Information Retrieval System

Abstract

The information retrieval system of the present invention comprises a database for storing business auxiliary information and an information management device capable of retrieving specified information from the business auxiliary information, the information management device comprising: a question accepting unit for accepting input of questions including keywords from a searcher; a partition information accepting unit for accepting input of partition information to which a screening keyword belongs; a screening keyword prompting unit for prompting screening keywords that belong to the input partition information and have a correlation with the keyword in a selectable manner; a retrieval execution unit for executing retrieval processing, the retrieval execution unit executing retrieval processing on the database based on the input keywords and the selected screening keywords.

Description

Technical Field

[0001] The present invention relates to an information retrieval system capable of efficiently retrieving desired information from a database storing business auxiliary information. Background Art

[0002] In order to retrieve desired information from a vast amount of business-related information, 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 result in the inability to retrieve the desired information. Against this backdrop, Patent Document 1 proposes an information retrieval system that performs searches using 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] Patent Document 1: Japanese Patent Application Laid-Open No. 2019-121392

[0004] However, the information retrieval system described in Patent Document 1 performs searches including synonyms, resulting in a large number of search results depending on the input keyword. As a result, even if the desired information is included in the search results, it takes a considerable amount of time to identify the desired information by individually checking the large number of search results. Summary of the Invention

[0005] The present invention has been made in view of the above-mentioned problems, and an object of the present invention is to provide an information retrieval system capable of efficiently identifying desired information from a vast amount of business-related information.

[0006] The information retrieval system of the present invention comprises a database for storing business auxiliary information and an information management device capable of retrieving specified information from the business auxiliary information, the information management device comprising: a question acceptance unit for accepting input of questions including keywords from a searcher; a partition information acceptance unit for accepting input of partition information to which a screening keyword belongs; a screening keyword prompting unit for prompting screening keywords that belong to the input partition information and have a correlation with the keyword in a selectable manner; and a retrieval execution unit for executing retrieval processing, the retrieval execution unit executing retrieval processing on the database based on the input keyword and the selected screening keyword.

[0007] It can also be that: the business auxiliary information is equipment maintenance information, the filtering keywords prompted by the filtering keyword prompting unit are keywords that are related to the keywords entered in the question acceptance unit in the equipment maintenance business, and the filtering keyword prompting unit prompts two or more filtering keywords together with the relevance.

[0008] The method may further include a storage unit that stores partition definition information that defines to which of the plurality of partition information a specific filtering keyword belongs.

[0009] According to the present invention, it is possible to provide an information retrieval system capable of efficiently identifying desired information from a vast amount of business-related information. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 This is a block diagram showing the configuration of an information retrieval system as one embodiment of the present invention.

[0011] Figure 2 This is a diagram showing an example of the standards for folder names and file names.

[0012] Figure 3A This is a diagram showing an example of definition information stored in the partition definition information storage unit.

[0013] Figure 3B This is a diagram showing an example of a creation system (definition) of words belonging to a partition.

[0014] Figure 4A This is a diagram showing an example of information type determination keywords stored in the information type determination dictionary.

[0015] Figure 4B This is a diagram showing an example of keywords stored in the keyword dictionary.

[0016] Figure 5 This is a flowchart showing the flow of information search processing as one embodiment of the present invention.

[0017] Figure 6A This is a diagram showing an example of a first screen displayed on the display device.

[0018] Figure 6B This is a diagram showing an example of the second screen displayed on the display device.

[0019] Figure 6C This is a diagram showing an example of the third screen displayed on the display device.

[0020] Figure 6D This is a diagram showing an example of the fourth screen displayed on the display device.

[0021] Figure 6E This is a diagram showing an example of the fifth screen displayed on the display device.

[0022] Figure 6F This is a diagram showing an example of the sixth screen displayed on the display device. DETAILED DESCRIPTION

[0023] The following describes an information retrieval system as one embodiment of the present invention. Furthermore, as an example of auxiliary business information, equipment maintenance business information for the manufacturing industry is described below. In the following description, the term "information" or "data" is assumed to represent "equipment maintenance business information."

