing performed in response to a search request from the terminal device 300 to the search suggestion management device 100 (FIG. 1 (8)). The search suggestion management device 100 refers to the search log and determines whether the user's search behavior was successful (FIG. 1 (9)). The search suggestion management device 100 calculates a score for each search query candidate based on the user's search behavior (FIG. 1 (10)).

The search suggestion management device 100 calculates a higher score for a search query candidate that constitutes a successful search behavior than for other search query candidates. This allows a higher score to be given to a search query when a user who is the target of training succeeds in a search. The higher the score, the more valid the search result presented to the user is considered to be. This allows a user performing a search to refer to the search behavior of a user who is the target of training when the search is successful.

The search suggestion management device 100 generates suggestion index information 123 by associating the user who performed the search with the score for the search query candidate (indexing: FIG. 1 (11)). Hereinafter, the processes from "(8) Log reception" to "(11) Indexing" executed by the search suggestion management device 100 will be described as the suggestion index generation process.

Figure 2 is a diagram showing an example of functional blocks of the search suggestion management system 1. The search suggestion management device 100, the search server device 200, and the terminal device 300 are connected via a network N. The search suggestion management system 1 has multiple terminal devices 300, including a terminal device 300 owned by a user who performs a search, and a terminal device 300 owned by a user who is to be trained.

The search suggestion management device 100 includes a processing unit 110, a storage unit 120, an input unit 130, an output unit 140, and a communication unit 150. The processing unit 110 controls the entire search suggestion management device 100. The storage unit 120 stores information necessary for processing by the processing unit 110. The input unit 130 accepts information input to the search suggestion management device 100 from an input device connected via an input IF 14 described later. The output unit 140 outputs information stored in the search suggestion management device 100 to an output device connected via an output IF 15 described later. The communication unit 150 mediates the transmission and reception of information to and from the search server device 200 and the terminal device 300.

The processing unit 110 includes a search log acquisition unit 111, a search success/failure determination unit 112, a score calculation unit 113, a suggestion index generation unit 114, a training target identification unit 115, and a suggestion extraction unit 116. The search log acquisition unit 111 acquires a search log transmitted from the terminal device 300. The search log includes information identifying a document output by the search server device 200 as a search result, and an operation log for the document. The search log acquisition unit 111 uses the acquired search log to generate search history information 121, which will be described later.

The search success determination unit 112 refers to the search history information 121 generated from the search log and determines whether the user's search behavior was successful. When the search success determination unit 112 determines that the operation log included in the search log satisfies a predetermined search success condition, it determines that the search behavior leading to obtaining the search results was successful. For example, the search success determination unit 112 determines that the search behavior was successful when the document output as the search result was viewed for a predetermined period of time or more. Alternatively, the search success determination unit 112 determines that the search behavior was successful when the document was scrolled to the end. Alternatively, the search success determination unit 112 determines that the search behavior was successful when the terminal device 300 receives an input operation indicating a search success. Note that the search success condition is not limited to this example.

The score calculation unit 113 calculates a score based on the user's search behavior for a search query candidate that includes one or more search queries input by the user. The score calculation unit 113 references the search log and generates search behavior information 122 (described later) that associates one or more input operations with one search behavior. For example, the score calculation unit 113 calculates a higher score for a search query candidate related to a successful search behavior than other search query candidates. In addition, for example, when calculating a score for a search query candidate, if there is another input operation related to the same search behavior as the input operation related to the search query candidate, the score calculation unit 113 calculates the score according to the other input operation.

The suggestion index generation unit 114 generates suggestion index information 123 that associates search query candidates including the search query acquired by the search log acquisition unit 111, the user who performed the search, and a score.

The mentoring target identification unit 115 identifies other users who will be mentors of the user performing the search. For example, the mentoring target identification unit 115 identifies the mentor input to the terminal device 300, as described using FIG. 1. The mentoring target identification unit 115 may refer to mentoring information (not shown) stored in the memory unit 120 to identify the mentor of the user performing the search. As an example, the mentoring information is organizational information, and the practical mentor (instructor) of the user performing the search can be the mentor.

When the suggestion extraction unit 116 acquires a search query input by a user to the terminal device 300, it extracts search query candidates from the suggestion index information 123 using other users to be trained, the input search query, and the score. For example, the suggestion extraction unit 116 associates a ranking based on the score with the extracted search query candidates and transmits them to the terminal device 300. For example, the suggestion extraction unit 116 may extract a predetermined number of search query candidates based on the score.

The storage unit 120 stores search history information 121, search behavior information 122, and suggest index information 123. The search history information 121 is information generated using a search log, and input operations accepted by the terminal device 300 are recorded. The search behavior information 122 is information generated based on the search history information 121, and one or more input operations are associated with one search behavior. The suggest index information 123 is information that associates search query candidates to be presented to a user performing a search, the user, and a score calculated based on the user's search behavior.

