JP6849904B2 - Search program, search device and search method - Google Patents

Description

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

A relational database is known as one of the databases used by a computer. A relational database is created by normalizing a data model based on an ER diagram (Entity Relationship Diagram) or the like. In the relational database, data is arranged in a table, and when searching for data, for example, a search using an index for improving search performance is performed. However, there are some searches for relational databases, such as partial match searches for Japanese sentences, where the index does not work effectively. Such a search on a relational database would be overkill and difficult to perform.

If it is difficult to search using a relational database, an unstructured database can be used. An unstructured database is created, for example, by denormalizing or lowering the normalization level based on a relational database. In a search against an unstructured database, the search performance depends on the data size of the Japanese text to be searched, not on the index. Therefore, by creating an unstructured database based on the relational database, it is possible to execute a search that is difficult to perform on the relational database on the unstructured database.

Here, a technique related to a database is proposed. For example, the workflow processing system extracts information on all data items of the electronic form and creates a matter attribute table. Further, for example, the common information extraction device is software that extracts 2 or more and n-1 or less software production support devices from n (n is a positive integer of 3 or more) software production support devices. Generate information common to production support devices as a common information file. The common information extraction device constitutes a common database including a common information file. The localized transfer device transfers the common database to one of the extracted software production support devices.

Japanese Unexamined Patent Publication No. 2000-163495 Japanese Unexamined Patent Publication No. 2000-330769

Unstructured databases can perform difficult searches on relational databases, but because of their low level of normalization, they contain redundant information. Therefore, the unstructured database has a larger amount of data than the relational database, and the increase in the amount of data hinders the shortening of the search time.

In one aspect, the present invention aims to speed up searches using unstructured databases.

In one aspect, a search program is provided. This search program acquires search information indicating the element name and the element content on the computer, and stores a plurality of parent elements having a plurality of child elements including the child element name and the child element content in an unstructured database. Searches for the first parent element having the first child element including the child element name and the child element content that match the element name and the element content indicated by the search information, and among the child elements of the first parent element, A second child element other than the first parent element that acquires a second child element other than the first child element and has a child element that includes a child element name and child element contents that match the second child element from an unstructured database. The process of searching the parent element and outputting the child element of the first parent element or the second parent element is executed.

Also, in one aspect, a search device is provided. This search device has a storage unit and a search unit. The storage unit stores an unstructured database that stores a plurality of parent elements having a plurality of child elements including a child element name and a child element content. The search unit acquires the search information indicating the element name and the element content, and from the unstructured database, the first child including the child element name and the child element content that match the element name and the element content indicated by the search information. The first parent element having an element is searched, the second child element other than the first child element among the child elements possessed by the first parent element is acquired, and the child matching the second child element is obtained from the unstructured database. A second parent element other than the first parent element having a child element including an element name and a child element content is searched, and the child element of the first parent element or the second parent element is output.

Also, in one aspect, a search method is provided. In this search method, a computer acquires search information indicating an element name and an element content, and an unstructured database in which a plurality of parent elements having a plurality of child elements including a child element name and a child element content are stored. Searches for the first parent element having the first child element including the child element name and the child element content matching the element name and the element content indicated by the search information, and among the child elements of the first parent element, A second child element other than the first parent element that acquires a second child element other than the first child element and has a child element that includes the child element name and the child element content that match the second child element from the unstructured database. The parent element is searched, and the child element of the first parent element or the second parent element is output.

On one side, searches using unstructured databases can be accelerated.

It is a figure which shows the search apparatus of 1st Embodiment. It is a figure which shows the information processing system of the 2nd Embodiment. It is a figure which shows the hardware example of a server. It is a figure which shows the functional example of a server. It is a figure (the 1) which shows the specific example of the creation process of a non-structured document. It is a figure (the 2) which shows the specific example of the creation process of a non-structured document. It is a figure which shows the specific example of the unstructured database. It is a figure which shows the example of the input definition information. It is a figure which shows the example of the document relation definition information. It is a flowchart which shows the example of the creation process of a non-structured document. It is a flowchart (the 1) which shows the example of the definition information creation process. It is a flowchart (2) which shows an example of the definition information creation process. It is a flowchart which shows the example of the search process. It is a figure which shows the specific example of a search process. It is a figure which shows the specific example of a search result.

Hereinafter, the present embodiment will be described with reference to the drawings.
[First Embodiment]
FIG. 1 is a diagram showing a search device according to the first embodiment. The search device 1 has a storage unit 1a and a search unit 1b. The storage unit 1a may be a volatile storage device such as a RAM (Random Access Memory) or a non-volatile storage device such as an HDD (Hard Disk Drive) or a flash memory. The search unit 1b is, for example, a processor. The processor may include a CPU (Central Processing Unit), a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), and the like. Further, the search unit 1b may be a multiprocessor.

