Information searching method and information searching system

Abstract

To provide an information searching method and an information searching system capable of efficiently searching desired information among an enormous amount of information.SOLUTION: An information searching method is executed by an information searching system in which an information managing device executes a searching process based on a search condition containing a keyword input by a searcher, and one or greater than or equal to two search keywords extracted by a correlation analysis on the keyword, and searches information matching the search condition from a database. The information managing device receives a selection operation for causing the searcher to select, in the search condition, whether the search keyword is set to be an AND condition to the keyword or not, and whether the search keyword is set to be a NOT condition to the keyword, and then executes the searching process.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present invention relates to an information retrieval method and an information retrieval system that can efficiently retrieve information desired by a searcher from a database that stores a plurality of pieces of information. [Background technology]

[0002] In order to search for desired information from a vast amount of information stored in a database, it is necessary to input appropriate keywords into an information retrieval system. However, for those with little knowledge or experience, it can be difficult to input appropriate keywords, which can lead to a problem of being unable to search for desired information. In light of this, for example, Patent Document 1 proposes an information retrieval system that performs a search using not only the input keyword but also its synonyms, thereby making it possible to search for desired information even if the input keyword is inappropriate. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2019-121392 Summary of the Invention [Problem to be solved by the invention]

[0004] However, in the information retrieval system proposed in Patent Document 1, since a search is performed including synonyms, the number of search results can be enormous depending on the keywords entered. Therefore, even if the desired information is included in the search results, it takes a lot of time to identify the desired information by checking each of the results one by one from the enormous number of search results.

[0005] The present invention has been made in view of the above, and has as its object to provide an information retrieval method and an information retrieval system that can efficiently retrieve desired information from a vast amount of information. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems and achieve the object, the information retrieval method of the present invention is an information retrieval method in which an information management device executes a search process based on search conditions including a keyword entered by a searcher and one or more narrowing-down keywords extracted by correlation analysis of the keyword, and searches a database for information that matches the search conditions, and the information management device executes the search process after accepting a selection operation that allows the searcher to select whether or not the narrowing-down keywords should be an AND condition for the keyword, or whether or not the narrowing-down keywords should be a NOT condition for the keyword, in the search conditions.

[0007] In addition, in the information search method of the present invention, when multiple narrowing-down keywords including a first narrowing-down keyword and a second narrowing-down keyword having the highest correlation to the keyword in that order are extracted as the narrowing-down keyword, the information management device accepts the selection operation in the order of the first narrowing-down keyword having the relatively highest correlation to the keyword, and then the second narrowing-down keyword.

[0008] In addition, in the information search method of the present invention, when the searcher selects whether or not to use the first narrowing-down keyword as an AND condition, or whether or not to use the first narrowing-down keyword as a NOT condition, the information management device executes the search process, and if the number of data items matching the search condition falls below a predetermined threshold, does not accept the selection operation for the second narrowing-down keyword.

[0009] In addition, in the information search method according to the present invention, in the above invention, the correlation analysis is performed based on category information arbitrarily set from a plurality of categories indicating the relationship between the keyword and the narrowing-down keywords.

[0010] In order to solve the above-mentioned problems and achieve the object, the information retrieval system of the present invention is an information retrieval system comprising a database and an information management device that executes a search process based on search criteria including a keyword entered by a searcher and one or more narrowing-down keywords extracted by correlation analysis of the keyword, and searches the database for information that matches the search criteria, and the information management device executes the search process after accepting a selection operation that allows the searcher to select whether or not the narrowing-down keywords should be an AND condition for the keyword, or whether or not the narrowing-down keywords should be a NOT condition for the keyword, in the search criteria. [Effects of the Invention]

