JP2021152709A - Retrieval support system, and retrieval support method - Google Patents

Abstract

To provide a retrieval support system and a retrieval support method for facilitating information retrieval by a user.SOLUTION: A retrieval support system 1 comprises: a data registration information creation unit 104 which creates retrieval object information which associates one or plural items with information indicating a content of the item; a retrieval information acquisition unit 102 which receives an input of a prescribed identifier indicating a prescribed item, and the input of retrieval information related to the item, as a retrieval condition, from a user; a query conversion unit 106 which converts the inputted identifier into the prescribed item; and a user presentation unit 101 which displays the information indicating the content of the item corresponding to the prescribed item in the retrieval object information, when it is determined that the retrieval object information satisfying the retrieval condition has been retrieved.SELECTED DRAWING: Figure 1

Description

The present invention relates to a search support system and a search support method.

Attempts have been made to improve the efficiency of operations by sharing the knowledge accumulated in relation to the operations of the organization with each member in the organization. For example, business information in a company is stored in a database, and each user can search this database to obtain necessary information.

A chat system equipped with a text input / output interface is used for this search system. By using the chat system, a plurality of users can share the same screen on each terminal, and each user can have a discussion while performing a text search as appropriate.

However, since various types of databases are accumulated in a large amount in a company, in order to effectively utilize the chat system, the search system must be adapted to the chat system.

For example, when discussing using a chat system, the user needs to perform a quick search while combining various search conditions. Therefore, when inputting the search conditions, it is preferable that the search conditions can be input as concisely as possible.

Also, when displaying search results, if too much information is displayed, it becomes difficult to grasp the necessary information, which may delay the discussion.

Here, in order to facilitate the construction and retrieval of a database, an automatic database system creation system that enables data definition has been proposed. For example, Patent Document 1 includes a data definition generation unit that generates data definition information based on input data, and an input screen file generation unit and an analysis screen file generation unit that automatically generate input screen and analysis screen HTML files, respectively. It has a DB registration program generation unit and a DB search program generation unit that automatically generate CGI programs for database registration and search, respectively, and a database generation unit that automatically generates a database, and each of these generation units is the same. It is described that each file etc. is automatically generated based on the data definition information of.

Japanese Unexamined Patent Publication No. 11-23290

However, according to Patent Document 1, since the user must specify the search condition according to a predetermined dedicated input form, it is difficult to apply it to a chat system in which the user interface is fixed. Further, regarding the output of the search result, Patent Document 1 does not have a mechanism for adjusting the amount of information to be output so as to be suitable for a chat system.

Further, in Patent Document 1, since the user must consider the definition of data by himself / herself, it may be difficult to search the user when the items in the database are complicated.

The present invention has been made in view of such a current situation, and an object of the present invention is to provide a search support system and a search support method capable of facilitating information retrieval by a user.

One of the present inventions for solving the above-mentioned problems is a data registration information creation process for creating search target information in which one or a plurality of items are associated with information indicating the contents of the items, and a user can use the data registration information creation process. As search conditions, a search information acquisition process that accepts input of a predetermined identifier representing the predetermined item and input of search information related to the item, and a query conversion process that converts the input identifier into the predetermined item. , A calculation device that executes a user presentation process for displaying information indicating the content of an item corresponding to the predetermined item in the search target information when it is determined that the search target information satisfying the search condition can be searched. It is a search support system equipped with.

Another one of the present invention for solving the above-mentioned problems is data registration information in which the information processing apparatus creates search target information in which one or a plurality of items are associated with information indicating the contents of the items. The creation process, the search information acquisition process that accepts the input of a predetermined identifier representing the predetermined item as the search condition from the user, and the input of the search information related to the item, and the input identifier as the predetermined item. Query conversion processing to convert to, and user presentation processing to display information indicating the content of the item corresponding to the predetermined item in the search target information when it is determined that the search target information satisfying the search condition can be searched. And, the search support method to execute.

According to the present invention, information retrieval by a user can be facilitated.

Issues, configurations and effects other than those described above will be clarified by the description of the following embodiments.

It is a figure explaining an example of the structure of the search support system which concerns on this embodiment. It is a figure explaining an example of the hardware provided in each information processing apparatus in a search support system. It is a flowchart explaining an example of an index creation process. It is a flowchart explaining an example of data definition information creation processing. It is a figure which shows an example of knowledge data. It is a figure which shows an example of the data definition information. It is a flowchart explaining an example of ID column setting processing. It is a flowchart explaining an example of a metadata column setting process. It is a flowchart explaining an example of a data type setting process. It is a figure which shows an example of the index DB schema. It is a flowchart explaining an example of a data registration process. It is a figure which shows an example of the process of creating data registration information. It is a flowchart explaining an example of a search execution process. It is a figure which shows an example of the search input information. It is a flowchart explaining an example of a query conversion process. It is a figure which shows an example of analysis information. It is a figure which shows an example of a search query. It is a flowchart explaining an example of a user presentation process. It is a figure which shows an example of the search result information. It is a figure which shows an example of the screen (chat screen) displayed on the user's terminal based on the chat output information.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

<System configuration>
FIG. 1 is a diagram illustrating an example of the configuration of the search support system 1 according to the present embodiment.