The storage unit 1a stores the unstructured database 1aa. The unstructured database 1aa stores a plurality of parent elements having a plurality of child elements including the child element name and the child element contents. For example, the unstructured database 1aa stores the parent elements 1a1,1a2. The parent element 1a1 has a child element including the child element name "customer name" and the child element content "X1". Further, the parent element 1a1 has a child element including the child element name "order number" and the child element content "Y1". The parent element 1a2 has a child element including the child element name "order number" and the child element content "Y1". Further, the parent element 1a2 has a child element including the child element name "product ID (identifier)" and the child element content "Z1".

Further, the unstructured database 1aa stores a plurality of parent elements in a state where the normalization level is high. That is, the redundancy of the unstructured database 1aa is small. The unstructured database 1aa may be created by the user using the search device 1 or another device, or may be created by the search device 1 based on the relational database.

The search unit 1b acquires the search information 2 (step S1). For example, the search unit 1b acquires the search information 2 directly input by the user to the search device 1. Alternatively, the search unit 1b acquires the search information 2 transmitted from another device that can be connected to the search device 1 via the network. The search information 2 indicates an element name and an element content. For example, the search information 2 indicates an element name “customer name” and an element content “X1”.

The search unit 1b searches the unstructured database 1aa for the first parent element having the first child element including the child element name and the child element content that match the element name and the element content indicated by the search information 2. For example, the search unit 1b searches for a parent element having a child element including a child element name and a child element content that match the element name “customer name” indicated by the search information 2 and the element content “X1”. Then, the search unit 1b identifies the parent element 1a1 (step S2).

The search unit 1b acquires the second child element other than the first child element among the child elements of the first parent element. For example, the search unit 1b acquires child elements other than the child elements including the child element name “customer name” and the child element content “X1” among the child elements possessed by the parent element 1a1. As a result, the search unit 1b acquires the child element including the child element name “order number” and the child element content “Y1” (step S3).

The search unit 1b searches the unstructured database 1aa for a second parent element other than the first parent element, which has a child element including a child element name and a child element content that match the second child element. For example, the search unit 1b searches for a parent element having a child element including an element name and an element content that matches the child element including the child element name “order number” and the child element content “Y1”. Further, the search unit 1b excludes the parent element 1a1 from the search target. Then, the search unit 1b identifies the parent element 1a2 (step S4).

The search unit 1b outputs the child element of the first parent element or the second parent element (step S5). For example, the search unit 1b outputs a search result 3 including a child element of the parent element 1a1 or the parent element 1a2. The output destination may be a display included in the search device 1 or another device that has transmitted the search information 2. The search unit 1b may output the child elements of the first parent element and the second parent element.

Here, the unstructured database can perform difficult searches on the relational database, but contains redundant information due to the low level of normalization. Therefore, the unstructured database has a larger amount of data than the relational database, and the increase in the amount of data hinders the shortening of the search time.

According to the first embodiment, the unstructured database 1aa stores a plurality of parent elements with a high normalization level. Therefore, the unstructured database 1aa has less redundancy than the unstructured database having a low normalization level. That is, the unstructured database 1aa has a smaller amount of data than the unstructured database having a low normalization level. The search device 1 can speed up the search using the unstructured database by searching using the unstructured database 1aa having a small amount of data.

Further, the search unit 1b may create an unstructured database 1aa having the same normalization level as the relational database based on the relational database as described above. For example, the search unit 1b generates a parent element corresponding to a record in a table included in a relational database, creates a child element corresponding to a value set in the record, and corresponds the created child element to the record. Include in parent element. As a result, the search device 1 can hold the unstructured database 1aa in a state where the normalization level is high.

The parent element may include the parent element name. For example, the parent element name may be the title name of the parent element (for example, customer information or shipping information in FIG. 1), or may be a table name when the parent element is created based on the relational database. The storage unit 1a may store the first definition in which the parent element name and the child element name are associated with each other. For example, when searching for the first parent element, the search unit 1b refers to the first definition and specifies the parent element name corresponding to the child element name that matches the element name indicated by the search information 2. The search unit 1b searches for the first parent element from the parent elements having the parent element name that matches the specified parent element name. In this way, the parent element name is added to the search condition for the unstructured database 1aa. The search device 1 can obtain accurate search results by increasing the number of search conditions for the unstructured database 1aa.

