JP6763837B2 - Search word suggestion device, search word suggestion method, and search word suggestion program - Google Patents

Description

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

The suggestion function that is generally used in the conventional information retrieval system suggests a search word that has been frequently input to the information retrieval system in the past.

Takao Kunioka et al., "Discovery of effective search terms by visualizing JST thesaurus map JST dictionary", Japan Science and Technology Agency, Information Management, 2012 vol.55 no.9

Since the amount of background knowledge about the content that the searcher wants to search here differs depending on the searcher, the content of the entered search word should also change. However, with the method of suggesting a search word that has been entered into the information retrieval system many times in the past as described above, for example, a searcher with a large amount of background knowledge may not be able to suggest the search word that he / she desires. is there. As a result, the search result desired by the searcher may not be obtained. Therefore, an object of the present invention is to solve the above-mentioned problems and to realize a suggestion of a search word according to a large amount of background knowledge possessed by the searcher.

In order to solve the above-mentioned problems, the present invention is a search word suggestion device that suggests a search word used for information retrieval, and establishes a hierarchical relationship between an input unit that receives input of a search word from a user and a word in a thesaurus. With reference to the indicated hierarchical table, the level of background knowledge of the user who input the search word is determined by whether or not the word group constituting the input search word includes a word having the hierarchical relationship. The determination unit, a recording unit that records a search word log in which the determined level of the background knowledge amount of the user and the search word input from the user are associated with each other in the storage unit, and after recording the search word log. When the user accepts the input of the search word, the determination unit determines the level of the background knowledge amount of the user who input the search word, and the search word corresponding to the determined level of the background knowledge amount of the user. It is characterized by including a suggestion unit that suggests a search word read from a log.

According to the present invention, it is possible to realize a suggestion of a search word according to the amount of background knowledge possessed by the searcher.

FIG. 1 is a diagram for explaining an operation example of the search word suggestion device. FIG. 2 is a diagram showing a configuration example of a search word suggestion device. FIG. 3 is a flowchart showing an example of a procedure in which the search word suggestion device records a keyword log. FIG. 4 is a flowchart showing an example of a procedure in which the search word suggestion device suggests a search word. FIG. 5 is a diagram showing an example of a document used by the hierarchical table creation unit to create a hierarchical table. FIG. 6 is a diagram showing a computer that executes a search word suggestion program.

Hereinafter, embodiments (embodiments) for carrying out the present invention will be described with reference to the drawings. The present invention is not limited to this embodiment.

[Overview]
Generally, in an information retrieval system, the search accuracy tends to be improved by searching by adding a lower layer word (a word of a subordinate concept of the word) of the word in the thesaurus to a word input as a search word. It is known to be in (see Non-Patent Document 1). In addition, a searcher who has a lot of background knowledge about the field to be searched tends to input a more specific word as a search word.

Therefore, the search word suggestion device 10 (see FIG. 2) of the present embodiment prepares in advance a hierarchical table (hierarchical information) 121 of words related to the search field that corresponds to a thesaurus, and inputs the search word. Determines the level of background knowledge of the person who entered the search word (searcher) depending on whether or not includes words having a hierarchical relationship in the hierarchical table 121. Then, the search word suggestion device 10 records a keyword log in which the level of knowledge of the determined searcher is associated with the search word input by the searcher. After that, when the search word suggestion device 10 receives the input of the search word from the searcher, the search word suggestion device 10 suggests the search word by using the keyword log corresponding to the knowledge amount of the searcher.

An operation example of the search word suggestion device 10 will be described with reference to FIG. In the hierarchical table 121 here, for example, "PH0.5", "PH1.0", and "PH1.0" are shown as lower layer words (lower concept words) adjacent to the upper layer word (upper concept word) "restart". It is assumed that words such as "PH2.0", ..., etc. are registered.

Here, when the search word suggestion device 10 receives the input of "restart effect" as the search word from the searcher, these words are not adjacent words (having a hierarchical relationship) in the hierarchical table 121, so that the searcher It is judged that the amount of background knowledge of is small. Then, the search word suggestion device 10 records the above search word in the "user keyword log with a small amount of background knowledge".

On the other hand, when the search word suggestion device 10 receives an input of "restart PH1.0 system switching service influence" as a search word from the searcher, since these words are adjacent words in the hierarchical table 121, the background knowledge of the searcher. Judge that the amount is large. Then, the search word suggestion device 10 records the above search word in the “background knowledge amount multi-user keyword log”.

After that, when only "Resume" is input as a search word and returned from the searcher, the search word suggestion device 10 determines that the search is performed by a searcher with a small amount of background knowledge, and "user keyword log with a small amount of background knowledge". Make a suggestion from the inside. For example, the search word suggestion device 10 makes a suggestion such as "restart effect" as a search word.