[0024] 〔structure〕

[0025] First, refer to Figure 1 The configuration of an information retrieval system as one embodiment of the present invention will be described. Figure 1 : is a block diagram showing the structure of an information retrieval system as one embodiment of the present invention. Figure 1 As shown, an information retrieval system as one embodiment of the present invention 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. Furthermore, the information management device 10, the database 12, the plurality of file servers 14, and the searcher-side terminal 20 are connected via a network 18.

[0026] The network 18 is a communication 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 communication network that partially includes a general 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.

[0027] The information management device 10 and the searcher-side terminal 20 are implemented as a computer and a program, respectively. Examples of the searcher-side terminal 20 include mobile 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.

[0028] The database 12 stores a collection of documents such as reports and manuals, drawings, and photos stored in each file server 14. Documents can be not only documents originally input as text, but also documents converted from sounds using a voice recognition engine, or documents converted from sentences in paper media using an OCR (Optical Character Reader). When storing drawings and photos, they are stored in a form that allows for retrieval of text or file names entered into the drawings. The information stored in the database 12 includes failures, malfunctions, and their causes of devices, etc., as well as information such as operating standards and manuals on their response methods. In this embodiment, it is assumed that a specific device exists in multiple regions (e.g., multiple countries, multiple counties), and file servers 14 are set up according to the regions.

[0029] The information management device 10 includes an information registration unit 110, an information retrieval unit 120, and various storage units (a keyword dictionary 50, a partition definition information storage unit 52, a retrieval information storage unit 54, and an information type determination dictionary 56). In this embodiment, for ease of understanding, the configuration for implementing the registration function and the retrieval function is described separately from the information registration unit 110 and the information retrieval unit 120. However, both functions may be implemented by a single configuration. As described above, the information registration unit 110 is configured to be able to perform the registration function, and more specifically, to perform the process of 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.

[0030] The information coordination unit 112 automatically coordinates the information stored in the database 12 with the information stored in each file server 14 on a regular or irregular basis. When new data that is not stored in the database 12 is stored in the file server 14, the information coordination unit 112 obtains the new data via the network 18. The information processing unit 114 performs predetermined data processing on the data obtained by the information coordination unit 112. In this embodiment, as data processing, information type information is added to the data. Information type information is information indicating to which information type the document belongs. The information processing unit 114 determines this based on, for example, the folder in which the data is stored in each file server 14 and / or the file name of the data. Figure 2 This is a diagram showing an example of a standard for specifying the folder names prepared in each file server 14 and the file names of the data stored in each folder. Figure 2The information processing unit 114 sets the information type of the data stored in the folder "Fault Report" to "Fault Case", the information type of the data stored in the folders "Work Standard Book", "Production Procedure Book" and "Safety Material" to "Manual", the information type of the data stored in the folder "Maintenance Technique" to "Maintenance Technique", the information type of the data stored in the folder "Drawing" to "Drawing", the information type of the data stored in the folder "Maintenance Script" to "Maintenance Script", and the information type of the data stored in the folder "Operation Standard" to "Operation Standard". In the case of determining by the file name, the explanation of each one is omitted, but for example, the creator of the data may determine the type of the data based on the file name. Figure 2 By creating a file name according to the file creation criteria shown, it is possible to determine that the data belongs to a "failure case" based on the file name.

[0031] Furthermore, the information processing unit 114 also adds partition information to data based on the partition information stored in the partition definition information storage unit 52 as data processing. Figure 3A FIG. 5 is a diagram showing an example of partition information stored in the partition definition information storage unit 52. Figure 3A As shown, the partition information is based on Figure 3B The shown base defines to which partition a particular word belongs among multiple partitions. Figure 3A For example, the word "abc" is shown as belonging to the "status" partition and is assigned "D51" as partition information. If the acquired data includes the word "abc," the data is assigned the partition information "D51" assigned to the word "abc." While this partition information assignment is not necessary, performing this process allows for efficient calculation of correlations by the correlation calculation unit 126, described later.