[0011] According to the present invention, desired information can be efficiently searched for from a vast amount of information. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a block diagram showing the configuration of an information retrieval system according to an embodiment. [Figure 2] FIG. 2 is a diagram showing an example of the criteria for folder names and file names. [Figure 3A] FIG. 3A is a diagram illustrating an example of definition information stored in a section definition information storage unit. [Figure 3B] FIG. 3B is a diagram showing an example of a creation system (definition) of words belonging to a category. [Figure 4A] FIG. 4A is a diagram showing an example of information type determination keywords stored in the information type determination dictionary. [Figure 4B] FIG. 4B is a diagram showing an example of keywords stored in the keyword dictionary. [Figure 5] FIG. 5 is a flowchart showing the flow of information search processing according to one embodiment of the present invention. [Figure 6A] FIG. 6A is a diagram showing an example of a first screen displayed on a display device in the information search method according to the embodiment. [Figure 6B] FIG. 6B is a diagram showing an example of a second screen displayed on the display device in the information search method according to the embodiment. [Figure 6C] FIG. 6C is a diagram showing an example of a third screen displayed on the display device in the information search method according to the embodiment. [Figure 6D] FIG. 6D is a diagram showing an example of the degree of correlation between the search keyword and each word (narrowing keywords) belonging to the category selected by the searcher in the information search method according to the embodiment. [Figure 6E] FIG. 6E is a diagram showing an example of a fourth screen displayed on the display device in the information search method according to the embodiment. [Figure 6F] FIG. 6F is a diagram showing an example of a fifth screen displayed on the display device in the information search method according to the embodiment. [Figure 6G] FIG. 6G is a diagram showing an example of a sixth screen displayed on the display device in the information search method according to the embodiment. [Figure 6H] FIG. 6H is a diagram showing an example of a seventh screen displayed on the display device in the information search method according to the embodiment. [Figure 6I] FIG. 6I is a diagram showing an example of an eighth screen displayed on the display device in the information retrieval method according to the embodiment. [Figure 7A] FIG. 7A is a diagram showing an example of a ninth screen displayed on the display device in the information search method according to the embodiment. [Figure 7B] FIG. 7B is a diagram showing an example of a tenth screen displayed on the display device in the information search method according to the embodiment. [Figure 7C] FIG. 7C is a diagram showing an example of an eleventh screen displayed on the display device in the information retrieval method according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0013] An information retrieval system and an information retrieval method according to an embodiment of the present invention will be described below. In the following, facility maintenance work information in the manufacturing industry will be described as an example of information to be searched. In the following description, when simply referring to "information" or "data," it refers to "facility maintenance work information."

[0014] [Information Retrieval System] The configuration of an information retrieval system according to an embodiment will be described with reference to Figures 1 to 4B. As shown in Figure 1, the information retrieval system according to an embodiment includes an information management device 10 that manages information, a database 12 that stores a plurality of pieces of facility maintenance work information to be searched, a plurality of file servers 14, and a searcher terminal 20.

[0015] The information management device 10, database 12, multiple file servers 14, and searcher terminal 20 are all connected via a network 18. The network 18 is a communications network, such as a LAN (Local Area Network), managed by the organization (company) to which the user belongs. However, the network 18 may be a public communications network such as the Internet, or a communications network that partially incorporates a general public line, such as a WAN (Wide Area Network) or VAN (Virtual Private Network).

[0016] The information management device 10 and the searcher terminal 20 are each realized by a computer and a program. Examples of the searcher terminal 20 include a mobile terminal such as a smartphone, a personal computer, and a tablet computer. An input device 22 such as a keyboard or a mouse, and a display device 24 such as a display are connected to or integrated with the searcher terminal 20.

[0017] The database 12 consolidates and stores documents such as reports and manuals, drawings, photographs, etc. stored in each file server 14. The information stored in the database 12 includes information on equipment failures and malfunctions, their causes, and work standards and manuals on how to deal with them. In this embodiment, it is assumed that a specific piece of equipment exists in multiple regions (e.g., multiple countries or multiple prefectures), and the file servers 14 are installed for each region.

[0018] The information management device 10 executes a search process based on search conditions including a search keyword entered by a searcher and one or more narrowing-down keywords extracted by correlation analysis of the search keyword, and searches the database 12 for information that matches the search conditions.

[0019] The information management device 10 has an information registration unit 110, an information search unit 120, and various storage units (keyword dictionary 50, category definition information storage unit 52, search information storage unit 54, and information type determination dictionary 56). In this embodiment, the configurations that perform the registration function and the search function are separated into the information registration unit 110 and the information search unit 120, but the two may be a single configuration.

[0020] The information registration unit 110 is configured to be able to perform the registration function described above, more specifically, to execute the process of storing information stored in each file server 14 in the database 12. The information registration unit 110 also includes an information linking unit 112, an information processing unit 114, and a storage processing unit 116.

[0021] The information linking unit 112 automatically links, periodically or irregularly, the information stored in the database 12 with the information stored in each file server 14. For example, when new data that is not stored in the database 12 is stored in the file server 14, the information linking unit 112 acquires the new data via the network 18.

[0022] The information processing unit 114 performs predetermined data processing on the data acquired by the information linking unit 112. As part of this data processing, the information processing unit 114 adds information type information to the data. The information type information indicates to which information type the document belongs.