On the other hand, when "restart PH1.0" is input and returned as a search word from the searcher, the search word suggestion device 10 returns "restart PH1.0" by inputting an adjacent word in the hierarchical table 121. Therefore, it is determined that the search is performed by a searcher with a large amount of background knowledge, and a suggestion is made from the "user keyword log with a large amount of background knowledge". For example, the search word suggestion device 10 makes a suggestion such as "restart PH1.0 system switching service influence" as a search word.

By doing so, the search word suggestion device 10 can suggest the search word according to the amount of background knowledge (level of background knowledge) of the searcher.

[Constitution]
Next, a configuration example of the search word suggestion device 10 will be described with reference to FIG. The search word suggestion device 10 includes an input / output unit (input unit and output unit) 11, a storage unit 12, and a control unit 13. The input / output unit 11 controls the input / output interface of the search word suggestion device 10. The input / output unit 11 receives, for example, input of a search word from a searcher, and outputs a suggestion result of the search word.

The storage unit 12 stores various information for the control unit 13 to suggest a search word. For example, the storage unit 12 stores the hierarchical table 121. Further, the storage unit 12 includes an area for storing a user keyword log (for example, a user keyword log having a large amount of background knowledge, a user keyword log having a small amount of background knowledge, etc.) output from the control unit 13.

As described above, the hierarchical table 121 is information showing the hierarchical relationship between words (the relationship between the superordinate concept and the subordinate concept). For example, the hierarchical relationship between words in the thesaurus and the word between words related to the search field. Information showing the hierarchical relationship. As the hierarchical table 121, those prepared in advance from the information to be searched (for example, document group) may be used. As an example, the hierarchical table 121 is information in which the upper layer word included in the document to be searched is associated with the lower layer word of the upper layer word.

The control unit 13 includes a background knowledge amount determination unit (determination unit) 131, a recording unit 132, and a suggestion processing unit 133. The hierarchical table creation unit 134 indicated by the broken line may or may not be equipped, and the case where it is equipped will be described later.

The background knowledge amount determination unit 131 refers to the search word input via the input / output unit 11 and the hierarchical table 121, and determines the level of the background knowledge amount of the searcher who input the search word. For example, the background knowledge amount determination unit 131 determines the background knowledge of the searcher who input the search word, depending on whether or not the word group constituting the input search word includes a word having a hierarchical relationship in the hierarchical table 121. Determine the level of quantity.

As an example, when the word group constituting the input search word includes a word having a hierarchical relationship in the hierarchical table 121, the background knowledge amount determination unit 131 uses the searcher who input the search word as background knowledge. Judge that the level of quantity is high (searcher with a large amount of background knowledge). On the other hand, when the word group constituting the input search word does not include the word having a hierarchical relationship in the hierarchical table 121, the background knowledge amount determination unit 131 sets the searcher who input the search word as the background knowledge amount. Judged as low level (searcher with little background knowledge).

In addition, the background knowledge amount determination unit 131 not only determines the level of the background knowledge amount of high / low in the determination of the level of the background knowledge amount of the searcher, but also determines the level of three or more stages. You may.

The recording unit 132 stores a keyword log (search word log) in the storage unit 12 in which the level of the background knowledge amount of the searcher determined by the background knowledge amount determination unit 131 and the search word input from the searcher are associated with each other. Record.

For example, when the recording unit 132 receives the input of the search word from the searcher, the recording unit 132 requests the background knowledge amount determination unit 131 to determine the background knowledge amount of the searcher. When the background knowledge amount determination unit 131 determines that the background knowledge amount of the searcher is large, the recording unit 132 records the search word input by the searcher as a "background knowledge amount multi-user keyword log". To do. On the other hand, when the background knowledge amount determination unit 131 determines that the background knowledge amount of the searcher is small, the recording unit 132 records the search word input by the searcher as the "background knowledge amount small user keyword log". ..

The suggestion processing unit 133 suggests the search word to the searcher. Specifically, when the suggestion processing unit 133 receives the input of the search word from the searcher via the input / output unit 11 after the keyword log is recorded by the recording unit 132, the suggestion processing unit 133 inputs the search word to the background knowledge amount determination unit 131. Requests the judgment of the level of background knowledge of the entered searcher. Then, when the suggestion processing unit 133 reads the keyword log corresponding to the level of the background knowledge amount of the searcher determined by the background knowledge amount determination unit 131 from the storage unit 12, the search word recorded in the keyword log is suggested. To do.

By doing so, the search word suggestion device 10 can realize the suggestion of the search word according to the high level of the background knowledge amount (the amount of background knowledge amount) possessed by the searcher.