[0032] return Figure 1 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 116 stores the stored data in the database 12 according to the type of information (see Figure 1 ). In addition, in addition to the regular or irregular storage of the information registration unit 110, the data can also be stored in the database 12 directly by the registrant via the searcher side terminal 20. In the case of direct registration by the registrant, on the screen for the registrant to operate, Figure 2 The indicated criteria are displayed on the display device 24 of the searcher-side terminal 20 , etc., so that the registrant can store the data in the database 12 after appropriately setting the file name of the data.

[0033] 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 question accepting unit 121, a language processing unit 122, a search candidate keyword reading unit 123, an information type determination unit 124, a partition information accepting unit 125, a correlation calculation unit 126, a filtering keyword suggesting unit 127, and a search execution unit 130.

[0034] Each functional module generally performs the following functions. The question accepting unit 121 receives questions from a searcher via the input device 22 of the searcher-side terminal 20 via the network 18. The format of the question input is not particularly limited; it can be a single word or a natural language sentence. For example, a sentence such as "What should I do if a BB device in the AA device fails?" or a question separated by multiple words such as "AA device BB device" separated by spaces is input from the searcher-side terminal 20.

[0035] The language processing unit 122 extracts words (independent words) by performing language processing such as known morpheme analysis on the question received by the question receiving unit 121. Specifically, the language processing unit 122 extracts input search keywords used in subsequent searches and information type judgment keywords used in judging the type of information as the search intention (purpose) from the input question. For example, in the above example, when the question sentence "What should be done when the BB device in the AA device fails?" is input, the language processing unit 122 extracts "AA device" and "BB device" as input search keywords. In addition, the language processing unit 122 refers to the information type judgment dictionary 56 ( Figure 4A ) to extract words that match the words registered in the information type determination dictionary 56 as information type determination keywords. That is, in the above example, the language processing unit 122 extracts the terms "failure" and "response" as information type determination keywords.

[0036] Next, the search candidate keyword reading unit 123 will be described, but before that, the keyword dictionary 50 will be described. Related keywords are pre-registered and stored in the keyword dictionary 50 in a correspondence relationship with each other. In this embodiment, synonymous keywords are registered in the keyword dictionary 50 in a correspondence relationship with each other. In addition, mutually antonymous keywords are registered in the keyword dictionary 50 in a correspondence relationship with each other. However, in the keyword dictionary 50, only synonymous keywords can be registered, or only antonymous keywords can be registered. Alternatively or in addition to the above, keywords that are considered to have high mutual relevance can be pre-registered in a correspondence relationship with each other. In addition, words registered in the keyword dictionary 50 can be appropriately added or deleted.

[0037] exist Figure 4B An example of keywords pre-registered in the keyword dictionary 50 is shown in FIG. Figure 4B In the example, keywords belonging to the same row (column) are related keywords and have established a corresponding relationship with each other.

[0038] Whenever the language processing unit 122 extracts an input search keyword, the search candidate keyword reading unit 123 refers to the keyword dictionary 50 to read out keywords related to the input search keyword as search candidate keywords. According to the above example, when the input search keyword is "AA device", the search candidate keyword reading unit 123 reads out keywords related to the input search keyword as search candidate keywords. Figure 4B The keyword dictionary 50 shown in the example performs a consistent search and reads out all other keywords belonging to the row including "AA device" (row 1 in the figure), or reads out specified keywords when there is a designation (for example, only synonyms) as search candidate keywords. In this example, "00 device", "A'A' device", and "A"A"" are read out as synonyms of "AA device", and "aaa", and "a'a'a'" are read out as synonyms of "AA device". The same is true for the "BB device" in the above example, refer to Figure 4B The second line of is used to read out the associated keywords as search candidate keywords.

[0039] The search information storage unit 54 stores the search candidate keywords and information category determination keywords obtained as described above as search keywords. The search information storage unit 54 cumulatively stores the search keywords each time an additional question is input and an additional search keyword is extracted.