The search server device 200 comprises a processing unit 210, a storage unit 220, an input unit 230, an output unit 240, and a communication unit 250. The processing unit 210 provides overall control of the entire search server device 200. The storage unit 220 stores information necessary for processing by the processing unit 210. The input unit 230 accepts information input from an input device to the search server device 200. The output unit 240 outputs information stored in the search server device 200 to an output device. The communication unit 250 mediates the transmission and reception of information between the search suggestion management device 100 and the terminal device 300.

The processing unit 210 includes a search unit 211. In response to a search request from the terminal device 300, the search unit 211 searches document index information 221, which is a list of documents to be searched, and extracts information that identifies a document corresponding to the search query included in the search request as a search result. The search unit 211 searches the document index information 221 using known means, and therefore a description thereof will be omitted.

The storage unit 220 stores document index information 221. The document index information 221 is information that includes a list of documents to be searched in the search server device 200. As an example, the documents included in the document index information 221 are knowledge information shared by each user. Note that in this embodiment, the search target is not limited to documents, and may be any searchable information, but for convenience, an example in which the search target is documents will be described.

The terminal device 300 includes a processing unit 310, an input unit 320, an output unit 330, and a communication unit 340. The processing unit 310 controls the entire terminal device 300. The input unit 320 accepts information input from an input device to the terminal device 300. The output unit 330 outputs information stored in the terminal device 300 to the output device. The output unit 330 displays a search screen including search query candidates extracted by the search suggestion management device 100. The communication unit 340 mediates the transmission and reception of information between the search suggestion management device 100 and the search server device 200.

The processing unit 310 includes a search request unit 311. The search request unit 311 accepts a search query input from a user and transmits a search request to the search server device 200.

Figure 3 is a diagram showing an example of the data structure of search history information 121. Search history information 121 is information generated based on a search log received from terminal device 300. It has a user ID 121a, a search date and time 121b, a search query 121c, a search result 121d, a selected document 121e, a viewing start date and time 121f, a viewing end date and time 121g, and an operation log 121h.

The user ID 121a is identification information that identifies the user of the terminal device 300. The search date and time 121b is information indicating the date and time when the user identified by the user ID 121a input a search instruction. The search query 121c is information indicating the search query that was input when the search request was made. The search result 121d is information indicating the search result transmitted from the search server device 200 to the terminal device 300 in response to the search request transmitted from the terminal device 300 to the search server device 200.

The selected document 121e is information that identifies a document selected by the user from the search results 121d. The viewing start date and time 121f is information that indicates the date and time when viewing of the document related to the selected document 121e began. The viewing end date and time 121g is information that indicates the date and time when viewing of the document related to the selected document 121e ended. The operation log 121h is information that identifies an operation performed on the document related to the selected document 121e. For example, the operation log 121h may be information that can identify whether the document was viewed to the end, or may be information that identifies that an input operation indicating that viewing of the document has ended has been received, or may be information that identifies that an input operation indicating that the search was successful has been received.

As shown in FIG. 3, the search history information 121 is generated in response to an input operation to the terminal device 300. Note that the viewing start date and time 121f and the viewing end date and time 121g are information corresponding to an operation on a document, and therefore can be treated as an operation log.

Figure 4 is a diagram showing an example of the data structure of the search behavior information 122. The search behavior information 122 has a search behavior ID 122a, a user ID 122b, a search query 122c, a reference document 122d, and a success flag 122e. The search behavior information 122 is generated in response to an input operation by the user.

Search behavior ID 122a is identification information that identifies the search behavior. User ID 122b is identification information that identifies the user who performed the search. Search query 122c is information that indicates the search query included in the search request. Reference document 122d is information that indicates the document selected by the user from the search results. Success flag 122e is flag information that identifies whether the search was successful or not. For example, if success flag 122e is "0", it indicates that the operation of searching for reference document 122d failed, and if success flag 122e is "1", it indicates that the operation of searching for reference document 122d was successful.

When a user starts a search, one or more input operations are performed before the user reaches the desired document. In this embodiment, a series of input operations from the start to the end of a search is treated as one search behavior. In the search behavior information 122, one or more input operations on the terminal device 300 are associated with one search behavior ID 122a.

Figure 5 is a diagram showing an example of the data structure of the suggestion index information 123. The suggestion index information 123 has a search query 123a, a user ID 123b, and a score 123c.

Search query 123a is information indicating a search query entered by a user identified by user ID 123b, and is treated as a search query candidate. User ID 123b is information identifying the user who performed the search. Score 123c is a numerical value calculated based on the search behavior for search query 123a. As described above, the higher the numerical value of score 123c, the more likely the search result to be presented is to be treated as being relatively valid.

