Information retrieval method and information retrieval system

Abstract

An information search system executes an information search method in which an information management device executes a search process on the basis of search conditions including a keyword input by a searcher and one or more screening keywords extracted by correlation analysis of the keyword. The information management device performs a search process after receiving a selection operation that causes a searcher to select, among the search conditions, whether or not to set the keyword for screening as an AND condition for the keyword and whether or not to set the keyword for screening as an NOT condition for the keyword, and retrieves information matching the search conditions from the database.

Description

Technical Field

[0001] The present invention relates to an information retrieval method and an information retrieval system capable of efficiently retrieving information required by a searcher from a database storing a plurality of information. Background Art

[0002] In order to retrieve desired information from the vast amount of information stored in a database, it is necessary to input appropriate keywords into an information retrieval system. However, for people with limited knowledge and experience, it is difficult to input appropriate keywords, which may lead to difficulties in retrieving the desired information. Against this background, for example, Patent Document 1 proposes an information retrieval system that performs a search including not only the input keywords but also their synonyms. This system can thus retrieve the desired information even if the input keywords are inappropriate.

[0003] [Prior art literature]

[0004] [Patent Document]

[0005] Patent Document 1: Japanese Patent Application Laid-Open No. 2019-121392 Summary of the Invention

[0006] [Problems to be Solved by the Invention]

[0007] However, the information retrieval system proposed in Patent Document 1 performs searches including synonyms, resulting in a large number of search results depending on the keyword input. Therefore, even if the search results contain the desired information, it takes a considerable amount of time to identify the desired information one by one from the large number of search results.

[0008] The present invention has been made in view of the above-mentioned circumstances, and an object of the present invention is to provide an information retrieval method and an information retrieval system capable of efficiently retrieving desired information from a vast amount of information.

[0009] [Means for solving the problem]

[0010] In order to solve the above-mentioned problems and achieve the purpose, the information retrieval method of the present invention is as follows: an information management device performs retrieval processing based on retrieval conditions including keywords input by a searcher and one or more filtering keywords extracted through correlation analysis of the keywords, and retrieves information matching the retrieval conditions from a database, wherein the information management device performs the retrieval processing after receiving a selection operation that allows the searcher to select whether to set the filtering keywords as AND conditions for the keywords or whether to set the filtering keywords as NOT conditions for the keywords in the retrieval conditions.

[0011] In addition, the information retrieval method according to the present invention is based on the above invention. When multiple filtering keywords including a first filtering keyword and a second filtering keyword whose relevance to the keyword is ordered from high to low are extracted as the filtering keyword, the information management device accepts the selection operation in the order of the first filtering keyword and the second filtering keyword whose relevance to the keyword is relatively high.

[0012] In addition, the information retrieval method involved in the present invention is based on the above-mentioned invention. When the searcher selects whether to set the first filtering keyword as an AND condition or whether to set the first filtering keyword as a NOT condition, the information management device executes the retrieval processing, and when the number of data entries matching the retrieval condition is lower than a specified threshold, the selection operation for the second filtering keyword is not accepted.

[0013] Furthermore, in the information search method according to the present invention, in the above invention, the correlation analysis is performed based on classification information arbitrarily set from a plurality of classifications indicating the relationship between the keyword and the screening keyword.

[0014] In order to solve the above-mentioned problems and achieve the purpose, the present invention relates to an information retrieval system comprising: a database; and an information management device, which performs retrieval processing based on retrieval conditions including keywords input by a searcher and one or more filtering keywords extracted through correlation analysis of the keywords, and retrieves information matching the retrieval conditions from the database, wherein the information management device performs the retrieval processing after receiving a selection operation that allows the searcher to select whether to set the filtering keywords as AND conditions for the keywords or whether to set the filtering keywords as NOT conditions for the keywords in the retrieval conditions.

[0015] [Effects of the Invention]

[0016] According to the present invention, desired information can be efficiently retrieved from a huge amount of information. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a block diagram showing the configuration of an information retrieval system according to an embodiment.

[0018] Figure 2 This is a diagram showing an example of the standards for folder names and file names.

[0019] Figure 3A This is a diagram showing an example of definition information stored in the classification definition information storage unit.

[0020] Figure 3B This is a diagram showing an example of a creation system (definition) of words belonging to a category.

[0021] Figure 4A This is a diagram showing an example of information type determination keywords stored in the information type determination dictionary.

[0022] Figure 4B This is a diagram showing an example of keywords stored in a keyword dictionary.