[0040] The information type determination unit 124 determines the information type that is the search intention of the searcher based on the information type determination keyword extracted from the question. Figure 4A The information type judgment dictionary 56 shown in the figure is used to judge the type of information to be retrieved based on the information type judgment keywords. The information type judgment keywords extracted from the question and the corresponding information types are pre-registered in the information type judgment dictionary 56 in a state of establishing a corresponding relationship. For example, for information type judgment keywords such as "fault", "similar examples" and "response", "fault case" is pre-registered in the information type judgment dictionary 56 as an information type in a state of establishing a corresponding relationship. For information type judgment keywords such as "type of fault" and "replacement steps", "manual" is pre-registered in the information type judgment dictionary 56 as an information type in a state of establishing a corresponding relationship. Although in Figure 4A The illustration is omitted, but the above reference Figure 2The information types described as “maintenance tips”, “drawings”, “maintenance scripts”, and “operation standards” also have information type determination keywords stored in the information type determination dictionary 56 .

[0041] As described above, the information stored in the database 12 is pre-classified into a plurality of groups A, B, C, etc. by the information registration unit 110 (the storage processing unit 116) according to the type of information. Figure 1 In the example shown, 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." Although not shown, 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 tips," "maintenance scripts," and "operation standards," respectively. Furthermore, if the information category determined by information category determination unit 124 differs from the searcher's intent, the searcher can change the information category through operation. For example, in the example above, Group A "fault cases" is initially assigned as the information category for information category determination keywords such as "fault" and "response." However, the searcher can change the information category to "manuals" to retrieve the desired information from information in Group B. Furthermore, if the information category determination keyword is not extracted from the input question, Group A "fault cases" is initially assigned as the information category, but the searcher can change the information category as appropriate.

[0042] The partition information receiving unit 125 displays the partitions to which the filtering keywords belong, allowing users to select them. In the above example, 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 result in a potentially large number of search results. To reduce the number of search results, additional keywords are required. The partition information receiving unit 125, the correlation calculation unit 126 (described later), and the filtering keyword display unit assist the searcher in adding filtering keywords.

[0043] The partition information receiving unit 125 displays a plurality of partitions in a selectable manner. In this embodiment, for example, a pull-down menu is displayed in which any partition can be selected from a plurality of partitions including "production line", "equipment", "device", "component", "state" and "cause" (for example, referring to Figure 6C ). In addition, the number of partitions is configured to be able to be added or changed as appropriate.

[0044] The correlation calculation unit 126 calculates the relevance between the search keyword and each word belonging to the partition selected by the searcher in the partition information reception unit 125. The relevance is calculated based on, for example, the frequency with which the search keyword and each word belonging to the partition are simultaneously described in the same report, manual, etc., or the number of times the search keyword and each word belonging to the partition are combined as search keywords stored in the search information storage unit 54. Furthermore, if there are multiple search keywords, the relevance is calculated for some or all of the search keywords based on the searcher's selection.

[0045] The filtering keyword suggestion unit 127 suggests each word belonging to the partition selected by the searcher in the partition information receiving unit 125 together with the relevance calculated by the relevance calculation unit 126 to the searcher in a selectable manner (for example, referring to Figure 6D The searcher can select one or more filtering keywords by referring to the relevance. When the searcher selects a specific filtering keyword, the selected filtering keyword is added and stored in the search information storage unit 54 as a search keyword.

[0046] The search execution unit 130 performs a text-based search for information stored in the database 12, based on a match between the search keyword and the search key words stored in the search information storage unit 54 (including the input search key words extracted by the language processing unit 122 and the filtering key words selected by the searcher in the filtering key word prompting unit 127). In this embodiment, the search execution unit 130 performs a text-based search for information in the groups A, B, C, etc. in the database 12 that correspond to the information type determined by the information type determination unit 124, based on a match between the search key words and the information.