The search support system 1 includes a search support device 100 for acquiring knowledge data 1000, a search system 200, and a chat system 300.

The knowledge data 1000 is, for example, information that stores information on business operations within a company. The knowledge data 1000 is, for example, a collection of questions and answers (Q & A collection) related to various tasks, data related to failures occurring in connection with the tasks, and the like.

The search support device 100 acquires the knowledge data 1000, and based on the acquired knowledge data 1000, predetermined index information (data registration information 4000 and index DB schema) for enabling the search system 200 to search the knowledge data 1000. 3000) is created, and the created index information is transmitted to the search system 200.

The knowledge data 1000 may be stored in the search support device 100, or may be stored in an information processing device external to the search support device 100.

The chat system 300 is a communication system that accepts data input from each terminal (not shown), which is an information processing device managed by a plurality of users, and transmits the input data to each terminal. For example, the chat system 300 can realize communication by a plurality of users by displaying the message information received from the terminals of each user on the terminals of all users.

Here, the chat system 300 receives the input of the predetermined search input information from the terminal of a certain user (search performer), and transmits the input search input information to the search support device 100. The search support device 100 converts the received search input information into a predetermined search request (search query) corresponding to the search system 200 based on the index information, and transmits the converted search query to the search system 200. After that, the chat system 300 transmits the search result information for the search query to the search system 200. The chat system 300 creates chat output information, which is screen information of search results suitable for viewing by the user, based on the data of the search results, and displays the created chat output information on the screen of each user's terminal.

The information processing devices in the search support system 1 are communicably connected by a wired or wireless network 5 such as a LAN (Local Area Network), WAN (Wide Area Network), the Internet, or a dedicated line.

Next, the details of the search support device 100 and the search system 200 will be described.

<Search system>
As shown in FIG. 1, the search system 200 stores the index DB 210. Further, the search system 200 includes a search unit 201 which is a functional unit.

The index DB 210 stores index data used for searching information of the knowledge data 1000.

The search unit 201 creates search result data (search result information) corresponding to the search query sent from the search support device 100 based on the index DB 210, and transmits the created search result information to the search support device 100. do.

<Search support device>
The search support device 100 includes a database acquisition unit 101, a search information acquisition unit 102, a data definition information creation unit 103, a data registration information creation unit 104, an index DB schema creation unit 105, a query conversion unit 106, and a user presentation unit 107. It has a functional unit.

The database acquisition unit 101 acquires the knowledge data 1000 having each item (column name).

In the present embodiment, it is assumed that the column name of the knowledge data 1000 is set by a Japanese or English character string.

The data registration information creation unit 104 provides search target information (data registration) in which one or more items (hereinafter referred to as physical names) corresponding to the column names in the knowledge data 1000 are associated with information indicating the contents of the items. Information 4000) is created.

The index DB schema creation unit 105 creates index information (index DB schema 3000) in which an item related to a physical name is associated with a data type of the item.

The search information acquisition unit 102 inputs, as search conditions, a predetermined identifier (hereinafter referred to as a column designation character) representing a predetermined item (physical name) related to the knowledge data 1000, and search information (search keyword) related to the item. ) Is accepted.

The query conversion unit 106 converts the identifier (column designation character) input by the search information acquisition unit 102 into a predetermined item (physical name).

The data definition information creation unit 103 creates an identifier (column designation character) representing each item for each of one or a plurality of items (physical names) in the knowledge data 1000 according to a predetermined rule.

In this case, the search information acquisition unit 102 inputs from the user any of the identifiers created by the data definition information creation unit 103 as an identifier (column designation character) representing a predetermined item (column name) in the knowledge data 1000. accept.

Further, the data definition information creation unit 103 sets each of the identifiers (column designation characters) to one character in a predetermined language (English in the present embodiment) representing each item.

Further, the data definition information creation unit 103 calculates the importance of each of one or a plurality of items (physical names), and assigns all identifiers (column designation characters) of the items whose calculated importance is less than a predetermined threshold value. Use the same identifier.

The user presentation unit 107 indicates the content of a predetermined item (physical name) in the search target information (data registration information 4000) when it is determined that the search target information (data registration information 4000) that satisfies the search condition can be searched. Create information (search result information).

Specifically, for example, when the user presentation unit 107 determines that the search target information (data registration information 4000) can be searched, the importance of the searched items is determined based on a predetermined determination condition, and the determination is made. Only when the importance is equal to or higher than a predetermined threshold, information (search result information) indicating the content of a predetermined item (physical name) in the search target information (data registration information 4000) is created.