Figure 6 is a diagram showing an example of the hardware configuration of the search suggestion management device 100. The search suggestion management device 100 includes a calculation device 11, a memory 12, an external storage device 13, an input IF (Interface) 14, an output IF 15, and a communication IF 16, and each component is connected by a bus.

The arithmetic unit 11 is a processing device such as a CPU (Central Processing Unit) and executes processing according to a program recorded in the memory 12 or the external storage device 13. In the search suggestion management device 100, processing is performed by the arithmetic unit 11 which operates according to a program read onto the memory 12 or the external storage device 13. The processing unit 110 realizes each function by the arithmetic unit 11 executing the program.

The memory 12 is a storage device such as a RAM (Random Access Memory) or a flash memory, and functions as a storage area from which programs and data are temporarily read. The external storage device 13 is, for example, a writable and readable storage medium and storage media drive, such as a HDD (Hard Disk Drive), a CD-R (Compact Disc-Recordable), or a DVD-RAM (Digital Versatile Disk-Random Access Memory). The functions of the storage unit 120 are realized by the memory 12 or the external storage device 13. Note that the functions of the storage unit 120 may also be realized by a storage device connected via the communication IF 16.

The input IF 14 is an interface for accepting input operations from an operator, and is connected to input devices such as a touch panel, keyboard, mouse, and microphone. The output IF 15 is an interface for outputting information to an output device such as an LCD (Liquid Crystal Display) display built into the search suggestion management device 100.

The communication IF 16 is an interface for connecting the search suggestion management device 100 to the network N, and is, for example, a communication device such as a LAN (Local Area Network) card. The search suggestion management device 100 may have a storage medium drive device (not shown) that inputs and outputs information from portable media.

The processing of each component of the search suggestion management device 100 may be executed by a single piece of hardware or by multiple pieces of hardware. The processing of each component of the search suggestion management device 100 may be realized by a single program or by multiple programs. One or more processes performed by the search suggestion management device 100 may be distributed and executed on physically separate pieces of hardware connected by a communication means.

The hardware configuration of the search server device 200 and the terminal device 300 is similar to that of the search suggestion management device 100, so a description is omitted.

FIG. 7 is a flowchart showing an example of a search behavior information generation process. This process is started, for example, when the search suggestion management device 100 receives information indicating that a search query has been entered into a search screen displayed on the terminal device 300. Note that this process may be executed, for example, periodically in the search suggestion management device 100.

First, the search log acquisition unit 111 receives a search log (step S101). Specifically, the search log acquisition unit 111 receives a search log transmitted from the terminal device 300. The search log acquisition unit 111 uses the acquired search log to generate the search history information 121 shown in FIG. 3.

Next, the search success determination unit 112 identifies the record with the most recent search date and time from among the unconfirmed records (step S102). Specifically, the search success determination unit 112 identifies one record from among the records in the search history information 121 that has not yet undergone the process shown in this figure and has the most recent search date and time 121b.

Next, the search success determination unit 112 determines whether a document has been selected (step S103). Specifically, the search success determination unit 112 references the selected document 121e of the record selected in step S102, and determines whether an input operation to select a document has been received. The search success determination unit 112 determines that a document has been selected if the selected document 121e contains information indicating a document.

When the search success determination unit 112 determines that a document has been selected (step S103: YES), the search success determination unit 112 refers to the operation log of the selected document (step S104). Specifically, the score calculation unit 113 refers to the viewing start date and time 121f, the viewing end date and time 121g, and the operation log 121h of the record selected in step S102.

Next, the search success determination unit 112 determines whether the search success condition is satisfied (step S105). Specifically, the search success determination unit 112 determines whether the operation log referenced in step S104 satisfies a predetermined search success condition. For example, the search success determination unit 112 calculates the viewing time of the document using the viewing start date and time 121f and the viewing end date and time 121g, and determines that the search success condition is satisfied if the viewing time is equal to or longer than a predetermined time. Alternatively, the search success determination unit 112 determines that the search success condition is satisfied if there is an operation log indicating that the document has been scrolled to the end. Alternatively, the search success determination unit 112 determines that the search success condition is satisfied if there is an operation log indicating that an input operation has been received indicating that the document that is the target of the search has been presented as a search result. Note that the search success condition is not limited to this example.

When the search success determination unit 112 determines that the search success condition is not met ("NO" in step S105), the score calculation unit 113 generates search behavior information 122 by associating a failure flag (step S106). Specifically, the score calculation unit 113 associates the user ID 121a, search query 122c, and selected document 121e of the search history information 121 selected in step S102 as user ID 122b, search query 122c, and reference document 122d, respectively, and generates a record of search behavior information 122 with the success flag 122e as information indicating failure. Note that the search behavior ID 122a of the generated record does not include information indicating a search behavior.