[0023] Figure 5 This is a flowchart showing the flow of information search processing as one embodiment of the present invention.

[0024] Figure 6A This is a diagram showing an example of a first screen displayed on a display device in the information search method according to the embodiment.

[0025] Figure 6B This is a diagram showing an example of a second screen displayed on the display device in the information search method according to the embodiment.

[0026] Figure 6C This is a diagram showing an example of a third screen displayed on the display device in the information search method according to the embodiment.

[0027] Figure 6D This is a diagram showing an example of the relevance between a search keyword and each word (filtering keyword) belonging to a category selected by a searcher in the information search method according to the embodiment.

[0028] Figure 6E This is a diagram showing an example of a fourth screen displayed on the display device in the information search method according to the embodiment.

[0029] Figure 6F This is a diagram showing an example of a fifth screen displayed on the display device in the information search method according to the embodiment.

[0030] Figure 6G This is a diagram showing an example of a sixth screen displayed on the display device in the information search method according to the embodiment.

[0031] Figure 6H This is a diagram showing an example of a seventh screen displayed on the display device in the information search method according to the embodiment.

[0032] Figure 6I This is a diagram showing an example of an eighth screen displayed on the display device in the information search method according to the embodiment.

[0033] Figure 7AThis is a diagram showing an example of a ninth screen displayed on the display device in the information search method according to the embodiment.

[0034] Figure 7B This is a diagram showing an example of a tenth screen displayed on the display device in the information search method according to the embodiment.

[0035] Figure 7C This is a diagram showing an example of an eleventh screen displayed on the display device in the information search method according to the embodiment. DETAILED DESCRIPTION

[0036] The following describes an information retrieval system and method according to embodiments of the present invention. It should be noted that, as an example of the information to be retrieved, the following description uses equipment maintenance information in the manufacturing industry. In the following description, the term "information" or "data" is used to refer to "equipment maintenance information."

[0037] Information Retrieval System

[0038] Reference Figures 1 to 4B The structure of the information retrieval system involved in the embodiment is described. Figure 1 As shown, the information search system according to the embodiment includes an information management device 10 for managing information, a database 12 storing maintenance work information of a plurality of devices to be searched, a plurality of file servers 14 , and a searcher-side terminal 20 .

[0039] The information management device 10, database 12, multiple file servers 14, and searcher-side terminals 20 are each connected via a network 18. Network 18 is, for example, a communication network managed by the organization (company) to which the user belongs, such as a LAN (Local Area Network). However, network 18 may be a communication network partially interposed with a general-purpose public line, such as a WAN (Wide Area Network) or a VAN (Virtual Private Network), in addition to a public communication network such as the Internet.

[0040] The information management device 10 and the searcher-side terminal 20 are implemented by a computer and a program, respectively. Examples of the searcher-side terminal 20 include portable devices such as smartphones, personal computers, and tablet computers. An input device 22 such as a keyboard and mouse, and a display device 24 such as a monitor, are connected to or integrated with the searcher-side terminal 20.

[0041] Database 12 aggregates and stores documents such as reports and manuals, drawings, and photographs stored in each file server 14. Information stored in database 12 includes information such as operating standards and manuals regarding equipment failures, malfunctions, their causes, and their solutions. In this embodiment, file servers 14 are deployed regionally, assuming that specific equipment exists in multiple locations (e.g., multiple countries or multiple counties).

[0042] The information management device 10 performs a search process based on a search condition including a search keyword input by a searcher and one or more screening keywords extracted by correlation analysis of the search keyword, and retrieves information matching the search condition from the database 12 .

[0043] The information management device 10 includes an information registration unit 110, an information search unit 120, and various storage units (a keyword dictionary 50, a classification definition information storage unit 52, a search information storage unit 54, and an information type determination dictionary 56). It should be noted that in this embodiment, the components implementing 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 also be a single component.

[0044] The information registration unit 110 is configured to execute the registration function described above, and more specifically, to execute processing for storing information stored in each file server 14 in the database 12. The information registration unit 110 includes an information coordination unit 112, an information processing unit 114, and a storage processing unit 116.

[0045] The information linking unit 112 automatically links the information stored in the database 12 with the information stored in each file server 14 regularly or irregularly. For example, when new data 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.

[0046] The information processing unit 114 performs predetermined data processing on the data acquired by the information linking unit 112. The information processing unit 114 adds information type information to the data as part of the data processing. The information type information is information indicating to which information type the document belongs.