Further, for example, when the user presentation unit 107 determines that the search target information (data registration information 4000) can be searched and the number of corresponding items (physical names) exceeds a predetermined number, a predetermined rule (for example, , Information indicating the content of the item in the search target information (data registration information 4000) (search result information) only for a predetermined number of the corresponding items based on the item having a high search score. ) Is created.

Here, FIG. 2 is a diagram illustrating an example of hardware included in each information processing device in the search support system 1. These information processing devices include a processor 11 such as a CPU, a storage device 12 such as a RAM (Random Access Memory) and a ROM (Read Only Memory), an HDD (Hard Disk Drive), an SSD (Solid State Drive), and a flash memory. It includes an auxiliary storage device 13, an input device 14 such as a keyboard or touch panel, an output device 15 such as a monitor or a display, and a communication device 16 such as a network interface card that communicates with other information processing devices.

Each function of each information processing device of the search support system 1 described above is realized by dedicated hardware or by reading a program stored in the storage device 12 or the auxiliary storage device 13 by the processor 11. .. Further, each program may be pre-recorded on a recording medium that can be read by each information processing device, or may be introduced when necessary via a storage medium or a predetermined communication network.

[Whole processing]
Next, the processing performed in the search support system 1 will be described.

First, the search support device 100 executes an index creation process for creating index information related to the search system 200 based on the knowledge data 1000.

Further, the search support device 100 executes a search execution process for causing the search system 200 to execute a search based on the search input information input from the user after executing the index creation process.

Hereinafter, the index creation process and the search execution process will be described.

<Indexing process>
FIG. 3 is a flowchart illustrating an example of the index creation process. The index creation process is started, for example, when the search support device 100 is activated, at a predetermined time, at a predetermined time interval, or when there is a predetermined input from the user.

First, the search support device 100 executes a data definition information creation process for creating data definition information 2000, which is attribute information of each item (column name) of the knowledge data 1000 and is information necessary for searching by the search system 200. (S1). The details of the data definition information creation process will be described later.

Then, the search support device 100 is based on the data definition information 2000 created in s1, and the content data in each item of the data definition information 2000 and each item of the knowledge data 1000 (
The index information (data registration information) associated with the character string) is created, and the data registration process of transmitting the created data registration information to the search system 200 is executed (s3). The details of the data registration process will be described later.

Further, the search support device 100 creates an index DB schema 3000 based on the data definition information 2000 created in s1, and executes an index DB schema creation process of transmitting the created index DB schema 3000 to the search system 200 (s5). ). Details of the index DB schema creation process will be described later.

This completes the index creation process. Next, the details of each process in the index creation process will be described.

<< Data definition information creation process >>
FIG. 4 is a flowchart illustrating an example of the data definition information creation process.

First, the database acquisition unit 101 of the search support device 100 acquires the knowledge data 1000 (s11).

Here, the details of the knowledge data 1000 will be described.

(Knowledge data)
FIG. 5 is a diagram showing an example of knowledge data 1000. The knowledge data 1000 includes a failure number 1001 (first item), which is an identifier that identifies a failure that has occurred in a predetermined task, a phenomenon 1002 and (second item), which are the contents of the failure related to the failure number 1001, and a failure. Cause 1003 of the failure related to the failure number 1001 (third item), countermeasure 1004 taken for the failure related to the failure number 1001 (fourth item), environment 1005 in which the failure related to the failure number 1001 occurred (third item) In this embodiment, the OS; Operating System) and (fifth item), the product name 1006 and (sixth item) that specify the business program (product in this embodiment) in which the failure related to the failure number 1001 has occurred, and the failure. It is composed of one or more records having each item including the registration date 1007 and (7th item) in which the content of the failure related to the number 1001 is registered.

Next, as shown in FIG. 4, the search support device 100 creates the data definition information 2000 based on the knowledge data 1000 acquired in s11 (s12-s17).

Here, a specific example of the data definition information 2000 will be described.

<Data definition information>
FIG. 6 is a diagram showing an example of data definition information 2000. In the data definition information 2000, the column name 2001, which is the name of each item in the knowledge data 1000, the physical name 2002, which is a character string representing the item related to the column name 2001, and the item related to the column name 2001 are key items (IDs). ID 2003, which is information indicating whether or not the data is, type 2004, which is the data type of the item related to the column name 2001, column designation character 2005, which is an identifier representing the item related to the column name 2001, and column name 2001. It consists of one or more records having each item including metadata 2006 indicating whether the item is metadata.

Subsequently, as shown in FIG. 4, the data definition information creation unit 103 registers the information representing the attributes of each item of the knowledge data 1000 in the data definition information 2000 (s12).

Specifically, for example, the data definition information creation unit 103 creates a new record in the data definition information 2000, and in the column name 2001 of each created record, the name of each item of the knowledge data 1000 acquired in s11 ( (Character string) is set (s12).