In step S105, if the search success determination unit 112 determines that the search success condition is met (if "YES" in step S105), the score calculation unit 113 generates search behavior information 122 by associating a success flag (step S107). Specifically, the score calculation unit 113 associates the user ID 121a, search query 122c, and selected document 121e of the search history information 121 selected in step S102 as user ID 122b, search query 122c, and reference document 122d, respectively, and generates a record of search behavior information 122 with the success flag 122e as information indicating success. Note that the search behavior ID 122a of the generated record does not include information indicating the search behavior.

In step S103, if the search success/failure determination unit 112 determines that a document is not selected (if "NO" in step S103), the search success/failure determination unit 112 proceeds to step S109.

After step S106, or after step S107, or if the result of step S103 is "NO", the search success/failure determination unit 112 determines whether or not there are any unconfirmed records (step S109). Specifically, the search success/failure determination unit 112 determines whether or not there are any records in the search history information 121 generated in step S101 that have not been identified in step S102. If the search success/failure determination unit 112 determines that there are any unconfirmed records (if the result of step S109 is "YES"), the search success/failure determination unit 112 transitions the process to step S102.

When the search success/failure determination unit 112 determines that there are no unconfirmed records ("NO" in step S109), the search success/failure determination unit 112 determines whether the search has ended (step S110). Specifically, the search success/failure determination unit 112 determines whether the search history information 121 satisfies a predetermined search end condition. For example, the search success/failure determination unit 112 refers to the search behavior information 122, and when there is a record to which the search behavior ID 122a has not been assigned and the success flag 122e is information indicating success, it determines that the user has reached the target document and that the search end condition is satisfied.

For another example, the search success/failure determination unit 112 determines that the search end condition is met when a predetermined time has elapsed since the search date/time 121b of the most recent record in the search history information 121. For another example, the search success/failure determination unit 112 refers to the search history information 121 and determines that the search end condition is met when the search history information 121 contains an operation log 121h indicating an input operation to close the search screen. Note that the search end condition is not limited to the example shown here.

If the search success/failure determination unit 112 determines that the search has not ended ("NO" at step S110), the process returns to step S109.

When the search success/failure determination unit 112 determines that the search has ended ("YES" in step S110), the score calculation unit 113 assigns a search behavior ID to the generated search behavior information 122 (step S111). Specifically, the score calculation unit 113 assigns a common search behavior ID 122a to one or more records of the search behavior information 122 generated in steps S106 and S107. Thereafter, the processing unit 110 of the search suggestion management device 100 ends the processing of this flowchart.

Figure 8 is a flowchart showing an example of a suggestion index generation process. This process is executed, for example, when the search behavior information generation process shown in Figure 7 is completed.

First, the suggestion index generation unit 114 identifies a search behavior ID to be processed (step S201). Specifically, the suggestion index generation unit 114 refers to the search behavior information 122 and identifies one search behavior ID 122a to be processed. As described above, one or more records are associated with one search behavior ID 122a. In other words, it can be said that one or more input operations are associated with one search behavior in the search behavior information 122.

Next, the suggestion index generation unit 114 calculates a score for each record associated with the search behavior ID according to a predetermined rule (step S202).

As an example, the suggestion index generation unit 114 refers to the success flag 122e of each record associated with the search behavior ID 122a identified in step S201, and treats the search behavior as successful if one or more pieces of information indicating success are included. In that case, the suggestion index generation unit 114 assigns a higher score to the search query 122c of each record that constitutes the successful search behavior than other search queries.

In this case, the suggestion index generation unit 114 can assign a score to each search query 122c by equally dividing a predetermined score to be assigned to a successful search behavior by the number of records associated with the search behavior. For example, a score of "100" is assigned to a successful search behavior, and "search behavior A" includes "record a", "record b", "record c", and "record d", with "record a", "record b", and "record c" having failure flags and "record d" having a success flag. In this case, a score of "25" can be assigned to each of the search queries for "record a", "record b", "record c", and "record d".

The suggestion index generation unit 114 may assign the highest score to the search query of the record that includes the success flag 122e among the records that constitute the successful search behavior, and assign relatively low scores to the other search queries. In the above example, the search query for "record d" may be assigned a score of "40," and the search queries for "record a," "record b," and "record c" may be assigned a score of "20."

For example, the suggestion index generation unit 114 may treat search queries contained in records that have a failure success flag 122e among records that constitute a successful search behavior as gradually approaching success, and assign a higher score to the later chronological records. In the above example, if the search dates and times are newer in the order of "record a", "record b", "record c", and "record d", the score of "record d" with the success flag can be set to "50", and the flags of "record c", "record b", and "record a" with the failure flag can be set to "30", "15", and "5", respectively. In other words, if there are multiple input operations related to the same search behavior, the score can be calculated according to the other input operations.