[0047] The information processing unit 114 specifies the information type of the data based on, for example, the name of the folder storing the data, the file name of the data, and the like. Figure 2 An example of a standard for defining the folder names prepared in each file server 14 and the file names of data stored in each folder is shown.

[0048] When the information type is determined based on the folder, the information processing unit 114 refers to, for example, Figure 2 , the information type of the data stored in the "Failure Report" folder is set to "Failure Case." Furthermore, the information processing unit 114 sets the information type of the data stored in the "Work Standards," "Production Procedures," and "Safety Materials" folders to "Manual," and the information type of the data stored in the "Maintenance Techniques" folder to "Maintenance Techniques." Furthermore, the information processing unit 114 sets the information type of the data stored in the "Drawings" folder to "Drawings," the information type of the data stored in the "Maintenance Ledger" folder to "Maintenance Ledger," and the information type of the data stored in the "Operation Standards" folder to "Operation Standards."

[0049] When the information processing unit 114 determines the type of information based on the file name, for example, the creator of the data Figure 2 The file name is created based on the file creation standard, and therefore, based on the file name, it is determined that the data belongs to the "failure case".

[0050] As data processing, the information processing unit 114 also adds classification information to the data based on the classification information stored in the classification definition information storage unit 52 . Figure 3A An example of the classification information stored in the classification definition information storage unit 52 is shown. Figure 3B The criteria shown are used to define to which of multiple categories a particular word belongs.

[0051] exist Figure 3A For example, it is shown that the word "abc" belongs to the category "status" and is assigned "D51" as the classification information. Therefore, for example, when the acquired data contains the word "abc", the classification information "D51" assigned to the word "abc" is assigned to the data. Figure 1 , followed by an explanation of the remaining components.

[0052] 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 according to the type of information (see Figure 1 ).

[0053] The information stored in the database 12 is pre-classified into a plurality of groups A, B, C, etc. according to the types of information. Figure 1In the illustrated database 12, Group A of the first information category stores information related to "Fault Cases," Group B of the second information category stores information related to "Manuals," and Group C of the third information category stores information related to "Drawings." Furthermore, although not shown in this figure, Group D of the fourth information category, Group E of the fifth information category, and Group F of the sixth information category store information related to "Maintenance Techniques," "Maintenance Books," and "Operation Standards," respectively.

[0054] The information retrieval unit 120 searches for and extracts specific equipment maintenance information from the database 12 based on a search instruction from the searcher-side terminal 20. The information retrieval unit 120 includes a query 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 filtering keyword suggestion unit 127, and a search execution unit 130.

[0055] The query receiving unit 121 receives queries input by a searcher via the input device 22 of the searcher-side terminal 20 via the network 18. The query input format is not particularly limited; for example, one or more words or a natural language sentence can be input. For example, a sentence such as "What should be done if a BB device in the AA device fails?" or a query consisting of multiple words such as "AA device BB device" separated by spaces can be input from the searcher-side terminal 20.

[0056] The language processing unit 122 extracts words (independent words) by performing language processing such as known morpheme analysis on the query received by the query receiving unit 121 .

[0057] Specifically, language processing unit 122 extracts input search keywords to be used in subsequent searches, as well as information category determination keywords used to determine the type of information intended for the search. For example, if the query "What should be done if the BB device in AA equipment fails?" is input, language processing unit 122 extracts "AA equipment" and "BB device" as input search keywords.

[0058] Furthermore, the language processing unit 122 considers the context of the query received by the query receiving unit 121 and extracts the input search keyword by distinguishing between positive and negative forms. For example, if the query contains the word "burned," the language processing unit 122 distinguishes between "burned" and "not burned" and extracts the input search keyword based on this.

[0059] In addition, the language processing unit 122 refers to the information type determination dictionary 56 (refer to Figure 4A), extracting words that match those registered in the information type determination dictionary 56 as information type determination keywords. For example, if a query sentence such as "What is the response when a BB device in AA equipment fails?" is input as described above, the language processing unit 122 extracts the terms "failure" and "response" as information type determination keywords.

[0060] Next, the keyword dictionary 50 will be described before the search candidate keyword reading unit 123. In the keyword dictionary 50, related keywords are registered and stored in advance in a manner of being associated with each other.

[0061] In the keyword dictionary 50, for example, mutually synonymous keywords are registered in a manner associated with each other. Furthermore, mutually synonymous keywords are registered in a manner associated with each other. However, the keyword dictionary 50 may only register mutually synonymous keywords or only register mutually synonymous keywords.