Next, the data definition information creation unit 103 sets the physical name (English) corresponding to the column name in the data definition information 2000 created in s12 (s13).

Specifically, for example, first, the data definition information creation unit 103 converts the character strings that are the names of each item of the knowledge data 1000 acquired in s11 into English character strings. When the converted English consists of a plurality of words, the data definition information creation unit 103 connects each word with a predetermined character string (for example, an underscore) to form one character string.

Then, the data definition information creation unit 103 sets each converted English character string to the physical name 2002 of each record in the data definition information 2000.

Subsequently, the data definition information creation unit 103 of the search support device 100 executes an ID column setting process for setting an ID (key item) in the data definition information 2000 created in s12 (s14). Details of the ID column setting process will be described later.

Further, the data definition information creation unit 103 of the search support device 100 executes a metadata column setting process for setting metadata information in the data definition information 2000 created in s12 (s15). Details of the metadata column setting process will be described later.

Further, the data definition information creation unit 103 of the search support device 100 executes a data type setting process for setting data type information in the data definition information 2000 created in s12 (s16). The details of the data type setting process will be described later.

Further, the data definition information creation unit 103 of the search support device 100 sets a column designation character in the data definition information 2000 created in s12 (s17), and then the data definition information creation process ends (s18).

Specifically, for example, the data definition information creation unit 103 uses a predetermined number of characters (for example, the first character) at a predetermined position in the character string of the physical name set in s13 as column designation characters.

If the column designation characters are duplicated in a plurality of items, an additional identifier may be added to the duplicate column designation characters. For example, when the character "P" is duplicated in a plurality of items, each column designation character is set to "P1", "P2", ....

Further, when there are a plurality of metadata items described later, all the column designation characters related to the plurality of metadata may be the same identifier or character (reference numeral 2007 in FIG. 2). This is because, as will be described later, the metadata items are of low importance and are not displayed as search results.

Next, the details of each process in the data definition information creation process will be described.

<<< ID column setting process >>>
FIG. 7 is a flowchart illustrating an example of the ID column setting process. First, the data definition information creation unit 103 of the search support device 100 sets the counter i to 1 (s141).

The data definition information creation unit 103 acquires all the data of the i-th item of the knowledge data 1000 (s142). Specifically, for example, the data definition information creation unit 103 acquires the character string of the failure number 1001, which is the first item of the knowledge data 1000 acquired in s11, from all the records.

The data definition information creation unit 103 confirms the number of characters of the character string acquired in s142, and determines whether or not all the character strings are a predetermined number of characters or less (for example, 10 characters or less) (s143).

When all the character strings are less than a predetermined number of characters (s143: YES), the data definition information creation unit 103 executes the process of s144 described later, and when there is a character string less than a predetermined number of characters (s143: NO). , The data definition information creation unit 103 executes the process of s147 described later.

In s144, the data definition information creation unit 103 confirms all the data (character strings) acquired in s143, and determines whether or not there are duplicate data (same character strings).

When there is duplicate data (same character strings) (s145: YES), the data definition information creation unit 103 executes the process of s146 described later, and when there is no duplicate data (same character strings) (s145: NO), the data definition information creation unit 103 executes the process of s147 described later.

In s146, the data definition information creation unit 103 sets in the data definition information 2000 information that the i-th item of the knowledge data 1000 is an ID (key item) and the other items are not IDs (key items). As a result, the ID column setting process ends (s149).

Specifically, for example, the data definition information creation unit 103 puts "○" in the ID 2003 of the record in which the i-th item of the knowledge data 1000 is set in the column name 2001 among the records of the data definition information 2000. Set, and set "x" to ID2003 of other records.

On the other hand, in s147, the data definition information creation unit 103 determines whether or not the current value of the counter i is the number of the last item in the knowledge data 1000 (7 in this embodiment).

When the value of the current counter i is the number of the last item in the knowledge data 1000 (s147: YES), the ID column setting process ends (s149).

On the other hand, if the value of the current counter i is not the number of the last item in the knowledge data 1000 (s147: NO), the data definition information creation unit 103 increments the value of the current counter i by 1 (s148). The processing after s142 is repeated.

<<< Metadata column setting process >>>
FIG. 8 is a flowchart illustrating an example of the metadata column setting process. First, the data definition information creation unit 103 of the search support device 100 sets a parameter t indicating a large number of variations in the data contents of each item in order to determine whether each item of the knowledge data 1000 is a metadata item. (S151).

Specifically, for example, the data definition information creation unit 103 sets the parameter t as "(total number of records of knowledge data 1000) / 10".

The data definition information creation unit 103 sets the value of the counter i to 1 (s152).

The data definition information creation unit 103 acquires the data of all the records of the i-th item of the knowledge data 1000 (s153).

Specifically, for example, the data definition information creation unit 103 acquires the contents of the failure number 1001, which is the first item of the knowledge data 1000 acquired in s11, for all the records.