The suggestion index generating unit 114 can also assign different scores to multiple successful search actions. For example, the suggestion index generating unit 114 can assign a higher score to a search action having a smaller number of reference documents 122d among multiple search action IDs 122a having information indicating success in the success flag 122e, as the target search result has been obtained more efficiently. That is, a case will be described in which there are "search action A" and "search action B" each having a success flag, and the number of reference documents 122d associated with "search action A" is smaller than the number of reference documents 122d associated with "search action B". In that case, a score of "100" can be assigned to "search action A" and a score of "70" can be assigned to "search action B". After that, the scores assigned to the search actions are distributed to the search queries that constitute each search action.

For example, the suggestion index generation unit 114 can assign a higher score to a search action ID 122a having fewer different search queries 122c, i.e., fewer times the search query has been changed, among the multiple search action IDs 122a having information indicating success in the success flag 122e. For example, the suggestion index generation unit 114 can refer to the search history information 121 for each search action to determine the required time, and assign a higher score to a search action that requires a short required time.

The suggestion index generating unit 114 may assign a lower score to search behaviors that do not have a success flag than to search behaviors that have a success flag, and may not assign a score to search behaviors that do not have a success flag. In addition, the score assignment method of the suggestion index generating unit 114 is not limited to this example. It is sufficient that a high score is assigned to search behaviors and search queries that are worth referring to.

Next, the suggestion index generation unit 114 generates suggestion index information 123 that associates the search query, score, and user ID included in the record (step S203). Specifically, the suggestion index generation unit 114 generates a record of suggestion index information 123 that associates the search query 122c included in the record of the search behavior information 122 for which the score was calculated in step S202, the user ID 122b, and the calculated score as search query 123a, user ID 123b, and score 123c, respectively. After that, the suggestion index generation unit 114 ends the processing of this flowchart.

As described above, according to this embodiment, a search behavior that leads to a successful search result can be determined to be successful, and an appropriate score can be assigned to the input operation that constitutes the search behavior. Furthermore, if the process of generating suggestion index information 123 shown in FIG. 8 is performed following the search behavior information generation process shown in FIG. 7, the suggestion index information 123 is generated at the time when the successful search behavior ends. This makes it possible to more quickly utilize the search behavior of the user to be trained in providing suggestions.

Figure 9 is a flowchart showing an example of the suggestion providing process.

First, the suggestion extraction unit 116 acquires a search query (step S301). When the search request unit 311 of the terminal device 300 accepts an input operation of a search query by the user, the search request unit 311 transmits a search request including the search query to the search suggestion management device 100. The suggestion extraction unit 116 acquires the search query included in the search request.

Next, the training target identification unit 115 identifies the training target (step S302). The search request unit 311 of the terminal device 300 accepts an input operation of information identifying the training target by the user and transmits it to the search suggestion management device 100. The training target identification unit 115 identifies the training target transmitted from the terminal device 300. As described above, instead of receiving the training target from the terminal device 300, the training target identification unit 115 may refer to training information (not shown) and identify the training target of the user performing the search.

Next, the suggestion extraction unit 116 refers to the suggestion index information 123 and extracts records (step S303). Specifically, the suggestion extraction unit 116 extracts records of the suggestion index information 123 that include the student identified in step S302 in the user ID 123b. From the extracted records, the suggestion extraction unit 116 further extracts records that include the search query acquired in step S301 in the search query 123a. Note that the suggestion extraction unit 116 may further extract a predetermined number of records with high scores 123c from the extracted records of the suggestion index information 123.

Next, the suggestion extraction unit 116 associates a ranking with the search query candidate and transmits it to the terminal device 300 (step S304). Specifically, the suggestion extraction unit 116 associates a ranking based on the score 123c with the search query 123a included in the record extracted in step S303 and transmits it to the terminal device 300. For example, the suggestion extraction unit 116 sets a ranking for the search query such that the higher the score, the higher the ranking. Note that the suggestion extraction unit 116 may transmit the search query candidate to the terminal device 300 without associating a ranking with it. After that, the processing unit 110 ends the processing of this flowchart.

In addition, the output unit 330 of the terminal device 300 that has received the search query candidates displays a search screen including the search query candidates. For example, the output unit 330 displays the search query candidates according to the ranking. As an example, the higher the ranking, the more the output unit 330 displays the search query candidate in a different position in the search window 351.

Figure 11 is a diagram showing an example of a search screen 350. Figure 11 (A) is an example of a search screen 350 in this embodiment, Figure 11 (B) is an example (part 1) of a search screen 350 in the first modified example, and Figure 11 (C) is an example (part 2) of a search screen 350 in the first modified example.