[0047] Information retrieval processing

[0048] Next, refer to Figure 5 、 Figures 6A to 6F , the operation of the information retrieval system when the information retrieval unit 120 performs information retrieval processing will be described. Figure 5 This is a flowchart showing the flow of information search processing as one embodiment of the present invention. Figures 6A to 6F 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 of the present invention. For example, when the searcher starts the above application on the searcher side terminal 20, Figure 5 Then, in step S101, it is determined whether the question accepting unit 121 has accepted the question. If the question has been accepted (yes in S101), the process proceeds to step S102. If the searcher starts the application, the display device 24 displays, for example, Figure 6AThe screen shown. The searcher can enter a question in the search space 200 via the input device 22. The searcher enters a question in the search space 200 and then selects (e.g., clicks) the send icon 201, whereupon the question receiving unit 121 receives the question. In this example, the question receiving unit 121 receives the question "What should I do if a failure occurs in the BB device of the AA device?" Furthermore, a history display area 300 is provided at the bottom of the screen. This area allows for copying all or part of questions previously entered into the search space 200, allowing users to select any range of past questions, copy them, and then paste them into the search space 200.

[0049] Next, the language processing unit 122 performs natural language processing such as morpheme analysis on the question sentence received by the question receiving unit 121, thereby extracting words (independent words) (step S102). In this example, "AA device," "BB device," "failure," and "response" are extracted from the question sentence "What should I do if a failure occurs in the BB device of the AA device?"

[0050] Next, the language processing unit 122 extracts (selects) input search keywords based on the words extracted by the language processing, and also extracts (selects) keywords for determining the type of information (step S103). In this example, the language processing unit 122 extracts "AA device" and "BB device" as input search keywords, and references the information type determination dictionary 56 to extract (select) the words "fault" and "response" included in the information type determination dictionary 56 as keywords for determining the type of information.

[0051] Next, the search candidate keyword reading unit 123 refers to the keyword dictionary 50 and reads out keywords related to the input search keyword as search candidate keywords (step S104). Figure 4B The keyword dictionary 50 illustrated in FIG is used to read out the synonyms and antonyms included in the first row belonging to the input search keyword "AA device" and the synonyms and antonyms included in the second row belonging to "BB device" extracted in the process of step S103 as search candidate keywords.

[0052] 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 device" and "BB device" and their synonyms and antonyms.

[0053] Next, the information type determination unit 124 determines the information type based on the information type determination keyword extracted in step S103 and the Figure 4AThe information type determination dictionary 56 shown in the example is used to determine the information type (search intent) (step S106). In this example, "failure" and "response" are extracted as keywords for information type determination. The information type determination unit 124 refers to the information type determination dictionary 56 and determines that "failure case" is an information type because it is associated with "failure" and "response".

[0054] Next, for example, Figure 6B As shown, the information search unit 120 displays the partition information, along with the search period, the information types extracted and determined in the above steps, and the input search keyword, on the display device 24 via the searcher-side terminal 20 (step S107). The information types, input search keywords, and partition information are displayed on the display device 24 in a selectable and deletable manner, allowing the searcher to change them as needed. For example, when changing the information type, the searcher can change the information type by selecting another type using the input device 22. The search period can be set to any desired period. If no period is specified, the search can be performed for the entire period. Search keywords can be excluded from the search by unchecking the checkboxes preceding each keyword. A word request icon 203 is displayed in a selectable manner in the search keyword display area, allowing the user to request the addition of other keywords similar to the displayed search keyword or the deletion of inappropriate keywords from the displayed search keyword. Keywords requested to be added or deleted can be added or deleted when the number of addition or deletion requests reaches a specified number, or they can be added or deleted as appropriate by a system administrator who has received the addition or deletion request. Partition information is composed of: Figure 6B As shown, "No designation" is selected in the initial state, but by operating the part using the input device 22, the Figure 6C The pull-down menu is displayed as shown, and the desired partition can be selected. In this example, Figure 6B This example explains the case where the information type and input search keyword remain unchanged, but the partition information changes from "Unspecified" to "Status".