The data definition information creation unit 103 specifies the total number n of data having different contents by comparing the data acquired in s153 with each other (s154).

Then, the data definition information creation unit 103 determines whether or not the total number n is smaller than the parameter t (s155).

When the total number n is smaller than the parameter t (s155: YES), the data definition information creation unit 103 executes the process of s156 described later.

On the other hand, when the total number n is not smaller than the parameter t (s155: NO), the data definition information creation unit 103 executes the process of s157 described later.

In s156, the data definition information creation unit 103 sets the data definition information 2000 to have information in which the i-th item of the knowledge data 1000 is a metadata item. After that, the process of s158 is performed.

Specifically, for example, the data definition information creation unit 103 adds "○" to the metadata 2006 of the record in which the i-th item of the knowledge data 1000 is set to the column name 2001 among the records of the data definition information 2000. "Is set.

In s157, the data definition information creation unit 103 sets the information that the i-th item of the knowledge data 1000 is not a metadata item in the data definition information 2000. After that, the process of s158 is performed.

Specifically, for example, the data definition information creation unit 103 adds "x" to the metadata 2006 of the record in which the i-th item of the knowledge data 1000 is set to the column name 2001 among the records of the data definition information 2000. "Is set.

In s158, the data definition information creation unit 103 determines whether or not the current i-th item of the knowledge data 1000 is the last item.

If the current i-th item of the knowledge data 1000 is not the last item (s158: NO), the data definition information creation unit 103 increments the value of the counter i by 1 (s159) and repeats the processing after s153.

On the other hand, if the current i-th item of the knowledge data 1000 is the last item (s158: YES), the metadata column setting process ends (s160).

<<< Data type setting process >>>
FIG. 9 is a flowchart illustrating an example of the data type setting process. First, the data definition information creation unit 103 of the search support device 100 sets the value of the counter i to 1 (s161).

The data definition information creation unit 103 acquires the data of all the records of the i-th item of the knowledge data 1000 (s162).

Specifically, for example, the data definition information creation unit 103 acquires the contents of the failure number 1001, which is the first item of the knowledge data 1000 acquired in s11, from all the records.

The data definition information creation unit 103 determines, for each data acquired in s162, whether or not the data has a date data format (s163).

Specifically, for example, the data definition information creation unit 103 analyzes the character string of the acquired data and determines whether or not this character string corresponds to the data format for storing the year, month, and day. ..

When all the data have the date data format (s164: YES), the data definition information creation unit 103 executes the process of s165 described later, and some data does not have the date data format. In the case (s164: NO), the data definition information creation unit 103 executes the process of s167 described later.

In s165, the data definition information creation unit 103 sets the information in which the data type of the i-th item of the knowledge data 1000 is the date type in the data definition information 2000. After that, the process of s166 described later is executed.

Specifically, for example, the data definition information creation unit 103 sets "date" in the record type 2004 in which the i-th item of the knowledge data 1000 is set in the column name 2001 in the data definition information 2000. do.

In s166, the data definition information creation unit 103 determines whether or not the current i-th item of the knowledge data 1000 is the last item.

If the current i-th item of the knowledge data 1000 is not the last item (s166: NO), the data definition information creation unit 103 increments the value of the counter i by 1 (s170) and repeats the processing after s162.

On the other hand, if the current i-th item of the knowledge data 1000 is the last item (s166: YES), the data type setting process ends (s171).

In s167, the data definition information creation unit 103 determines whether the i-th data item of the knowledge data 1000 acquired in s162 is an ID item or a metadata item based on the results of the ID column setting process and the metadata setting process. Judge whether or not.

When the item acquired in s162 is an ID or metadata item (s167: YES), the data definition information creation unit 103 executes the process of s168 described later, and even if the item acquired in s162 is an ID item, it is meta. If it is not a data item (s167: NO), the data definition information creation unit 103 executes the process of s169 described later.

In s168, the data definition information creation unit 103 sets the data type of the i-th item of the knowledge data 1000 to the keyword type (if this item is the search target from the user, the search keyword from the user is the item of the item. Unless all the data contents match, that is, unless the whole sentence matches, it will not be a search result). After that, the process of s166 is performed.

Specifically, for example, the data definition information creation unit 103 sets "keyword" in the record type 2004 in which the i-th item of the knowledge data 1000 is set in the column name 2001 in the data definition information 2000. do.

In s169, the data definition information creation unit 103 sets the data type of the current i-th item of the knowledge data 1000 to the text type (when this item is the search target from the user, the data type corresponds to the search keyword from the user. The data definition information 2000 is set to the information to be determined by a predetermined analyzer as to whether or not the data of the items match. After that, the process of s166 is performed.

Specifically, for example, the data definition information creation unit 103 sets "text" in the record type 2004 in which the i-th item of the knowledge data 1000 is set in the column name 2001 in the data definition information 2000. do.