The search screen 350 shown in FIG. 11(A) includes a search window 351, a search button 352, and a search query candidate display area 353. The search window 351 is an area that accepts input of a search query by the user. The search button 352 is an operation button that accepts input of a search instruction for the search query entered in the search window 351. The search query candidate display area 353 is an area that displays search query candidates received by the terminal device 300 from the search suggestion management device 100.

As described above, this embodiment can extract search queries that utilize the search behavior of users to be trained and present them to the user conducting the search, making effective use of the user's search skills. For example, by performing suggestion management according to this embodiment in an organization that shares knowledge information, the search behavior of excellent users can be used as a reference for other users, improving the work efficiency of the entire organization.

<First modified example>

Next, we will explain the search suggestion management system 1 in the first modified example. Below, we will explain the differences from the above-mentioned embodiment.

The search log acquisition unit 111 in this modified example acquires a search log that includes at least one of a search target period or a search target category along with the search query. When the search log includes a search target period or a search target category, the score calculation unit 113 in this modified example calculates a score for the combination of the input search query and at least one of the input search target period or search target category.

In addition, in this modified example, when a search target period or a search target category is associated with a search query candidate extracted using the score, the suggestion extraction unit 116 transmits the search target period or the search target category to the terminal device 300 together with the extracted search candidate.

FIG. 11(B) is an example of a search screen 350 in the first modified example, where a search query has not yet been input. As shown in FIG. 11(B), the search screen 350 in the first modified example has a search window 351, a search period input area 354, and a search category input area 355.

When a user inputs a search query into the search window 351, a search period into the search period input area 354, or a search category into the search category input area 355, the search request unit 311 sends a search request including the input information to the search server device 200. The search unit 211 of the search server device 200 searches the document index information 221 for documents corresponding to the search request and sends the search results to the terminal device 300.

Figure 10 is a diagram showing an example of the data structure of the suggestion index information 123 in the first modified example. In this modified example, the suggestion index information 123 includes a search query 123a, a user ID 123b, a score 123c, a period designation 123d, and a category 123e. The period designation 123d is a search target period input in the search target period input field 354. The category 123e is information indicating a search target category input in the search target category input field 355.

When information is included in at least one of the period designation 123d or the category 123e, the score 123c is a numerical value calculated for the combination of the search query 123a and at least one of the period designation 123d or the category 123e. When transmitting the extracted search query candidate to the terminal device 300 in step S304 of the above-mentioned suggestion providing process, the suggestion extraction unit 116 in this modified example determines whether at least one of the period designation 123d or the category 123e is associated with the search query candidate extracted in the suggestion index information 123. When the period designation 123d or the category 123e is associated, the suggestion extraction unit 116 transmits the search target period or the search target category together with the search query candidate to the terminal device 300.

FIG. 11(C) is an example of a search screen 350 in the first modified example, which is an example of the search screen 350 when the terminal device 300 receives search query candidates from the search suggestion management device 100. The search query candidate display area 353 in the first modified example has a search target period display area 3531 and a search target category display area 3532.

The search target period display area 3531 is an area that displays the search target period included in the period specification 123d when the period specification 123d is associated with the search query 123a displayed in the search query candidate display area 353 in the suggest index information 123. The search target category display area 3532 is an area that displays the search target category included in the category 123e when the category 123e is associated with the search query 123a displayed in the search query candidate display area 353 in the suggest index information 123.

As shown in FIG. 11C, the search target period or the search target category is associated with the search query candidates displayed in the search query candidate display area 353. Note that the information displayed in the search target period display area 3531 or the search target category display area 3532 may be displayed in a color different from the search query candidates.

In addition, when a user selects a search query candidate displayed in the search query candidate display area 353, if a search target period or search target category is associated with the selected search query candidate, the associated search target period is input into the search target period display area 3531, and the associated search target category is input into the search target category input area 355.

With this modified example, if the user of the student enters a search period or search category and the search is successful, the user can recognize the search period or search category entered by the student. In addition, when the user selects a search query candidate, the search period or search category is entered, improving efficiency when performing a search with reference to the student.

The search unit 211 of the search server device 200 narrows down the document index using the information input in the search target period input area 354 and the search target category input area 355, and performs a search using the character string of the search query. The search suggestion management device in this modified example only needs to present to the terminal device 300 a search query for searching using a character string and information for narrowing down the document index as suggestions, and the presented information is not limited to the search target period and search target category.

<Second modified example>

Next, we will explain the search suggestion management system 1 in the second modified example. Below, we will explain the differences from the above-mentioned embodiment.

Figure 12 is an example of a functional block of the search suggestion management system 1 in the second modified example. The search suggestion management device 100 in the second modified example includes a search log acquisition unit 111, a search success/failure determination unit 112, a score calculation unit 113, a suggestion index generation unit 114, a training target identification unit 115, a suggestion extraction unit 116, and a deletion information identification unit 117.