The storage unit 1a may store a second definition in which the parent element names of a plurality of parent elements, each of which has a child element having the same child element name, are associated with each other. For example, when the search unit 1b searches for the second parent element, the search unit 1b refers to the second definition, has a relationship with the parent element name of the first parent element, and is a child element that matches the element name indicated by the second child element. Identify the parent element name of the parent element that has the name. The search unit 1b searches for the second parent element from the parent elements having the parent element name that matches the specified parent element name. In this way, the search device 1 can specify the second parent element by having the second definition. Further, the search device 1 refers to the second definition, and has a relationship with the parent element name of the first parent element, and the parent element name of the parent element having a child element name that matches the element name indicated by the second child element. If cannot be specified, the search for the parent element based on the element name indicated by the second child element is not executed. As a result, the search device 1 does not have to perform an extra search process.

The search unit 1b may create the first definition and the second definition based on the parent element name and the child element included in the plurality of parent elements. Since the search device 1 automatically creates the first definition and the second definition, the user can save the trouble of creating the first definition and the second definition.

[Second Embodiment]
Next, the second embodiment will be described. In the second embodiment, an unstructured database is created based on the relational database stored in the server, and the unstructured database can be used for searching.

FIG. 2 is a diagram showing an information processing system according to the second embodiment. The information processing system of the second embodiment includes a server 100 and a terminal device 200. The server 100 and the terminal device 200 are connected via the network 300. The network 300 may be a LAN (Local Area Network) or a wide area network such as a WAN (Wide Area Network) or the Internet.

The server 100 stores a relational database. Server 100 creates an unstructured database based on the relational database. When the server 100 receives the search request from the terminal device 200, the server 100 performs a search using the unstructured database. The server 100 transmits the search result to the terminal device 200.

The server 100 may perform a search using an unstructured database in response to a search request directly input to the server 100 by the user.
The terminal device 200 is a client computer used by the user. The terminal device 200 can display the search result acquired from the server 100 on its own device.

Next, the hardware of the server 100 will be described.
FIG. 3 is a diagram showing an example of server hardware. The server 100 includes a processor 101, a RAM 102, an HDD 103, an image signal processing unit 104, an input signal processing unit 105, a reading device 106, and a communication interface 107. Each unit is connected to the bus of server 100.

The processor 101 controls the entire server 100. The processor 101 is, for example, a CPU, DSP, ASIC, FPGA, or the like. Further, the processor 101 may be a multiprocessor including a plurality of processing elements. Further, the processor 101 may be a combination of two or more elements such as a CPU, a DSP, an ASIC, and an FPGA.

The RAM 102 is the main storage device of the server 100. The RAM 102 temporarily stores at least a part of an OS (Operating System) program or an application program to be executed by the processor 101. Further, the RAM 102 stores various data used for processing by the processor 101.

The HDD 103 is an auxiliary storage device of the server 100. The HDD 103 magnetically writes and reads data from the built-in magnetic disk. The OS program, application program, and various data are stored in the HDD 103. The server 100 may include other types of auxiliary storage devices such as a flash memory and an SSD (Solid State Drive), or may include a plurality of auxiliary storage devices.

The image signal processing unit 104 outputs an image to the display 11 connected to the server 100 in accordance with an instruction from the processor 101. As the display 11, various displays such as a CRT (Cathode Ray Tube) display, a liquid crystal display (LCD: Liquid Crystal Display), and an organic EL (Electro-Luminescence) display can be used.

The input signal processing unit 105 acquires an input signal from the input device 12 connected to the server 100 and outputs the input signal to the processor 101. As the input device 12, various input devices such as a pointing device such as a mouse and a touch panel and a keyboard can be used. A plurality of types of input devices may be connected to the server 100.

The reading device 106 is a device that reads programs and data recorded on the recording medium 13. As the recording medium 13, for example, a magnetic disk such as a flexible disk (FD) or HDD, an optical disk such as a CD (Compact Disc) or a DVD (Digital Versatile Disc), or a magneto-optical disk (MO) can be used. Can be used. Further, as the recording medium 13, for example, a non-volatile semiconductor memory such as a flash memory card can be used. The reading device 106 stores, for example, a program or data read from the recording medium 13 in the RAM 102 or the HDD 103 in accordance with an instruction from the processor 101.

The communication interface 107 communicates with the terminal device 200 via the network 300. The communication interface 107 may be a wired communication interface or a wireless communication interface.

The terminal device 200 can also be realized by the same hardware as the server 100.
Next, the function of the server 100 will be described.
FIG. 4 is a diagram showing a functional example of the server. The server 100 has a storage unit 110, an unstructured processing unit 120, a definition information creation unit 130, a registration unit 140, and a search unit 150.