[0062] Figure 4B 1 shows an example of keywords pre-registered in the keyword dictionary 50. In the figure, keywords belonging to the same row (column) are related to each other and are associated with each other.

[0063] The search candidate keyword reading unit 123 refers to the keyword dictionary 50 every time the language processing unit 122 extracts an input search keyword, and reads keywords related to the input search keyword as search candidate keywords.

[0064] For example, when the input search keyword is "AA device" as described above, the search candidate keyword reading unit 123 reads Figure 4B The keyword dictionary 50 shown is searched for a match. All other keywords in the row containing "AA device" (the first row in the figure) or, if specified (for example, only synonyms), the specified keywords are read as search candidate keywords.

[0065] In this example, as synonyms of "AA device", "00 device", "A'A' device", "A"A"" are read out, and as synonyms of "AA device", "aaa", "a'a'a'" are read out. In addition, the same is true for "BB device". Figure 4B In the second line, the associated keywords are read out as search candidate keywords.

[0066] Next, the search information storage unit 54 will be described before the information type determination unit 124. The search information storage unit 54 stores, as search keywords, the input search keyword and information type determination keyword extracted by the language processing unit 122, and the search candidate keywords read by the search candidate keyword reading unit 123. For example, each time an additional query is input and the additional search keyword is extracted (read), the search information storage unit 54 cumulatively stores the search keyword.

[0067] The information type determination unit 124 determines the type of information that represents the search intention of the searcher based on the information type determination keyword extracted from the query. Figure 4A The information type determination dictionary 56 shown determines the type of information to be searched based on the information type determination keywords.

[0068] In the information type determination dictionary 56, information type determination keywords assumed to be extracted from a query and the corresponding information types are registered in advance in a state of being associated with each other. Figure 4A In the information type judgment dictionary 56, the information type "fault case" is registered in advance with the information type judgment keywords such as "fault", "similar example" and "response". In addition, in the information type judgment dictionary 56 of the figure, the information type "manual" is registered in advance with the information type judgment keywords such as "fault type" and "replacement procedure". It should be noted that although the illustration is omitted in the figure, Figure 2 The displayed information types "maintenance techniques", "drawings", "maintenance books", and "operation standards" also have keywords registered for information type judgment.

[0069] The classification information receiving unit 125 displays the categories to which the filtering keywords belong in a selectable format. For example, in the example above, searching for information belonging to group A, "Fault Cases," using search keywords such as "AA Equipment" and "BB Device" (including synonyms and near-synonyms) would likely result in an overwhelming number of search results. Therefore, the classification information receiving unit 125 (and the correlation calculation unit 126 and filtering keyword display unit, described later) allows the searcher to add or exclude filtering keywords to narrow the search results.

[0070] The classification information receiving unit 125 displays a plurality of classifications in a selectable manner. Specifically, the classification information receiving unit 125 displays, for example, a pull-down menu so that any one of the classifications including "production line", "equipment", "device", "component", "state", and "cause" can be selected (see the following description). Figure 6C ).

[0071] The correlation calculation unit 126 calculates the 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 correlation based on, for example, the frequency with which the search keyword and each word belonging to the category are both listed in the same report, manual, etc. Alternatively, the correlation calculation unit 126 can calculate the correlation based on the number of times the search keyword and each word belonging to the category are combined as the search keyword stored in the search information storage unit 54.

[0072] It should be noted that when there are multiple search keywords, the relevance is calculated for some or all of the search keywords based on the searcher's selection. It should be noted that the "search keyword" mentioned above specifically refers to the input search keyword extracted by the language processing unit 122 and the search candidate keyword read by the search candidate keyword reading unit 123.

[0073] The filtering keyword prompting unit 127 prompts the searcher in a selectable manner whether to add each word (hereinafter referred to as "filtering keyword") belonging to the category selected by the searcher in the category information receiving unit 125 to the search condition. That is, the filtering keyword prompting unit 127 accepts a selection operation for the searcher to select whether to set the above-mentioned filtering keyword as an AND condition ("and" condition) with respect to the search keyword in the search condition of the database 12 (see the following description). Figures 6E to 6G ). In addition, the filtering keyword prompting unit 127 accepts a selection operation for the searcher to select whether to set the above-mentioned filtering keyword as a NOT condition ("non" condition) for the search keyword in the search condition of the database 12 (refer to the following description). Figures 6E to 6G Specifically, this selection operation includes three modes: (1) setting the filter keyword as an AND condition for the search keyword, (2) setting it as a NOT condition, and (3) setting it neither as an AND condition nor as a NOT condition (skip).