<<< Index DB schema creation process >>>
The data definition information creation unit 103 creates an index DB schema in the index DB schema creation process.

FIG. 10 is a diagram showing an example of the index DB schema 3000. The index DB schema 3000 is a data group in which the character string 3001 indicating the physical name 2002 in the data definition information 2000 and the character string 3002 indicating the data type of the item related to the physical name 2002 are associated and stored.

If the data type is text type, information 3003 that identifies the analyzer to be used is displayed.

Next, the data registration process will be described.

<< Data registration process >>
FIG. 11 is a flowchart illustrating an example of the data registration process. First, the database acquisition unit 101 of the search support device 100 acquires the knowledge data 1000 (s31).

The data registration information creation unit 104 sets the value of the counter i to 1 (s32).

The data registration information creation unit 104 acquires all the data of the i-th item (hereinafter referred to as a selection item) of the knowledge data 1000 (s33).

Specifically, for example, the data registration information creation unit 104 acquires the contents of the failure number 1001, which is the first item of the knowledge data 1000 acquired in s31, from all the records.

The data registration information creation unit 104 creates data registration information based on the data acquired in s33 and the data definition information 2000 (s34).

Specifically, for example, the data registration information creation unit 104 acquires the physical name 2002 in the record in which the i-th item is set in the column name 2001 in the data definition information 2000, and obtains the acquired physical name 2002. , Creates information associated with the content of the i-th item in each record of the knowledge data 1000 acquired in s33.

(Data registration information)
FIG. 12 is a diagram showing an example of the process of creating the data registration information 4000. The data registration information 4000 is, for example, CSV format data. The figure is an example of associating a record with knowledge data 1000 with data definition information 2000.

The data registration information 4000 is information in which each physical name 4001 in the data definition information 2000 and the data 4002 of the knowledge data 1000, which is the content of each physical name 4001, are associated with each other by a predetermined delimiter.

Next, as shown in FIG. 11, the data registration information creation unit 104 transmits the data registration information created in s34 to the index DB 210 of the search system 200 (s35).

In s36, the data registration information creation unit 104 determines whether or not the i-th item of the knowledge data 1000 is the last item.

If the i-th item of the knowledge data 1000 is not the last item (s36: NO), the data registration information creation unit increments the value of the counter i by 1 (s38) and repeats the processing after s33.

On the other hand, when the i-th item of the knowledge data 1000 is the last item (s36: YES), the data registration process ends (s37).

<Search execution process>
Next, the search execution process executed after the index creation process will be described.

FIG. 13 is a flowchart illustrating an example of the search execution process. The search execution process is started after the index creation process is executed, for example, when a predetermined input is received from the user.

First, the search information acquisition unit 102 of the search support device 100 waits for the input (reception) of the search input information 5000 from any terminal of each user (s7: NO).

The search input information 5000 is a character or character string corresponding to the column designation character 2005 of the data definition information 2000, and a search character string (keyword, sentence) which is the data content of the item (physical name) indicated by the column designation character 2005. , Words, etc.) information.

(Search input information)
FIG. 14 is a diagram showing an example of the search input information 5000. The search input information 5000 has one or a plurality of search expressions 5003 including a column designation character 5001 and a search character string 5002 related to the column designation character 5001. A plurality of column designation characters 5001 can be associated with each search character string 5002.

Next, as shown in FIG. 13, when the user inputs the search input information 5000 (s7: YES), the search information acquisition unit 102 creates a search query based on the search input information 5000, and at the same time, creates a search query. A query conversion process for transmitting the created search query to the search system 200 is executed (s8). The details of the query conversion process will be described later.

After that, when the user presentation unit 107 receives the search result information corresponding to the search query from the search system 200, the user presentation unit 107 executes a user presentation process of converting the search result information into chat output information (s9). The details of the user presentation process will be described later. This completes the search execution process.

Next, the details of each process in the search execution process will be described.

<< Query conversion process >>
FIG. 15 is a flowchart illustrating an example of the query conversion process.

First, the query conversion unit 106 converts each column designation character 5001 in each search expression 5003 in the search input information 5000 input in s7 into the corresponding physical name 2002 based on the data registration information 4000 (s91).

Then, the query conversion unit 106 creates a search query by analyzing each search formula 5003 that has undergone this conversion according to the analysis information 6000 described later (s92).

(Analysis information)
FIG. 16 is a diagram showing an example of analysis information 6000. The analysis information 6000 stores one or more conversion rules.

When the search expression 5003 is a pair of the column specification character 5001 and the search character string 5002, the query conversion unit 106 follows the first conversion rule 6001 and includes the character string indicated by the search character string 5002. Create a search query to search for the item related to 5001. When a plurality of items are indicated by the column designation character 5001, the query conversion unit 106 creates a search query for each item.