[0023] The information processing unit 114 identifies the information type of the data, for example, based on the name of the folder in which the data was stored, the file name of the data, etc. Fig. 2 shows an example of the criteria for defining the folder names prepared in each file server 14 and the file names of the data stored in each folder.

[0024] When specifying the information type by folder, the information processing unit 114, for example, with reference to FIG. 2, determines the information type of data stored in the "Failure Report" folder to be "Failure Case." The information processing unit 114 also determines the information type of data stored in the "Operation Standards," "Creation Procedures," and "Safety Materials" folders to be "Manuals," and the information type of data stored in the "Maintenance Know-How" folder to be "Maintenance Know-How." The information processing unit 114 also determines the information type of data stored in the "Drawings" folder to be "Drawings," the information type of data stored in the "Maintenance Register" folder to be "Maintenance Register," and the information type of data stored in the "Operation Standards" folder to be "Operation Standards."

[0025] When the information processing unit 114 identifies the type of information by the file name, the file name is created by the creator of the data based on the file creation criteria of Figure 2, for example, and the data is identified as belonging to a "fault case" based on this file name.

[0026] As a further data processing step, the information processing unit 114 assigns category information to the data based on the category information stored in the category definition information storage unit 52. Fig. 3A shows an example of category information stored in the category definition information storage unit 52. This category information defines to which category out of multiple categories a specific word belongs based on the criteria shown in Fig. 3B.

[0027] In Figure 3A, for example, it is shown that the word "abc" belongs to the "state" category and that "D51" is assigned as category information. Therefore, for example, if the word "abc" is included in acquired data, the category information "D51" assigned to the word "abc" is assigned to the data. Returning to Figure 1, we will continue to explain the remaining configuration.

[0028] The storage processing unit 116 stores the data processed by the information processing unit 114 in the database 12. At this time, the storage processing unit 16 stores the data to be stored in the database 12 by classifying the data by information type (see FIG. 1).

[0029] The information stored in database 12 is divided in advance into a plurality of groups A, B, C, etc., according to the type of information. For example, in database 12 shown in Fig. 1, group A of the first information type stores information related to "fault cases," group B of the second information type stores information related to "manuals," and group C of the third information type stores information related to "drawings." Furthermore, although not shown in the figure, group D of the fourth information type, group E of the fifth information type, and group F of the sixth information type store information related to "maintenance know-how," "maintenance ledger," and "operating standards," respectively.

[0030] The information search unit 120 searches for and extracts specific equipment maintenance information from the database 12 based on a search instruction from the searcher terminal 20. The information search unit 120 includes a question receiving unit 121, a language processing unit 122, a search candidate keyword reading unit 123, an information type determination unit 124, a classification information receiving unit 125, a correlation calculation unit 126, a narrowing-down keyword presentation unit 127, and a search execution unit 130.

[0031] The question receiving unit 121 receives a question input by a searcher via the input device 22 of the searcher terminal 20 via the network 18. The input format of the question is not particularly limited, and for example, one or more words may be input, or a sentence in natural language format may be input. For example, a sentence such as "What should be done if the BB device in the AA facility breaks down?" or a question consisting of multiple words separated by spaces, such as "AA device BB device," is input from the searcher terminal 20.

[0032] The language processing unit 122 extracts words (independent words) by performing language processing such as known morphological analysis on the question received by the question receiving unit 121.

[0033] Specifically, the language processing unit 122 extracts, from the input question, input search keywords to be used in a later search and information type determination keywords to be used to determine the information type, which is the search intent (purpose). For example, as described above, when the question sentence "What should be done if the BB device in the AA facility breaks down?" is input, the language processing unit 122 extracts "AA facility" and "BB device" as input search keywords.

[0034] Furthermore, the language processing unit 122 extracts input search keywords by distinguishing between affirmative and negative forms, taking into consideration the context of the question received by the question receiving unit 121. For example, if the question contains the word "burned," the language processing unit 122 distinguishes between "it is burned" and "it is not burned," and extracts this as the input search keyword.

[0035] Furthermore, the language processing unit 122 refers to the information type determination dictionary 56 (see FIG. 4A) and extracts, as information type determination keywords, words that match words registered in the information type determination dictionary 56. For example, as described above, when the question "What should be done if the BB device in the AA facility breaks down?" is input, the language processing unit 122 extracts the terms "breakdown" and "action" as information type determination keywords.

[0036] Next, before describing the search candidate keyword reading unit 123, a description will be given of the keyword dictionary 50. In this keyword dictionary 50, related keywords are registered and stored in advance in association with each other.