[0074] It should be noted that the above-mentioned filtering keywords are words that are distinguished between positive and negative forms. For example, the filtering keyword related to "burned" can be either "burned" or "not burned".

[0075] Here, there is also a case where, through the processing in the correlation calculation unit 126, a plurality of screening keywords including the first screening keyword and the second screening keyword in descending order of relevance to the search keyword are extracted as screening keywords (see the following description). Figure 6D ). For example, Figure 6D In the example, the first filter keyword is "abc" and the second filter keyword is "def".

[0076] In this case, the filtering keyword suggestion unit 127 receives a selection operation as to whether to set the first filtering keyword and the second filtering keyword, which have relatively high relevance to the search keyword, as an AND condition for the search keyword in this order (see the following description). Figure 6E ). In addition, the filtering keyword prompting unit 127 receives a selection operation related to whether to set the first filtering keyword and the second filtering keyword with a relatively high relevance to the search keyword as the NOT condition for the search keyword in this order (refer to the following description). Figure 6F Specifically, this selection operation includes three modes: (1) setting the first and second filtering keywords as AND conditions for the search keyword, (2) setting them as NOT conditions, and (3) setting them neither as AND conditions nor as NOT conditions (skip).

[0077] In addition, when the searcher selects whether to set the first filtering keyword as an AND condition or as a NOT condition, the filtering keyword prompting unit 127 performs a temporary search process (temporary search process). Furthermore, when the number of entries of data matching the search condition is less than a predetermined threshold, the filtering keyword prompting unit 127 does not accept the selection operation of the second filtering keyword. That is, when the result of performing a temporary search by adding or excluding the first filtering keyword is that the search results become sufficiently small, the filtering keyword prompting unit 127 does not display the screen for selecting whether to add the second filtering keyword to the search condition (described later). Figure 6F ).

[0078] The search execution unit 130 performs a text-based matching search process between the search keywords stored in the search information storage unit 54 and the filtering keywords selected by the searcher, for the information stored in the database 12. Furthermore, the search execution unit 130 performs the matching search process described above on information within a group corresponding to the information type determined by the information type determination unit 124, among the plurality of groups A, B, C, etc. within the database 12.

[0079] [Information Retrieval Method]

[0080] Reference Figures 5 to 7C An information retrieval method executed by the information retrieval system according to the embodiment will be described. Figure 5 Flowchart showing the process of the information retrieval method according to the embodiment. Figures 6A to 7C This is a screen displayed on the display device 24 of the searcher-side terminal 20 and is an example of an application screen for using the information search system according to the embodiment. The information search method according to the embodiment starts when the searcher activates the above application on the searcher-side terminal 20, for example.

[0081] First, the query receiving unit 121 determines whether a query has been received (step S101). For example, when the searcher starts the application, the display device 24 displays Figure 6A Then, when the searcher inputs a query into the search space 200 via the input device 22 and selects (for example, clicks) the send icon 201, the query accepting unit 121 accepts the query.

[0082] In this example, the query receiving unit 121 receives the query "What should I do if a failure occurs in the BB device of the AA device?" Note that a history display area 300 is provided at the bottom of the screen, allowing all or part of previously entered queries into the search space 200 to be copied.

[0083] In step S101 , when the query accepting unit 121 accepts a query (Yes in step S101 ), the process proceeds to step S102 . When no query is accepted (No in step S101 ), the process returns to step S101 .

[0084] Next, the language processing unit 122 performs natural language processing such as morpheme analysis on the query sentence received by the query receiving unit 121 to extract words (independent words) (step S102). In this example, the language processing unit 122 extracts "AA device," "BB device," "failure," and "response" from the query sentence "What is the response when the BB device in the AA device fails?"

[0085] Next, the language processing unit 122 extracts (selects) input search keywords and information type identification keywords from the words extracted by the language processing (step S103). In this example, the language processing unit 122 extracts "AA equipment" and "BB device" as input search keywords and "fault" and "response" as information type identification keywords.

[0086] Next, the search candidate keyword reading unit 123 refers to the keyword dictionary 50 (refer to Figure 4B ), and keywords associated with the input search keyword are read out as search candidate keywords (step S104). In this example, the search candidate keyword reading unit 123 reads out synonyms and near synonyms belonging to "AA device" as the input search keyword extracted in step S103, and synonyms and near synonyms belonging to "BB device" as search candidate keywords.