The storage unit 110 is implemented as, for example, a storage area secured in the RAM 102 or the HDD 103. The storage unit 110 stores the relational database group 111, the unstructured database 112, the input definition information 113, and the document relational definition information 114.

The relational database group 111 includes a plurality of relational databases. The unstructured database 112 is a database created based on the relational database group 111. The input definition information 113 and the document-related definition information 114 will be described in detail later.

The unstructured processing unit 120, the definition information creation unit 130, the registration unit 140, and the search unit 150 are implemented as, for example, modules of a program executed by the processor 101.
The unstructured processing unit 120 creates an unstructured document based on the relational database group 111. The definition information creation unit 130 creates the input definition information 113 and the document-related definition information 114 based on the non-structured document. The registration unit 140 registers the unstructured document in the unstructured database 112. When the search unit 150 receives the search request from the terminal device 200, the search unit 150 searches using the unstructured database 112.

Next, a process in which the unstructured processing unit 120 creates an unstructured document based on the relational database group 111 will be described with reference to a specific example.
FIG. 5 is a diagram (No. 1) showing a specific example of the process of creating a non-structured document. FIG. 5 shows that the customer information 111a and the shipping information 111b are included in the relational database group 111.

The customer information 111a includes items of a customer ID, a customer name, and an order number. The customer ID item indicates the customer's identifier. The customer name item indicates the customer's name. The order number item indicates the order number identifier.

The shipping information 111b includes items of an order number and a product ID. The order number item indicates the order number identifier. The product ID item indicates the product identifier.
The unstructured processing unit 120 creates the unstructured documents 400a and 400b based on each record of the customer information 111a without changing the normalization level. For example, the unstructured processing unit 120 creates the unstructured document 400a based on the records of the customer ID “C001”, the customer name “A company”, and the order number “ORDER001” of the customer information 111a. The non-structural documents 400a and 400b are created in XML (Extensible Markup Language) format. The unstructured processing unit 120 adds the table name “customer information” of the customer information 111a to the start tag and the end tag of the unstructured documents 400a and 400b.

The unstructured processing unit 120 creates the unstructured documents 401a to 401e based on each record of the shipping information 111b without changing the normalization level. The non-structural documents 401a to 401e are created in XML format. The unstructured processing unit 120 adds the table name “shipping information” of the shipping information 111b to the start tag and the end tag of the unstructured documents 401a to 401e.

FIG. 6 is a diagram (No. 2) showing a specific example of the process of creating a non-structured document. FIG. 6 shows that the product information 111c and the inquiry history 111d are included in the relational database group 111.

Product information 111c includes items of product ID and product name. The product ID item indicates the product identifier. The item of product name indicates the name of the product.
The inquiry history 111d includes items such as an inquiry ID, a customer ID, an order number, a product ID, and an inquiry content. The item of the inquiry ID indicates the identifier of the inquiry. The customer ID item indicates the customer's identifier. The order number item indicates the order number identifier. The product ID item indicates the product identifier. Inquiry content items indicate questions about the product, product failures, and so on.

The unstructured processing unit 120 creates the unstructured documents 402a to 402e based on each record of the product information 111c without changing the normalization level. The non-structural documents 402a to 402e are created in XML format. The unstructured processing unit 120 adds the table name “product information” of the product information 111c to the start tag and the end tag of the unstructured documents 402a to 402e.

The unstructured processing unit 120 creates the unstructured documents 403a to 403e based on each record of the inquiry history 111d without changing the normalization level. The non-structural documents 403a to 403e are created in XML format. The unstructured processing unit 120 adds the table name “inquiry history” of the inquiry history 111d to the start tag and the end tag of the unstructured documents 403a to 403e.

Next, the unstructured database 112 will be described with reference to specific examples.
FIG. 7 is a diagram showing a specific example of an unstructured database. The registration unit 140 registers the unstructured document created by the unstructured processing unit 120 in the unstructured database 112. For example, the unstructured database 112 has information (inside the frame 112a) registered in the unstructured document 400a.

Each of the frames 112a to 112e shown in the unstructured database 112 of FIG. 7 is hereinafter referred to as a document. The document is an example of the parent element of the first embodiment. The document contains items and item contents. The item is an example of the child element name of the first embodiment. The item content is an example of the child element content of the first embodiment. FIG. 7 shows that the unstructured database 112 has 5 or more documents. The frames 112a to 112e shown on the unstructured database 112 are not included in the unstructured database 112 because they are for explaining one document.