[0037] For example, synonymous keywords are associated with each other and registered in the keyword dictionary 50. Also, similar keywords are associated with each other and registered in the keyword dictionary 50. However, the keyword dictionary 50 may register only synonymous keywords or only similar keywords.

[0038] 4B shows an example of keywords pre-registered in the keyword dictionary 50. In the figure, keywords belonging to the same row (horizontal column) are related to each other and are associated with each other.

[0039] The search candidate keyword reading unit 123 refers to the keyword dictionary 50 each time the language processing unit 122 extracts an input search keyword, and reads out keywords related to the input search keyword as search candidate keywords.

[0040] For example, as described above, when the input search keyword is "AA equipment," the search candidate keyword readout unit 123 searches for a match in the keyword dictionary 50 shown in Fig. 4B. Then, all other keywords belonging to the row containing "AA equipment" (the first row in the figure) or, if specified (for example, only synonyms), the specified keyword are read out as search candidate keywords.

[0041] In this example, "XX equipment," "A'A' equipment," and "A"A"" are read out as synonyms of "AA equipment," and "aaa" and "a'a'a'" are read out as synonyms of "AA equipment." Similarly, for "BB equipment," related keywords are read out as search candidate keywords by referring to the second line in FIG. 4B.

[0042] Next, before the information type determination unit 124, the search information storage unit 54 will be described. The search information storage unit 54 stores, as search keywords, the input search keywords and information type determination keywords extracted by the language processing unit 122, and the search candidate keywords read out by the search candidate keyword reading unit 123. For example, each time a question is additionally input and an additional search keyword is extracted (read out), the search information storage unit 54 cumulatively stores the search keyword.

[0043] The information type determination unit 124 determines the information type that indicates the searcher's search intention based on the information type determination keywords extracted from the question. Specifically, the information type determination unit 124 refers to the information type determination dictionary 56 shown in FIG. 4A and determines the information type to be searched for based on the information type determination keywords.

[0044] The information type determination dictionary 56 has information type determination keywords expected to be extracted from questions and their corresponding information types registered in advance in a corresponding state. In the information type determination dictionary 56 of FIG. 4A, the information type determination keywords "failure," "similar examples," and "measures" are registered in advance in a corresponding state with the information type "failure case." Also, in the information type determination dictionary 56 of the same figure, the information type determination keywords "failure type" and "replacement procedure" are registered in advance in a corresponding state with the information type "manual." Although not shown in the same figure, information type determination keywords are also registered for the information types "maintenance know-how," "drawing," "maintenance ledger," and "operating standard" shown in FIG. 2.

[0045] The category information receiving unit 125 displays the category to which the narrowing-down keywords belong in a selectable manner. For example, in the above example, information belonging to Group A "Fault Cases" is searched using the search keywords "AA Equipment" and "BB Device" (including synonyms and similar words), but the search results may be enormous. Therefore, the category information receiving unit 125 (and the correlation calculation unit 126 and narrowing-down keyword display unit described later) supports the searcher in adding or excluding narrowing-down keywords in order to narrow down the search results.

[0046] The category information receiving unit 125 displays a plurality of selectable categories. Specifically, the category information receiving unit 125 displays, for example, a pull-down menu so that one of a plurality of categories including "line," "facility," "device," "part," "status," and "cause" can be selected (see FIG. 6C, which will be described later).

[0047] The correlation calculation unit 126 calculates the degree of correlation between the search keyword and each word belonging to the category selected by the searcher in the category information receiving unit 125. The correlation calculation unit 126 can calculate the degree of correlation based on the frequency with which the search keyword and each word belonging to the category are mentioned together in, for example, the same report, manual, etc. The correlation calculation unit 126 can also calculate the degree of correlation based on the number of times the search keyword and each word belonging to the category are combined as search keywords stored in the search information storage unit 54.

[0048] If there are multiple search keywords, the correlation degree is calculated for some or all of the search keywords selected by the searcher. The above "search keywords" specifically refer to the input search keywords extracted by the language processing unit 122 and the search candidate keywords read out by the search candidate keyword reading unit 123.

[0049] The narrowing-down keyword presentation unit 127 presents to the searcher, in a selectable manner, whether or not to add each word (hereinafter referred to as a "narrowing-down keyword") belonging to the category selected by the searcher in the category information receiving unit 125 to the search conditions. That is, the narrowing-down keyword presentation unit 127 accepts a selection operation that allows the searcher to select whether or not to use the narrowing-down keyword as an AND condition with respect to the search keyword, or whether or not to use the narrowing-down keyword as a NOT condition with respect to the search keyword, in the search conditions of the database 12 (see FIGS. 6E to 6G described later). Specifically, this selection operation includes three patterns: (1) using the narrowing-down keyword as an AND condition with respect to the search keyword, (2) using the narrowing-down keyword as a NOT condition, and (3) not using the narrowing-down keyword as either an AND condition or a NOT condition (skip).