[0087] Next, the search information storage unit 54 stores the input search keyword and the search candidate keywords read by the search candidate keyword reading unit 123 (step S105). In this example, the search information storage unit 54 stores "AA device" and "BB device" and their synonyms and antonyms.

[0088] Next, the information type determination unit 124 determines the information type based on the information type determination keywords extracted in step S103 and the information type determination dictionary 56 (see Figure 4A ), to determine the information type (search intention) (step S106). In this example, the information type determination unit 124 determines that "fault case" associated with "fault" and "response" is the information type.

[0089] Next, the information search unit 120 uses the searcher-side terminal 20, for example, Figure 6B The search period, information type, search keyword, and classification information are displayed on the display device 24 as shown (step S107). Note that the "search keyword" mentioned above specifically refers to the input search keyword extracted by the language processing unit 122 and the search candidate keyword read by the search candidate keyword reading unit 123.

[0090] exist Figure 6B The screen displays information types, search keywords, and classification information in a selectable format, allowing the searcher to modify these as needed. Furthermore, an arbitrary search period can be set. If no period is specified, the search will be executed for the entire period. Furthermore, search keywords can be removed from the search by unchecking the checkboxes preceding each keyword.

[0091] In the search keyword display area, a word request icon 203 is displayed in a selectable manner, and it is possible to request to add or delete other keywords similar to the keyword displayed as the search keyword. In addition, as for the classification information, although "unspecified" is selected in the initial state, by operating this part using the input device 22, it is possible to, for example, Figure 6C In this example, the case where the information type and the input search keyword have not changed and the classification information has changed from "unspecified" to "status" is explained as an example.

[0092] If the classification information is changed from "unspecified" to another classification in step S107 and the classification information receiving unit 125 receives 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.

[0093] Next, the correlation calculation unit 126 calculates the correlation between the search keyword selected by the searcher in step S107 and each word belonging to the category received by the category information receiving unit 125 (step S109). Figure 6D As shown, the relevance between the two keywords "AA equipment" and "BB device" and each word (abc, def, ...) belonging to the category "state" is calculated.

[0094] Next, the filtering keyword prompting unit 127 displays the filtering keyword with the highest relevance to the search keyword (Nth (N=1 for the first time)) on the display device 24 via the searcher-side terminal 20 (step S110). Figure 6D As shown, the words (abc, def, ...) belong to the category "state".

[0095] Next, the filtering keyword prompting unit 127 performs a provisional search based on the search conditions selected by the searcher (step S111). Next, the filtering keyword prompting unit 127 determines whether a prescribed condition is met (step S112). If the prescribed condition is met, the process proceeds to step S113. If the prescribed condition is not met, the process returns to step S110 and sets "N = N + 1." The "prescribed condition" mentioned above refers to whether or not the filtering keyword is prompted to the searcher. Examples of prescribed conditions include the absence of relevant filtering keywords and the number of search hits during the provisional search being less than a few dozen (e.g., 50).

[0096] The following describes the details of the processing of steps S110 to S112. Figure 6E As shown, it is prompted in a selectable manner whether to set the screening keyword "abc" with the highest correlation "1" with the search keyword as an exclusion condition (NOT condition) or as an additional condition (AND condition). In addition, the screen of the figure shows the current search HIT number when only the search keyword is temporarily searched, "exclusion keyword setting", "additional keyword setting" and "skip" icons, etc. The "skip" icon is a button selected when neither the screening keyword is excluded nor the screening keyword is added. In this example, the case of setting the screening keyword "abc" as an additional keyword is used as an example for explanation.

[0097] Next, the filtering keyword suggestion unit 127 performs a temporary search based on the search keyword and the additional keyword "abc". Figure 6FAs shown, the user is prompted to select whether to set the filtering keyword "def," which has the second highest relevance to the search keyword, as an exclusion condition (NOT condition) or as an addition condition (AND condition). In this example, the case where the filtering keyword "def" is set as an exclusion keyword is explained.

[0098] Next, the filtering keyword suggestion unit 127 performs a temporary search based on the search keyword, the additional keyword "abc" and the exclusion keyword "def". Figure 6G As shown, it is prompted in a selectable manner whether to set the screening keyword "ghi" with the highest relevance "3" to the search keyword as an exclusion condition (NOT condition) or as an addition condition (AND condition). In this example, the case where the screening keyword "ghi" is set as an exclusion keyword is explained as an example. The screening keyword prompting unit 127 repeatedly executes Figures 6E to 6G The displayed filtering keywords are presented until the above-mentioned predetermined conditions are met.