When the search expression 5003 is a pair of the column designation character 5001 and the search character string 5002, and a predetermined symbol (for example, a minus symbol) is specified immediately before the column designation character 5001, the query conversion unit 106 uses the second According to the conversion rule 6002 of the above, a search query for searching the item related to the column designation character 5001 that does not include the character string indicated by the search character string 5002 is created. When a plurality of items are indicated by the column designation character 5001, the query conversion unit 106 creates a search query for each item.

When the search input information 5000 includes a plurality of search expressions 5003, the query conversion unit 106 searches for a character string that satisfies all the search results of each search expression 5003 according to the third conversion rule 6003 (AND condition). ..

(Search query)
Next, FIG. 17 is a diagram showing an example of a search query. The search query 7000 has a physical name character string 7001 and a search character string 7002 as information corresponding to the search expression 5003.

When there are a plurality of search formulas 5003, predetermined identifiers 7003 and 7004 indicating that all the search conditions of each search formula 5003 should be satisfied are set.

Next, the details of the user presentation process will be described.

<< User presentation processing >>
FIG. 18 is a flowchart illustrating an example of user presentation processing.

First, the user presentation unit 107 receives the search result information corresponding to the search query 7000 from the search system 200 (s91).

(Search result information)
FIG. 19 is a diagram showing an example of search result information. The search result information 8000 records the data of each item searched by the search query 7000 among the data of the knowledge data 1000.

That is, the search result information 8000 is configured to include one or a plurality of records having data of items related to the physical names 8001, 8002, 8003.

The physical names 8001, 8002, and 8003 include the physical name 8002 related to the metadata item.

Further, each record of the search result information 8000 has a physical name 8003 related to a search score (a parameter indicating the degree of matching of the search result with respect to the search condition).

Next, as shown in FIG. 18, the user presentation unit 107 chat outputs text information obtained by extracting only a predetermined number of search results related to a predetermined item from the search results indicated by the search result information 8000 received in s91. It is output as information (s92).

Specifically, for example, the user presentation unit 107 can change the data of each record of the search result information 8000 into a metadata item by referring to the column name 2001 and the metadata 2006 of each record of the data definition information 2000. Create record data (text data) excluding such data. Further, the user presentation unit 107 extracts the record data whose search score satisfies a predetermined condition (for example, the top predetermined number of search scores) from the created record data, and outputs the extracted data as chat output information. do.

Then, the user presentation unit 107 transmits the chat output information created in s92 to each user's terminal via the chat system 300 (s93). The terminal of each user displays this chat output information on a predetermined screen. This completes the user presentation process (s94).

(Chat screen)
FIG. 20 is a diagram showing an example of a screen (chat screen) displayed on the user's terminal based on the chat output information.

The chat screen 9000 is provided with a display field 9001 for displaying the search input information 5000 input by the user.

Further, the chat screen 9000 is provided with a display field 9002 for chat output information, which is a search result.

The search result block 9003 is displayed in the display field 9002. In each search result block 9003, the contents of the record corresponding to the search condition are displayed. Specifically, the column name 9004 corresponding to the physical name 8001 of the search result information 8000 and the character string representing the contents of the column name are displayed. 9005 is displayed.

The number of search results 9003 displayed is the search result block 9003 with a maximum of 3 search scores. As a result, only important search result information can be presented to the user, and the work efficiency of the user can be improved.

Here, in the search result 9003, the content of the physical name (information representing the metadata) related to the metadata among the physical names 8003 in the search result information 8000 is not displayed. In this way, by presenting only the information of important items to the user, the work efficiency of the user can be improved.

As described above, the search support system 1 of the present embodiment creates data registration information 4000 in which each item is associated with information indicating the content of the item, and the item is set as a search condition by the user. Accepts the input of the identifier (column designation character) representing the above and the input of the search information related to the item, converts the input identifier into the item of the data registration information 4000, and inputs the item of the data registration information 4000 that satisfies the above search condition. If the search is possible, the item of the data registration information 4000 and the content of the item in the data registration information 4000 are displayed.

As described above, according to the search support system 1 of the present embodiment, the user can obtain a desired search result by inputting an identifier representing the item when inputting the search condition. That is, according to the search support system 1 of the present embodiment, information retrieval by the user can be facilitated.

Although the embodiment for carrying out the present invention has been specifically described above, the present invention is not limited to this, and various modifications can be made without departing from the gist thereof.

For example, the content of the knowledge data 1000 is not particularly limited as long as it includes a predetermined item and its content data.

Further, in the present embodiment, the physical name is created by translating the column name from Japanese to English, but other methods may be used. For example, another language may be used, simplified Japanese may be used, or a predetermined symbol may be set.

Further, the data described in the database format in the present embodiment may be in another data format.

Further, in the present embodiment, the metadata is set as an item for which the display is omitted, but other items may be omitted. For example, it may be a preset item with a low priority.

Further, in the present embodiment, the metadata is not always displayed as the search result, but it may be displayed by the user's selection.