[0050] The narrowing-down keywords are words that can be either affirmative or negative. For example, narrowing-down keywords for "burned" are either "burned" or "not burned."

[0051] Here, through the processing in the correlation calculation unit 126, a plurality of narrowing-down keywords including a first narrowing-down keyword and a second narrowing-down keyword having the highest correlation with the search keyword in this order may be extracted as the narrowing-down keywords (see FIG. 6D described later). For example, in the example of FIG. 6D, the first narrowing-down keyword is "abc" and the second narrowing-down keyword is "def".

[0052] In this case, the narrowing-down keyword presentation unit 127 accepts a selection operation regarding whether to use the first narrowing-down keyword and the second narrowing-down keyword, which have a relatively high degree of correlation with the search keyword, as an AND condition for the search keyword, or whether to use the second narrowing-down keyword as a NOT condition for the search keyword, in that order (see FIGS. 6E and 6F, which will be described later). Specifically, this selection operation includes three patterns: (1) use the first and second narrowing-down keywords as AND conditions for the search keyword, (2) use them as NOT conditions, and (3) do not use them as either AND conditions or NOT conditions (skip).

[0053] Furthermore, when the searcher selects whether to use the first narrowing-down keyword as an AND condition or a NOT condition, the narrowing-down keyword presentation unit 127 executes a provisional search process (provisional search process). Then, when the number of data items matching the search condition falls below a predetermined threshold, the narrowing-down keyword presentation unit 127 does not accept a selection operation for the second narrowing-down keyword. In other words, when the search results are sufficiently small as a result of performing a provisional search by adding or excluding the first narrowing-down keyword, the narrowing-down keyword presentation unit 127 does not display a screen (FIG. 6F described below) for selecting whether to add the second narrowing-down keyword to the search condition.

[0054] The search execution unit 130 performs a text-based match search process between the search keywords stored in the search information storage unit 54 and the narrowing-down keywords selected by the searcher for the information stored in the database 12. The search execution unit 130 also performs the match search process on information in a group corresponding to the information type determined by the information type determination unit 124, out of multiple groups A, B, C, etc. in the database 12.

[0055] [Information search method] An information search method executed by the information search system according to the embodiment will be described with reference to Figs. 5 to 7C. Fig. 5 is a flowchart showing the flow of the information search method according to the embodiment. Figs. 6A to 7C show screens displayed on the display device 24 of the searcher terminal 20, which are examples of application screens for using the information search system according to the embodiment. The information search method according to the embodiment is started, for example, when a searcher starts the application on the searcher terminal 20.

[0056] First, question receiving unit 121 determines whether a question has been received (step S101). For example, when a searcher starts an application, a screen such as that shown in Fig. 6A is displayed on display device 24. Then, when the searcher inputs a question into search space 200 via input device 22 and selects (e.g., clicks) send icon 201, question receiving unit 121 receives the question.

[0057] In this example, a case will be described in which the question receiving unit 121 receives a question such as, "What should be done if the BB device in the AA facility breaks down?". At the bottom of the screen, a history display area 300 is provided, which is configured so that all or part of questions previously entered in the search space 200 can be duplicated.

[0058] In step S101, if question receiving unit 121 receives a question (Yes in step S101), the process proceeds to step S102, and if no question is received (No in step S101), the process returns to step S101.

[0059] Next, the language processing unit 122 extracts words (independent words) by performing natural language processing such as morphological analysis on the question sentence received by the question receiving unit 121 (step S102). In this example, the language processing unit 122 extracts "AA facility," "BB device," "failure," and "measure" from the question sentence, for example, "What should be done when the BB device breaks down in the AA facility?"

[0060] Next, the language processing unit 122 extracts (selects) input search keywords and information type determination keywords from the words extracted by the language processing (step S103). In this example, the language processing unit 122 extracts, for example, "AA equipment" and "BB device" as input search keywords, and "fault" and "handling" as information type determination keywords.

[0061] Next, the search candidate keyword reading unit 123 refers to the keyword dictionary 50 (see FIG. 4B) and reads out keywords related to the input search keyword as search candidate keywords (step S104). In this example, the search candidate keyword reading unit 123 reads out, as search candidate keywords, synonyms and similar words to which the input search keyword "AA equipment" extracted in step S103 belongs and synonyms and similar words to which "BB device" belongs.