[0099] As described above, in the information search system according to the embodiment, by suggesting to the searcher to add or exclude filtering keywords until the number of search results becomes equal to or less than a predetermined number, the information stored in the database 12 can be filtered easily and visually.

[0100] Next, the filtering keyword suggestion unit 127 uses the searcher-side terminal 20, for example, Figure 6H The information type, search keyword, classification information, and filtering keywords (additional keywords, exclusion keywords) are displayed on the display device 24 as shown (step S113). It should be noted that at this stage, additional keywords and exclusion keywords can be removed from the search target by unchecking the check boxes located before the additional keywords and exclusion keywords.

[0101] Next, the search execution unit 130 selects the search icon 202 (see Figure 6H ), a search process is performed on the information stored in the database 12 (step S114). Specifically, the search execution unit 130 performs a text-based matching search process on the information in the database 12 that belongs to the group corresponding to the selected information type based on the input search keyword, search candidate keywords, and screening keywords.

[0102] Next, the search execution unit 130 performs the following operations, for example: Figure 6I The search results are displayed on the display device 24 on the searcher side as shown (step S115), and this process is terminated.

[0103] Here, in Figures 6E to 6G In the order of relevance to the search keyword from high to low, the searcher is prompted to decide whether to set the filtering keyword as an exclusion condition or an additional condition, but the prompting method is not limited to this. Figures 7A to 7C As shown, a plurality of filtering keywords, relevance of the filtering keywords to the search keywords, and check boxes for adding or excluding the filtering keywords may be presented.

[0104] In this case, if the searcher adds or excludes any of the filtering keywords, the filtering keyword prompting unit 127 performs a temporary search under the conditions and displays the current search HIT number when the temporary search is performed on the screen. Figure 7A In the example, when the filter keyword "abc" is added, the number of search hits is "××××× items". However, Figure 7B As shown in , by further excluding the screening keyword "def", the number of search HITs is reduced to "×××× items". Figure 7C As shown, by further excluding the screening keyword "ghi", the number of search HITs is reduced to "××× items".

[0105] in addition, Figures 7A to 7C The "Add to Keywords" icon in each screen is used to register the word selected by the check box as an additional keyword or an excluded keyword. Figure 7A As shown in FIG. 1 , when the “add to keyword” icon is pressed while “abc” is selected in the additional check box, “abc” is officially registered as an additional keyword and transferred to Figure 6H In addition, for example, Figure 7B and Figure 7C As shown in FIG, when the "add to keywords" icon is pressed in the state where "def" and "ghi" are selected in the exclusion check box, "def" and "ghi" are officially registered as exclusion keywords and transferred to Figure 6H 's picture.

[0106] in addition, Figures 7A to 7C The "skip" icon in each screen is used to skip the selection of each word. Figure 7A As shown in FIG, when the "skip" icon is pressed in the state where "abc" can be added or excluded (the row of "abc" is highlighted), the addition or exclusion of "abc" is skipped. Then, the row of "def" in the next section is highlighted, and the state is moved to where "def" can be added or excluded. In addition, for example, Figure 7BAs shown in FIG, when the "skip" icon is pressed in the state where "def" can be added or excluded (the state where the "def" line is highlighted), the addition or exclusion of "def" is skipped. Then, the line of "ghi" in the lower section is highlighted, and the state is moved to where "ghi" can be added or excluded. In addition, for example, Figure 7C As shown, when the "skip" icon is pressed in a state where "ghi" can be added or excluded (a state where the row of "ghi" is highlighted), the addition or exclusion of "ghi" is skipped. Figure 6H 's picture.

[0107] Thus, in the information search system according to the embodiment, the number of search results when adding or excluding filtering keywords is displayed, and the searcher is advised to add or exclude filtering keywords.

[0108] In the information retrieval method and system according to the embodiments described above, input search keywords are extracted from a query entered by the searcher, and the searcher is then allowed to select whether to use the filtering keywords extracted through correlation analysis as an AND condition or a NOT condition. This allows efficient retrieval of desired information from the vast amount of information stored in database 12. As a result, for example, if a failure or malfunction is reported, the searcher can enter the condition in sentence form, efficiently retrieving reference information such as past examples from database 12. Based on this retrieved information, the searcher can quickly perform equipment recovery operations.