The suggested search word is displayed, for example, in the lower part of the screen area where the searcher has entered the search word (see FIG. 1). After that, the searcher selects and inputs the search word to be used for the search from the search words displayed on the screen and the suggested search words. Then, the search word suggestion device 10 or the search device (not shown) performs an information search using the search word selected by the searcher.

[Processing procedure]
Next, the processing procedure of the search word suggestion device 10 will be described. First, an example of a procedure in which the search word suggestion device 10 records a keyword log will be described with reference to FIG. 3, and then, using FIG. 4, the search word suggestion device 10 will use the keyword log to describe a search word. An example of the procedure for making a suggestion will be described.

For example, when the input / output unit 11 of the search word suggestion device 10 accepts the input of the search word (S11 in FIG. 3), the background knowledge amount determination unit 131 determines whether or not the search word is an adjacent word in the hierarchical table 121. Judgment (S12). Here, when the background knowledge amount determination unit 131 determines that the search word is an adjacent word in the hierarchical table 121 (Yes in S12), and determines that the searcher has a large amount of background knowledge (S13), the search word. Is recorded in the background knowledge amount multi-user keyword log (S14). On the other hand, when the background knowledge amount determination unit 131 determines that the search word is not an adjacent word in the hierarchical table 121 (No in S12), it determines that the background knowledge amount of the searcher is small (S15), and the search The word is recorded in the user keyword log with a small amount of background knowledge (S16).

The explanation moves to FIG. After that, when the input / output unit 11 of the search word suggestion device 10 receives the input of the search word (S21 in FIG. 4), the background knowledge amount determination unit 131 shows the search word in the hierarchical table 121 as in S12 in FIG. Determines whether or not the words are adjacent to each other (S22). Here, when the background knowledge amount determination unit 131 determines that the search word is an adjacent word in the hierarchical table 121 (Yes in S22), it is determined that the searcher has a large amount of background knowledge (S23), and the suggestion processing unit 133 suggests a search word from the user keyword log having a large amount of background knowledge (S24). On the other hand, when the background knowledge amount determination unit 131 determines that the search word is not an adjacent word in the hierarchical table 121 (No in S22), it determines that the background knowledge amount of the searcher is small (S25), and suggests processing. Section 133 suggests a search word from the user keyword log having a small amount of background knowledge (S26).

By doing so, the search word suggestion device 10 can realize the suggestion of the search word according to the amount of background knowledge possessed by the searcher.

[Other Embodiments]
The search word suggestion device 10 may further include a hierarchical table creation unit (hierarchical information creation unit) 134 (see the broken line portion in FIG. 2).

For example, when the information to be searched by the search word is a document such as a specification, the hierarchical relationship of the search word is often represented in a figure or a table in the above document. For example, as in documents 501 and 502 shown in FIG. 5, lower layer words (“PH0.5”, “PH1.0”, “PH2.0”, etc.) with respect to words representing figures or tables (“restart”, etc.), etc. ) Often appear in the figure or table.

Therefore, the hierarchical table creation unit 134 is included in the explanatory text of the chart such as "*** is shown in Figure XX" or "*** is shown in Table XX" in the document relating to the field to be searched. A word (first word) is extracted from the part corresponding to "***" in the wording indicating the chart. Further, the hierarchical table creation unit 134 extracts words (second words) included in the chart, and counts the number of co-occurrence between the first word and the second word. Then, when the counted number of co-occurrence is equal to or greater than a predetermined value, the hierarchy table creation unit 134 determines that the first word and the second word are in a hierarchical relationship. That is, the hierarchical table creation unit 134 creates an associated hierarchical table 121 with the first word as an upper layer word and the second word as a lower layer word of the first word.

By doing so, the search word suggestion device 10 can automatically create a hierarchical table 121 that matches the field to be searched.

[program]
Further, it can be implemented by installing a program that realizes the function of the search word suggestion device 10 described in the above embodiment on a desired information processing device (computer). For example, the information processing device can function as the search word suggestion device 10 by causing the information processing device to execute the above program provided as package software or online software. The information processing device referred to here includes a desktop type or notebook type personal computer. In addition, the information processing device includes smartphones, mobile phones, mobile communication terminals such as PHS (Personal Handyphone System), and PDA (Personal Digital Assistant). Further, the search word suggestion device 10 may be mounted on the cloud server.

An example of a computer that executes the above program (search word suggestion program) will be described with reference to FIG. As shown in FIG. 6, the computer 1000 has, for example, a memory 1010, a CPU 1020, a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. Each of these parts is connected by a bus 1080.