[0062] Next, the search information storage unit 54 stores the input search keyword and the search candidate keyword read by the search candidate keyword reading unit 123 (step S105). In this example, the search information storage unit 54 stores "AA equipment" and "BB device," as well as their synonyms and similar words.

[0063] Next, the information type determination unit 124 determines the information type (search intent) based on the information type determination keywords extracted in step S103 and the information type determination dictionary 56 (see FIG. 4A) (step S106). In this example, the information type determination unit 124 determines that the information type is "fault case" associated with "fault" and "measure."

[0064] Next, the information search unit 120 displays the search period, information type, search keyword, and category information on the display device 24 via the searcher terminal 20, for example, as shown in FIG. 6B (step S107). Note that the above "search keyword" specifically refers to the input search keyword extracted by the language processing unit 122 and the search candidate keyword read out by the search candidate keyword readout unit 123.

[0065] On the screen shown in Figure 6B, the information type, search keywords, and classification information are displayed as selectable items, and the searcher can change them as needed. The search period can be set to any period, and if no period is specified, the search will be performed for the entire period. Search keywords can also be excluded from the search by unchecking the checkbox in front of each keyword.

[0066] In the search keyword display area, a selectable word request icon 203 is displayed, allowing the user to request the addition or deletion of other keywords similar to the keyword displayed as the search keyword. Furthermore, the category information is initially set to "Not Specified," but by operating the corresponding area using the input device 22, a pull-down menu is displayed, as shown in FIG. 6C, allowing the user to select the desired category. In this example, a case will be described in which the information type and input search keyword remain unchanged, but the category information is changed from "Not Specified" to "Status."

[0067] In step S107, if the classification information is changed from "unspecified" to another classification, and the classification information accepting unit 125 accepts a classification other than "unspecified" (Yes in step S108), the process proceeds to step S109. On the other hand, if the classification information is unspecified (No in step S108), the process proceeds to step S112.

[0068] Next, in step S109, the correlation calculation unit 126 calculates the degree of correlation between the search keyword selected by the searcher in step S107 and each word belonging to the category accepted by the category information acceptance unit 125. In this example, the correlation calculation unit 126 calculates the degree of correlation between both the keywords "AA equipment" and "BB device" and each word (abc, def, ...) belonging to the category "status", as shown in Fig. 6D, for example.

[0069] Next, the narrowing-down keyword presentation unit 127 displays the narrowing-down keywords having the Nth highest correlation with the search keyword (N=1 the first time) on the display device 24 via the searcher terminal 20, together with setting candidates for search conditions (step S110). Note that the above "narrowing-down keywords" refer to, for example, each word (abc, def, ...) belonging to the category "state" as shown in Fig. 6D.

[0070] Next, the narrowing-down keyword presentation unit 127 executes a tentative search based on the search conditions selected by the searcher (step S111). Next, the narrowing-down keyword presentation unit 127 determines whether or not a predetermined condition is met (step S112). If the predetermined condition is met, the process proceeds to step S113. If the predetermined condition is not met, the process returns to step S110, and "N=N+1" is set. The above "predetermined condition" refers to a condition for whether or not narrowing-down keywords are presented to the searcher. Examples of the predetermined condition include when there are no correlated narrowing-down keywords, when the number of search hits when a tentative search is performed is several tens (for example, 50), or less, and so on.

[0071] The processing of steps S110 to S112 will be described in detail below. As shown in FIG. 6E, for example, the narrowing-down keyword presentation unit 127 presents whether or not to set "abc," the narrowing-down keyword with the "first" highest correlation with the search keyword, as an exclusion condition (NOT condition) or an additional condition (AND condition), in a selectable manner. The screen in the same figure also displays the current number of search hits when a tentative search is performed using only the search keyword, icons such as "Set Exclusion Keyword," "Set Additional Keyword," and "Skip." The "Skip" icon is a button to be selected when neither excluding nor adding a narrowing-down keyword. In this example, a case where the narrowing-down keyword "abc" is set as an additional keyword will be described.

[0072] Next, the narrowing-down keyword presentation unit 127 executes a tentative search based on the search keyword and the additional keyword "abc." Then, as shown in FIG. 6F, for example, the narrowing-down keyword presentation unit 127 presents "def," which is the narrowing-down keyword having the second highest correlation with the search keyword, in a selectable manner as to whether it is to be set as an exclusion condition (NOT condition) or as an additional condition (AND condition). In this example, a case where the narrowing-down keyword "def" is set as an exclusion keyword will be described.