Further, as a method of omitting the display, in the present embodiment, the omission of the display of the metadata item and the limitation of the number of items to be displayed have been described, but the amount of information to be displayed may be changed from other viewpoints. For example, the information displayed on the screen size of each terminal may be adjusted.

The above description of the present specification clarifies at least the following. That is, in the search support system 1 of the present embodiment, the arithmetic unit performs data definition information creation processing for each of the one or a plurality of items to create an identifier representing each item according to a predetermined rule. It may be executed, and in the search information acquisition process, input of any one of the created identifiers may be accepted from the user as an identifier representing the predetermined item.

In this way, by creating an identifier (column designation character) representing each item and accepting input of the created identifier as an identifier (column designation character) representing the item from the user, the user can use it. Search input using an identifier can be performed easily and reliably.

Further, in the search support system 1 of the present embodiment, even if the arithmetic unit sets each of the identifiers to one character in a predetermined language representing each item in the data definition information creation process. good.

In this way, by setting each of the identifiers (column designation characters) to one character representing each item in the predetermined language, the user can easily recall the column designation character and search using the identifier. Input can be performed easily and reliably.

Further, in the search support system 1 of the present embodiment, when the arithmetic unit determines that the search target information can be searched in the user presentation process, the importance of the searched items is determined as a predetermined determination condition. The information indicating the content of the item corresponding to the predetermined item in the search target information may be displayed only when the determination is made based on the above and the determined importance is equal to or higher than the predetermined threshold value.

In this way, when an item of search target information (data registration information 4000) can be searched, the user can perform an important search by displaying the search result of the item only when the importance of the item is high. Only the resulting information can be known.

Further, in the search support system 1 of the present embodiment, the arithmetic unit calculates the importance of each of the one or a plurality of items in the data definition information creation process, and the calculated importance is less than a predetermined threshold value. It is also possible that all the above-mentioned identifiers of the items are the same identifier.

By using the same identifier for all non-important items (for example, metadata items) in this way, it is possible to suppress the consumption of resources related to the identifier when there are many such non-important items. ..

Further, in the search support system 1 of the present embodiment, when the arithmetic unit determines that the search target information can be searched in the user presentation process, the number of the corresponding items exceeds a predetermined number. , Information indicating the content of the item in the search target information may be displayed only for the predetermined number of the corresponding items based on the predetermined rule.

In this way, when the number of searchable items exceeds a predetermined number, the user can display the search results only for the predetermined number of searchable items in a limited screen. You can get the information of the search result that is especially necessary.

1 search support system, 100 search support device, 200 search system, 300 chat system, 1000 knowledge data

Claims (7)

Data registration information creation process that creates search target information that associates one or more items with information that indicates the content of the item.
Search information acquisition processing that accepts input of a predetermined identifier representing the predetermined item and input of search information related to the item as search conditions from the user.
A query conversion process that converts the input identifier into the predetermined item, and
When it is determined that the search target information satisfying the search condition can be searched, a user presentation process for displaying information indicating the content of the item corresponding to the predetermined item in the search target information, and a user presentation process.
A search support system equipped with an arithmetic unit that executes. The arithmetic unit
For each of the one or a plurality of items, a data definition information creation process for creating an identifier representing each item according to a predetermined rule is executed.
In the search information acquisition process, an input of any of the created identifiers is accepted from the user as an identifier representing the predetermined item.
The search support system according to claim 1. The arithmetic unit
In the data definition information creation process, each of the identifiers is set to one character in a predetermined language representing each of the items.
The search support system according to claim 2. The arithmetic unit
When it is determined that the search target information can be searched in the user presentation process, the importance of the searched item is determined based on a predetermined determination condition, and the determined importance is equal to or higher than a predetermined threshold value. Information indicating the content of the item corresponding to the predetermined item in the search target information is displayed only in
The search support system according to claim 1. The arithmetic unit
In the data definition information creation process, the importance of each of the one or a plurality of items is calculated, and all the identifiers of the items whose calculated importance is less than a predetermined threshold value are set as the same identifier.
The search support system according to claim 1. The arithmetic unit
If it is determined in the user presentation process that the search target information can be searched and the number of the corresponding items exceeds a predetermined number, the predetermined number of the corresponding items among the corresponding items are based on the predetermined rule. Information indicating the content of the item in the search target information is displayed only for
The search support system according to claim 1. Information processing device
Data registration information creation process that creates search target information that associates one or more items with information that indicates the content of the item.
Search information acquisition processing that accepts input of a predetermined identifier representing the predetermined item and input of search information related to the item as search conditions from the user.
A query conversion process that converts the input identifier into the predetermined item, and
When it is determined that the search target information satisfying the search condition can be searched, a user presentation process for displaying information indicating the content of the item corresponding to the predetermined item in the search target information, and a user presentation process.
Search support method to execute.

Images