The memory 1010 includes a ROM (Read Only Memory) 1011 and a RAM (Random Access Memory) 1012. The ROM 1011 stores, for example, a boot program such as a BIOS (Basic Input Output System). The hard disk drive interface 1030 is connected to the hard disk drive 1090. The disk drive interface 1040 is connected to the disk drive 1100. A removable storage medium such as a magnetic disk or an optical disk is inserted into the disk drive 1100. For example, a mouse 1110 and a keyboard 1120 are connected to the serial port interface 1050. A display 1130 is connected to the video adapter 1060, for example.

Here, as shown in FIG. 6, the hard disk drive 1090 stores, for example, the OS 1091, the application program 1092, the program module 1093, and the program data 1094. The various data and information described in the above-described embodiment are stored in, for example, the hard disk drive 1090 or the memory 1010.

Then, the CPU 1020 reads the program module 1093 and the program data 1094 stored in the hard disk drive 1090 into the RAM 1012 as needed, and executes each of the above-described procedures.

The program module 1093 and program data 1094 related to the above search word suggestion program are not limited to the case where they are stored in the hard disk drive 1090, for example, they are stored in a removable storage medium and are stored via a disk drive 1100 or the like. It may be read by the CPU 1020. Alternatively, the program module 1093 and the program data 1094 related to the above program are stored in another computer connected via a network such as LAN (Local Area Network) or WAN (Wide Area Network), and are stored via the network interface 1070. It may be read by the CPU 1020.

10 Search word suggestion device 11 Input / output unit 12 Storage unit 13 Control unit 121 Hierarchical table 131 Background knowledge amount judgment unit 132 Recording unit 133 Suggestion processing unit 134 Hierarchical table creation unit

Claims (5)

A search word suggestion device that suggests search words used for information retrieval.
An input unit that accepts search word input from the user,
Referencing the hierarchical information showing the hierarchical relationship between the superordinate concept and the subordinate concept between words in the thesaurus, at least two words included in the word group constituting the input search word are the hierarchical relationship between the superordinate concept and the subordinate concept. and a by determining lifting mounds not, the search word determining section which level of background knowledge of the user who entered the,
A recording unit that records a search word log in which the determined level of the background knowledge amount of the user and a search word input from the user are associated with each other in the storage unit.
When the input of the search word is received from the user after recording the search word log, the determination unit determines the level of the background knowledge amount of the user who has input the search word, and the background knowledge amount of the determined user is determined. A search word suggestion device including a suggestion unit that suggests a search word read from the search word log corresponding to the level. The determination unit
When at least two words included in the word group constituting the input search word have a hierarchical relationship between the superordinate concept and the subordinate concept with reference to the hierarchical information, the amount of background knowledge of the user who input the search word. The search word suggestion device according to claim 1, wherein the level of is determined to be a high level. The co-occurrence number of the first word included in the description of the figure or table in the predetermined document and the second word included in the figure or table described by the first word is counted, and the co-occurrence is counted. When the number is equal to or greater than a predetermined value, the first word and the second word are extracted as words having a hierarchical relationship, so that a hierarchical information creation unit for creating the hierarchical information is further provided. The search word suggestion device according to claim 1. Search word suggestion device,
An input step that accepts search word input from the user,
Referencing the hierarchical information showing the hierarchical relationship between the superordinate concept and the subordinate concept between words in the thesaurus, at least two words included in the word group constituting the input search word are the hierarchical relationship between the superordinate concept and the subordinate concept. the by either lifting mounds not, the determination step of determining the level of background knowledge of the user who entered the search word,
A recording step of recording a search word log in which the level of the background knowledge amount of the user determined by the determination step and the search word input from the user are associated with each other in the storage unit.
When the input of the search word is received from the user after recording the search word log, the level of the background knowledge amount of the user who input the search word is determined by executing the determination step, and the determined user A search word suggestion method including a suggestion step for suggesting a search word read from the search word log corresponding to the level of background knowledge. The step of accepting the input of the search word from the user,
Referencing the hierarchical information showing the hierarchical relationship between the superordinate concept and the subordinate concept between words in the thesaurus, at least two words included in the word group constituting the input search word are the hierarchical relationship between the superordinate concept and the subordinate concept. the by either lifting mounds not, the determination step of determining the level of background knowledge of the user who entered the search word,
A recording step of recording a search word log in which the level of the background knowledge amount of the user determined by the determination step and the search word input from the user are associated with each other in the storage unit.
When the input of the search word is received from the user after recording the search word log, the level of the background knowledge amount of the user who input the search word is determined by executing the determination step, and the determined user A search word suggestion program characterized by causing a computer to execute a suggestion step that suggests a search word read from the search word log corresponding to the level of background knowledge.

Images