[0055] In step S107, if the partition information has been changed from "Unspecified" to another partition, that is, if the partition information receiving unit 125 has accepted a partition other than "Unspecified" (Yes in step S108), the process proceeds to step S109. If the partition information remains "Unspecified" (No in step S108), the process proceeds to step S113. In step S109, 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 partition accepted by the partition information receiving unit 125. In this example, the correlation between the two keywords "AA device" and "BB device" and each word belonging to the partition "status" is calculated.

[0056] Then, the filtering keyword suggestion unit 127 sends the searcher-side terminal 20, for example, Figure 6D As shown in FIG. 1 , each word belonging to the partition accepted by the partition information accepting unit 125 is used as a filtering keyword and displayed on the display device 24 together with the relevance calculated in step S109 (step S110). The searcher can add one or more filtering keywords that he thinks are appropriate among the filtering keywords displayed on the display device 24 to the search keyword. In this example, it is assumed that Figure 6D The word "abc" shown in FIG. 1 is used as a keyword for screening. In addition, the display form of the relevance is not particularly limited and can be as follows: Figure 6D Although it is shown in a graph (chart) as shown, it can also be shown in a numerical value as an example of another display form.

[0057] In step S112, the information search unit 120 displays the information type extracted, determined, or selected by the searcher in the above steps, the input search keyword, and the partition information on the display device 24 via the searcher-side terminal 20, depending on whether the filtering keyword is added in step S110 (step S111). Specifically, in the case where one or more filtering keywords are added to the keywords in step S110, Figure 6E As shown, a screen that can recognize the added filtering keywords is displayed. If the filtering keywords are not added to the keywords, the screen Figure 6B The screen shown.

[0058] Then, select the search icon 202 (see Figure 6B or Figure 6E ), whereby the search execution unit 130 performs a search process on the information stored in the database 12. Specifically, in a case where a filtering keyword is further added based on the input search keyword based on the question input by the searcher, the search execution unit 130 also performs a text-based consistent search process on the information stored in the database 12 belonging to the group corresponding to the selected information type based on the filtering keyword (step S113). At this time, in a case where the information type is a drawing, the search execution unit 130 performs a search on the file name marked on the drawing and / or the text information filled into the drawing as text data. And, finally, as Figure 6F As shown, the information search unit 120 displays the search result on the display device 24 on the searcher side (step S114).

[0059] exist Figure 6FIn the search result screen shown, when multiple information (documents) are hit by the search, they are displayed according to the prescribed criteria (for example, the order of the information's storage date from the latest to the earliest). However, information with prescribed rating information attached is displayed at the top with priority over the prescribed criteria. Here, as prescribed rating information, Figure 6F The number of stars assigned to "useful for problem solving" and "applicable to rare problems" is shown in the example. This rating information can be assigned by the searcher, and can be assigned by selecting stars at any position (in Figure 6F In the , you can assign rating information by selecting any star among the 5 stars. Figure 6F In the example, for the first information displayed, "4" is assigned to the item "Useful for dealing with problems", and "2" is assigned to the item "Applicable to rare problems". In addition, for the second information displayed, "2" is assigned to the item "Useful for dealing with problems", and "3" is assigned to the item "Applicable to rare problems". The larger the value, the higher the rating, that is, the more useful the information. In this embodiment, two items ("Useful for dealing with problems" and "Applicable to rare problems") are illustrated as rating information, but it can be either one, or there can be more than three items. In the case where there are multiple items in the rating information, the display order can be determined based on the rating information of one of the items, or the display order can be determined based on the rating information of all items (for example, the total number of stars of all items).

[0060] According to the present invention, input search keywords are extracted from the searcher's question, and filtering keywords are presented along with their relevance based on the input search keywords and the partitioning of filtering keywords. Therefore, even those with limited knowledge and experience can select appropriate keywords as filtering keywords. The searcher's requested data can then be accurately extracted from the database based on the input search keywords and filtering keywords, enabling efficient retrieval. Thus, in the event of a malfunction or failure, the searcher can input the situation as a statement, and information serving as a reference for past incidents, etc., can be efficiently retrieved from the database. Based on this retrieved information, the searcher can quickly perform equipment malfunction repair operations.