[0073] Next, the narrowing-down keyword presentation unit 127 executes a tentative search based on the search keyword, the additional keyword "abc", and the exclusion keyword "def". Then, as shown in FIG. 6G, for example, the narrowing-down keyword presentation unit 127 presents "ghi", which is the narrowing-down keyword having the third highest correlation with the search keyword, in a selectable manner as to whether it is to be set as an exclusion condition (NOT condition) or as an additional condition (AND condition). In this example, a case will be described in which the narrowing-down keyword "ghi" is set as an exclusion keyword. The narrowing-down keyword presentation unit 127 repeatedly presents narrowing-down keywords as shown in FIGS. 6E to 6G until the above-mentioned predetermined condition is met.

[0074] In this way, in the information search system of the embodiment, the searcher is prompted to add or remove narrowing-down keywords until the number of search results falls below a predetermined number, thereby allowing the searcher to easily and visually narrow down the information stored in database 12.

[0075] Next, the narrowing-down keyword presentation unit 127 displays the information type, search keywords, classification information, and narrowing-down keywords (additional keywords, exclusionary keywords) on the display device 24 via the searcher terminal 20, as shown in Fig. 6H (step S113). Note that even at this stage, additional keywords and exclusionary keywords can be excluded from the search by unchecking the checkbox in front of them.

[0076] Next, when search icon 202 (see FIG. 6H) is selected on the screen displayed on display device 24, search execution unit 130 executes a search process on the information stored in database 12 (step S114). Specifically, search execution unit 130 executes a text-based match search process on information belonging to a group corresponding to the selected information type in database 12, based on the input search keyword, search candidate keywords, and narrowing-down keywords.

[0077] Next, the search execution unit 130 displays the search results on the display device 24 on the searcher's side (step S115), as shown in FIG. 6I, for example, and then completes this process.

[0078] 6E to 6G, whether or not to use the narrowing-down keywords as exclusion conditions or whether or not to use the narrowing-down keywords as additional conditions is presented to the searcher in order of the degree of correlation with the search keyword, but the method of presentation is not limited to this. For example, as shown in FIGS. 7A to 7C, multiple narrowing-down keywords, the degree of correlation of the narrowing-down keywords with the search keyword, and check boxes for adding or excluding these narrowing-down keywords may be presented.

[0079] In this case, when a searcher adds or removes a narrowing keyword, the narrowing keyword presentation unit 127 performs a provisional search under that condition and displays on the screen the current number of search hits when the provisional search is performed. For example, in the example of FIG. 7A, when the narrowing keyword "abc" is added, the number of search hits is "xxxxxx." However, as shown in FIG. 7B, by further removing the narrowing keyword "def," the number of search hits decreases to "xxx." Then, as shown in FIG. 7C, by further removing the narrowing keyword "ghi," the number of search hits decreases to "xxx."

[0080] 7A to 7C, the "Add to Keywords" icon is used to register a word selected in the check box as an added keyword or an excluded keyword. For example, as in FIG. 7A, when "abc" is selected in the add check box and the "Add to Keywords" icon is pressed, "abc" is officially registered as an added keyword, and the screen transitions to the screen in FIG. 6H. Also, as in FIG. 7B and 7C, when "def" and "ghi" are selected in the exclude check box and the "Add to Keywords" icon is pressed, "def" and "ghi" are officially registered as excluded keywords, and the screen transitions to the screen in FIG. 6H.

[0081] 7A to 7C. The "Skip" icon on each screen is used to skip the selection of each word. For example, when the "Skip" icon is pressed in a state where "abc" can be added or removed (the "abc" line is highlighted) as shown in FIG. 7A, the addition or removal of "abc" is skipped. Then, the "def" line below is highlighted, and the screen transitions to a state where "def" can be added or removed. For example, when the "Skip" icon is pressed in a state where "def" can be added or removed (the "def" line is highlighted) as shown in FIG. 7B, the addition or removal of "def" is skipped. Then, the "ghi" line below is highlighted, and the screen transitions to a state where "ghi" can be added or removed. For example, when the "Skip" icon is pressed in a state where "ghi" can be added or removed (the "ghi" line is highlighted) as shown in FIG. 7C, the addition or removal of "ghi" is skipped. Then, the screen transitions to the screen shown in FIG. 6H.

[0082] In this way, the information search system of the embodiment presents the number of search results when narrowing-down keywords are added or removed, and suggests to the searcher to add or remove narrowing-down keywords, thereby making it easy and visually easy to narrow down the information stored in database 12.