[0109] Furthermore, the information search method and information search system according to the embodiment can exclude highly relevant filtering keywords, thereby efficiently narrowing down the number of search results. Therefore, even in a situation where there are many similar cases in the past, the target information can be accurately found.

[0110] While the information retrieval method and information retrieval system according to the present invention have been specifically described above through the embodiments and examples for implementing the invention, the gist of the present invention is not limited to the above description and must be broadly interpreted based on the claims. Furthermore, various modifications and variations based on the above description are naturally also included in the gist of the present invention.

[0111] For example, the above embodiment describes an example of searching for equipment maintenance information in the manufacturing industry, but the present invention is not limited to this. The information retrieval method and information retrieval system of the present invention can be applied to all types of Q&A (Question & Answer) information, such as information indicating corresponding treatment methods and disease names for disease symptoms, and information indicating corresponding treatment methods for error names and statuses displayed by machinery such as vehicles.

[0112] In the above embodiment, the relevance to the search keyword is calculated only once (see Figure 5 Step S109), query the first to third most relevant screening keywords in sequence (refer to Figures 6E to 6G ), but the relevance can be calculated multiple times. In this case, the filtering keyword prompting unit 127 calculates the relevance of multiple filtering keywords to the search keyword (refer to Figure 6D ), the display device 24 prompts whether to add or exclude the most relevant screening keywords.

[0113] When the searcher selects to add or exclude the filtering keyword, the filtering keyword prompting unit 127 recalculates the relevance under these conditions and prompts the display device 24 whether to add or exclude the most relevant filtering keyword. The filtering keyword prompting unit 127 repeats the relevance calculation and filtering keyword prompting process until the aforementioned conditions are met. By calculating the relevance each time a filtering keyword is prompted, search accuracy can be improved.

[0114] Description of Reference Signs

[0115] 10Information management device

[0116] 12 Database

[0117] 14 File Server

[0118] 18 Network

[0119] 20Searcher side terminal

[0120] 22 Input devices

[0121] 24 display devices

[0122] 50 keyword dictionary

[0123] 52 Classification definition information storage unit

[0124] 54 retrieval information storage unit

[0125] 56 Dictionary for determining information type

[0126] 110 Information Registration Department

[0127] 112 Information Cooperation Department

[0128] 114 Information Processing Department

[0129] 116 Storage Processing Unit

[0130] 120 Information Retrieval Department

[0131] 121 Inquiry Receiving Department

[0132] 122 Language Processing Department

[0133] 123 search candidate keyword reading unit

[0134] 124 Information Type Judgment Unit

[0135] 125 Classified Information Receiving Department

[0136] 126 Related Calculation Department

[0137] 127 Keyword Prompt Department for Screening

[0138] 130 Search Execution Department

Claims

1. An information retrieval method, comprising: an information management device performing a retrieval process based on retrieval conditions including keywords input by a searcher and one or more screening keywords extracted by correlation analysis of the keywords, and retrieving information matching the retrieval conditions from a database; The information management device executes the search process after receiving a selection operation for the searcher to select whether to set the narrowing-down keyword as an AND condition for the keyword or as a NOT condition for the keyword in the search condition.

2. The information retrieval method according to claim 1, wherein: When a plurality of filtering keywords including a first filtering keyword and a second filtering keyword whose relevance to the keyword is ordered from high to low are extracted as the filtering keyword, the information management device accepts the selection operation in the order of the first filtering keyword and the second filtering keyword whose relevance to the keyword is relatively high.

3. The information retrieval method according to claim 1, wherein: The information management device executes the search process when the searcher selects whether to use the first screening keyword as an AND condition or as a NOT condition. When the number of data items matching the search condition is less than a predetermined threshold, the selection operation for the second narrowing-down keyword is not accepted.

4. The information retrieval method according to claim 1, wherein: The correlation analysis is performed based on classification information arbitrarily set from a plurality of classifications indicating a relationship between the keyword and the filtering keyword.

5. An information retrieval system comprising: a database; and an information management device, which performs a search process based on a search condition including a keyword input by a searcher and one or more screening keywords extracted by correlation analysis of the keyword, and retrieves information matching the search condition from the database, wherein: The information management device executes the search process after receiving a selection operation for the searcher to select whether to set the narrowing-down keyword as an AND condition for the keyword or as a NOT condition for the keyword in the search condition.

Images