[0061] In the above embodiment, an information retrieval system for equipment maintenance business information in the manufacturing industry is described, but the present invention is not limited to this embodiment. For example, it can be applied to a retrieval system for Q&A (Question & Answer) information as a whole, such as information indicating coping methods corresponding to disease symptoms, disease names, and information indicating coping methods corresponding to error names and states displayed on machinery such as automobiles.

[0062] In the above embodiment, each data item in the database 12 is stored in groups according to information type. During a search, in addition to the input search keyword and the filter keyword, the search is performed based on the information type. However, grouping by information type and searching based on the information type are not required. According to the present embodiment, as described above, it is possible to search based on the filter keyword in addition to the input search keyword. Even without performing a search based on the information type, it is possible to efficiently search and extract the desired information.

[0063] exist Figure 6A In the example, the past questions input by the searcher are displayed in the history display area 300, but a predetermined number of past questions are displayed according to the size of the history display area 300. Therefore, it is also possible to configure the system so that when the searcher wants to confirm the past questions that are not displayed in the history display area 300, log download can be performed separately. Log download can be obtained, for example, from a predetermined setting screen (such as a setting screen for searcher information, etc., not shown in the figure), and log data of the search actions of oneself (if the administrator has administrator authority of the system, it is for oneself and others) in any period can be obtained. The log data includes various operations performed by the above-mentioned searcher, that is, search conditions (questions, search period, additional filter keywords, etc.), browsed data information, rating additional operation information, etc.

[0064] While the embodiments of the invention developed by the present inventors have been described above, the present invention is not limited to the description and drawings that constitute part of the disclosure of the present invention based on these embodiments. In other words, other embodiments, examples, and application techniques developed by those skilled in the art based on these embodiments are all within the scope of the present invention.

[0065] According to the present invention, it is possible to provide an information retrieval system capable of efficiently identifying desired information from a vast amount of business-related information.

[0066] Description of reference numerals:

[0067] 10…information management device; 12…database; 14…file server; 18…network; 20…searcher-side terminal; 22…input device; 24…display device; 50…keyword dictionary; 52…partition definition information storage unit; 54…retrieval information storage unit; 56…information type judgment dictionary; 110…information registration unit; 112…information collaboration unit; 114…information processing unit; 116…storage processing unit; 120…information retrieval unit; 121…question acceptance unit; 122…language processing unit; 123…retrieval candidate keyword reading unit; 124…information type judgment unit; 125…partition information acceptance unit; 126…correlation calculation unit; 127…keyword prompting unit for screening; 130…retrieval execution unit.

Claims

1. An information retrieval system comprising a database storing device maintenance information and an information management device capable of retrieving predetermined information from the device maintenance information. The information retrieval system is characterized in that The information management device comprises: A question receiving unit receives input of questions including keywords from searchers; A partition information receiving unit receives input of partition information to which a keyword for screening belongs; a screening keyword presenting unit configured to present screening keywords that belong to the input section information and are related to the keyword in a selectable manner; and A search execution unit executes search processing. The search execution unit executes a search process on the database based on the input keyword and the selected filtering keyword. The screening keyword suggested by the screening keyword suggestion unit is a keyword related to the keyword input by the question receiving unit in the equipment maintenance business, The filtering keyword prompting unit prompts two or more filtering keywords together with their relevance. The relevance is the relevance between the filtering keyword and the input keyword, and is calculated based on the number of times the search process has been performed in combination with the keyword in the past. When the partition information receiving unit receives input of partition information and the partition information is changed, the correlation is calculated.

2. The information retrieval system according to claim 1, characterized in that The information search system includes a storage unit that stores section definition information that defines to which section information a specific filtering keyword belongs among a plurality of section information.

Images