The deleted information identifying unit 117 identifies the deleted information that has been deleted from the information to be searched. Specifically, the deleted information identifying unit 117 acquires, from the search unit 211 of the search server device 200, the deleted information that identifies the document that has been deleted from the document index information 221 of the search server device 200.

FIG. 13 is a flowchart showing an example of a suggestion index generation process in the second modified example. The suggestion index generation process in this diagram is executed, for example, periodically in the search suggestion management device 100.

First, the deletion information identification unit 117 identifies a deleted document (step S2001). Specifically, the deletion information identification unit 117 receives deletion information identifying a deleted document from the search unit 211 of the search server device 200. The process performed in step S201 is similar to step S201 of the suggest index generation process in FIG. 8.

Next, the suggestion index generating unit 114 determines whether a record including the identified document and a success flag exists in the search behavior information 122. Specifically, the suggestion index generating unit 114 refers to the search behavior information 122 and determines whether there is a record that includes the document identified in step S2001 in the reference document 122d and whose success flag 122e is information indicating success.

If the suggestion index generation unit 114 determines that a record including the identified document and a success flag exists in the search behavior information 122 (YES in step S2011), the suggestion index generation unit 114 deletes the corresponding record (step S2012). Specifically, the suggestion index generation unit 114 deletes the record determined to exist in step S2011 from the search behavior information 122. As a result, generation of suggestion index information 123 is suppressed for the search query candidate for which the deletion information has been searched.

After step S2012 or in step S2011, if the suggestion index generation unit 114 determines that a record including the identified document and a success flag does not exist in the search behavior information 122 (if "NO" in step S2011), the suggestion index generation unit 114 transitions the process to step S202. The processes in steps S202 and S203 are similar to the suggestion index generation process shown in FIG. 8.

Next, the suggestion index generation unit 114 determines whether there is a search behavior ID for which processing has not yet been completed (step S2031). Specifically, the suggestion index generation unit 114 determines whether there is a search behavior ID in the search behavior information 122 that has not yet been identified as a search behavior ID to be processed in step S201.

If the suggestion index generation unit 114 determines that there is a search behavior ID for which processing has not yet been completed (YES in step S2031), the suggestion index generation unit 114 returns the process to step S201. Then, in step S201, a search behavior ID for which processing has not yet been completed is selected.

If the suggestion index generation unit 114 determines that there is no search behavior ID for which processing has not been completed ("NO" in step S2031), the suggestion index generation unit 114 ends the processing of this flowchart.

As described above, in this modified example, when information in the document index information 221 is deleted, the generation of search behavior information 122 related to search query candidates determined to be successful in searching the document is suppressed. In other words, when the operation log related to the deleted information satisfies the search success condition, the suggestion extraction unit 116 suppresses the extraction of search query candidates that searched the deleted information. That is, when presenting search query candidates to the user, suggestions of search query candidates related to the deleted document are suppressed.

Therefore, when a document is deleted, this can be reflected in the search query candidate suggestions, improving the immediate response to the situation and making it more efficient. In other words, even when a document is deleted, it is possible to effectively utilize search skills and suggest appropriate search query candidates.

Although each embodiment of the present invention has been described above, the present invention is not limited to the above-mentioned embodiment, and various modified examples are included. For example, the above-mentioned embodiment has been described in detail to make the present invention easier to understand, and the present invention is not limited to having all the configurations described here. In addition, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment. It is also possible to add the configuration of another embodiment to the configuration of one embodiment. It is also possible to add, delete, or replace other configurations with respect to a part of the configuration of each embodiment. In addition, each of the above-mentioned configurations, functions, processing units, processing means, etc. may be realized in hardware, part or all of them, for example, by designing them as integrated circuits. In addition, the control lines and information lines in the figure show those that are considered necessary for explanation, and do not necessarily show all of them. It may be considered that almost all of the configurations are connected to each other.

Furthermore, the functional configurations of the above-mentioned search suggestion management device 100, search server device 200, and terminal device 300 are classified according to the main processing contents for ease of understanding. The method of classification or names of the components does not limit the present invention. As described above, the configurations of the search suggestion management device 100, search server device 200, and terminal device 300 can be further classified into more components according to the processing contents. Also, one component can be classified to perform even more processes.