[0083] In the information search method and information search system according to the above-described embodiment, input search keywords are extracted from a question entered by a searcher, and the searcher is then prompted to select whether to use the narrowing-down keywords extracted by correlation analysis as an AND condition or a NOT condition. This allows for efficient search for desired information from the vast amount of information stored in the database 12. As a result, for example, when a breakdown or malfunction occurs, the searcher can enter a description of the situation and efficiently search the database 12 for useful information such as past cases, and based on the searched information, the searcher can quickly carry out work to restore the equipment from the breakdown.

[0084] Furthermore, in the information retrieval method and information retrieval system according to the embodiment, it is possible to efficiently narrow down the number of search results by excluding highly correlated narrowing-down keywords, so that the desired information can be accurately found even when there are many similar cases in the past.

[0085] The information retrieval method and information retrieval system according to the present invention have been specifically described above using the mode and examples for carrying out the invention, but the gist of the present invention is not limited to these descriptions and should be broadly interpreted based on the claims. It goes without saying that various changes and modifications based on these descriptions are also included in the gist of the present invention.

[0086] For example, in the above embodiment, an example of searching for facility maintenance work information in the manufacturing industry has been described, but the present invention is not limited to this. The information search method and information search system according to the present invention can be applied to Q&A (Question & Answer) information in general, such as information indicating how to deal with symptoms of an illness or the name of the illness, or information indicating the name of an error displayed on machinery such as a car or how to deal with the error depending on the condition.

[0087] In the above embodiment, the degree of correlation with the search keyword is calculated only once (see step S109 in FIG. 5), and a query is sequentially made for the narrowing-down keywords with the first to third highest degrees of correlation (see FIGS. 6E to 6G). However, the degree of correlation may be calculated multiple times. In this case, narrowing-down keyword presentation unit 127 calculates the degree of correlation of multiple narrowing-down keywords with the search keyword (see FIG. 6D), and presents, via display device 24, whether to add or remove the narrowing-down keyword with the highest degree of correlation.

[0088] When the searcher selects whether to add or remove the narrowing-down keyword, the narrowing-down keyword presentation unit 127 recalculates the correlation degree under the selected condition and presents, via the display device 24, whether to add or remove the narrowing-down keyword with the highest correlation degree. The narrowing-down keyword presentation unit 127 then repeats the calculation of the correlation degree and the presentation of the narrowing-down keywords until the above-mentioned predetermined condition is met. In this way, by calculating the correlation degree each time a narrowing-down keyword is presented, it is possible to improve search accuracy. [Explanation of symbols]

[0089] 10 Information management device 12 Databases 14 File Server 18 Network 20 Searcher's terminal 22 Input Devices 24 Display device 50 Keyword Dictionary 52 Classification definition information storage section 54 Search information storage unit 56 Dictionary for determining information type 110 Information Registration Department 112 Information and Communications Department 114 Information Processing Department 116 Storage Processing Unit 120 Information Search Department 121 Question Reception Department 122 Language Processing Unit 123 Search candidate keyword reading section 124 Information type judgment unit 125 Classification Information Reception Department 126 Correlation calculation unit 127 Keyword display section for narrowing down search results 130 Search execution unit

Claims

1. An information retrieval method comprising an information management device that performs a search process based on search conditions including keywords entered by a searcher and one or more filtering keywords extracted by correlation analysis of those keywords, and retrieves information from a database that matches the search conditions, The aforementioned information management device is The search process is executed after accepting a selection operation from the searcher to choose whether or not to use the filtering keyword as an AND condition for the keyword, or whether or not to use the filtering keyword as a NOT condition for the keyword. Prior to executing the aforementioned search process, The following are presented: the filtering keywords, the correlation between the filtering keywords and the original keywords, and the number of hits in the hypothetical search when the filtering keywords are added or removed. Information retrieval methods.

2. The information retrieval method according to Claim 1, wherein the information management device accepts the addition or removal of the filtering keywords in order of decreasing correlation, performs a preliminary search each time, and presents the number of hits in the preliminary search.

3. An information retrieval system comprising: a database; an information management device that performs a search process based on search conditions including keywords entered by a searcher and one or more filtering keywords extracted by correlation analysis of the keywords, and retrieves information from the database that matches the search conditions, The aforementioned information management device is The search process is executed after accepting a selection operation from the searcher to choose whether or not to use the filtering keyword as an AND condition for the keyword, or whether or not to use the filtering keyword as a NOT condition for the keyword. Prior to executing the aforementioned search process, The following are presented: the filtering keywords, the correlation between the filtering keywords and the original keywords, and the number of hits in the hypothetical search when the filtering keywords are added or removed. Information retrieval system.

Images