1: Search suggestion management system, 11: Calculation device, 12: Memory, 13: External storage device, 14: Input IF, 15: Output IF, 16: Communication IF, 100: Search suggestion management device, 110, 210, 310: Processing unit, 111: Search log acquisition unit, 112: Search success/failure determination unit, 113: Score calculation unit, 114: Suggestion index generation unit, 115: Apprentice target identification unit, 116: Suggestion extraction unit, 117: Deleted information identification unit, 120, 220: Storage unit, 121: Search history information, 122: Search behavior information, 123: Suggestion document index information, 130, 230, 320: input section, 140, 240, 330: output section, 150, 250, 340: communication section, 200: search server device, 211: search section, 221: document index information, 300: terminal device, 311: search request section, 350: search screen, 351: search window, 352: search button, 353: search query candidate display area, 354: search target period input area, 355: search target category input area, 3531: search target period display area, 3532: search target category display area, N: network

Claims (10)

A search log acquisition unit that acquires a user's search log;
a score calculation unit that calculates a score for each search query candidate including one or more search queries input by the user based on the search behavior of the user;
a suggestion index generating unit that generates suggestion index information that associates the search query candidates, the users, and the scores;
a study subject identification unit for identifying a study subject under which other users will study in the search;
and a suggestion extraction unit that, when a search query input from the other user is obtained, extracts the search query candidates from the suggestion index information using the user to be trained, the input search query, and the score. The search suggestion management device according to claim 1 ,
a search success/failure determination unit that refers to the search log and determines whether the user's search behavior is successful;
The search suggestion management device, wherein the score calculation unit calculates the score for the search query candidate related to the successful search action to be higher than the scores for other search query candidates. 3. The search suggestion management device according to claim 2,
The search log includes information output as a search result and an operation log for the information,
The search suggestion management device, characterized in that the search success determination unit determines that the search behavior leading to obtaining the search results was successful when the operation log satisfies a predetermined search success condition. 4. The search suggestion management device according to claim 1,
the score calculation unit references the search log and generates search behavior information in which one or more input operations are associated with one of the search behaviors;
A search suggestion management device characterized in that, when calculating the score for the search query candidate, if there is another input operation associated with the same search behavior as the input operation related to the search query candidate, the score is calculated according to the other input operation. 5. The search suggestion management device according to claim 1,
The search suggestion management device, wherein the suggestion extraction unit outputs the extracted search query candidates in association with a ranking based on the score. 6. The search suggestion management device according to claim 1,
The search log acquisition unit acquires the search log including at least one of a search target period or a search target category,
the score calculation unit, when at least one of the search target period or the search target category is input, calculates the score for a combination of the search query candidate and at least one of the input search target period or the input search target category;
The search suggestion management device is characterized in that, when the search target period or the search target category is associated with the search query candidate extracted using the score, the suggestion extraction unit outputs the associated search target period or the search target category together with the search query candidate. The search suggestion management device according to claim 3,
a deletion information identifying unit that identifies information that has been deleted from information to be searched,
The search suggestion management device is characterized in that the suggestion extraction unit suppresses extraction of the search query candidate that searched for the deletion information when the operation log related to the deletion information satisfies the search success condition. A search suggestion management system including a terminal device, a search server device, and a search suggestion management device, which are communicatively connected to each other,
The terminal device
a search request unit that receives a search query input from a user and transmits a search request to the search server device;
an output unit that receives and displays search query suggestions from the search suggestion management device;
The search server device
a search unit that searches for information to be searched using the search query related to the search request;
The search suggestion management device,
A search log acquisition unit that acquires a search log of the user;
a score calculation unit that calculates a score for each search query candidate including one or more search queries input by the user based on the search behavior of the user;
a suggestion index generating unit that generates suggestion index information that associates the search query candidates, the users, and the scores;
a study subject identification unit that identifies a study subject under which other users will study in the search;
a suggestion extraction unit that, when a search query input from the other user is acquired, extracts the search query candidate from the suggestion index information by using the user to be trained, the input search query, and the score;
The search suggestion management system according to claim 1, wherein an output unit of the terminal device displays the search query candidates extracted by the suggestion extraction unit of the search suggestion management system. A search suggestion management method by a search suggestion management device, comprising:
A procedure for acquiring a user's search log;
a score calculation step of calculating a score for each search query candidate including one or more search queries input by the user based on the search behavior of the user;
a suggestion index generating step of generating suggestion index information that associates the search query candidates, the user, and the score;
a study subject identification unit for identifying a study subject under which other users will study in the search;
a suggestion extraction step of, when a search query input from the other user is obtained, extracting search query candidates from the suggestion index information using the user to be trained, the input search query, and the score. A program for causing a processing unit of a computer to execute a search suggestion management method,
A procedure for acquiring a user's search log;
a score calculation step of calculating a score for each search query candidate including one or more search queries input by the user based on the search behavior of the user;
a suggestion index generating step of generating suggestion index information that associates the search query candidates, the user, and the score;
a study subject identification unit that identifies a study subject under which other users will study in the search;
When a search query input from the other user is obtained, a suggestion extraction procedure is executed to extract search query candidates from the suggestion index information using the user to be trained, the input search query, and the score.

Images