In addition, the upper tag in the document is hereinafter referred to as the document name. For example, the document name of the document in the frame 112a is customer information.
The unstructured database 112 does not have to be ordered. For example, the document showing the customer information in the frame 112a and the document showing the customer information in the frame 112c are not continuously registered in the unstructured database 112.

Here, the unstructured processing unit 120 creates an unstructured document at the same normalization level as the relational database. The registration unit 140 registers an unstructured document having the same normalization level as the relational database in the unstructured database 112. Therefore, the unstructured database 112 has the same level of normalization as the relational database. As a result, the amount of data in the unstructured database 112 is smaller than the amount of data in the unstructured database with a low normalization level.

Next, the input definition information 113 and the document-related definition information 114 will be described.
FIG. 8 is a diagram showing an example of input definition information. The input definition information 113 includes an input item and a search target document name / item. The input items correspond to the input items included in the search request acquired by the server 100 from the terminal device 200. The search target document name / item indicates that the search target document name and the item included in the document with the document name are the search target items.

For example, the input definition information 113 indicates an input item “customer ID”, a search target document name “customer information” / item “customer ID”, and a search target document name “inquiry history” / item “customer ID”. This means that when the input item included in the search request acquired from the terminal device 200 by the server 100 is the "customer ID", the "customer ID" of the "customer information" and the "customer ID" of the "inquiry history" are searched. Indicates that it should be an item.

FIG. 9 is a diagram showing an example of document-related definition information. The document relationship definition information 114 includes a search target document name / item and a search target document name (related document name) / item. The search target document name / item indicates the document name searched and the item included in the document with the document name. The search target document name (related document name) / item indicates a document name related to the search target document name of the search target document name / item and an item included in the document of the document name.

For example, the document-related definition information 114 indicates a search target document name “customer information” / item “customer ID” and a search target document name (related document name) “inquiry history” / item “customer ID”. This indicates that the document name related to "customer information" is "inquiry history". It also indicates that the "customer ID" of the "customer information" and the "customer ID" of the "inquiry history" have a relationship.

Next, the process executed by the server 100 will be described with reference to a flowchart.
FIG. 10 is a flowchart showing an example of a non-structured document creation process. For example, the unstructured processing unit 120 executes the processing of FIG. 10 by receiving an instruction for creating an unstructured document from the terminal device 200. Hereinafter, the process shown in FIG. 10 will be described along with the step numbers.

(S11) The unstructured processing unit 120 selects one table from the relational database group. For example, the unstructured processing unit 120 selects customer information 111a from the relational database group.

(S12) The unstructured processing unit 120 selects one record from the table. For example, the unstructured processing unit 120 selects a record including the customer ID “C001”, the customer name “A company”, and the order number “ORDER001” of the customer information 111a.

(S13) The unstructured processing unit 120 creates an unstructured document in which record information is converted into XML format.
(S14) The unstructured processing unit 120 adds the start tag of the table name and the end tag of the table name to the unstructured document.

(S15) The unstructured processing unit 120 stores an XML-format unstructured document in the storage unit 110.
(S16) The unstructured processing unit 120 determines whether or not all the records have been selected. When all the records are selected, the unstructured processing unit 120 advances the processing to step S17. If not all records have been selected, the unstructured processing unit 120 advances the processing to step S12.

(S17) The unstructured processing unit 120 determines whether or not all the tables have been selected. When all the tables are selected, the unstructured processing unit 120 ends the processing. If all the tables are not selected, the unstructured processing unit 120 advances the processing to step S11.

FIG. 11 is a flowchart (No. 1) showing an example of the definition information creation process. The definition information creation unit 130 executes the process of FIG. 11 before the registration unit 140 registers the unstructured document in the unstructured database 112. Hereinafter, the process shown in FIG. 11 will be described along with the step numbers.

(S21) The definition information creation unit 130 reads a non-structured document.
(S22) The definition information creation unit 130 selects one item from the non-structured document. For example, the definition information creation unit 130 selects the “customer ID” from the non-structural document 400a. The definition information creation unit 130 excludes the document name when selecting an item from the non-structured document. For example, the definition information creation unit 130 excludes the document name “customer information” in the non-structural document 400a.

(S23) In the definition information creation unit 130, the selected item exists in the input item of the input definition information 113, and the document name of the non-structural document read in step S21 exists in the search target document of the input definition information 113. Judge whether or not. If the determination condition is satisfied, the definition information creation unit 130 advances the process to step S31. If the determination condition is not satisfied, the definition information creation unit 130 advances the process to step S24.

(S24) The definition information creation unit 130 determines whether or not the selected item exists in the input item of the input definition information 113. If the selected item exists in the input item of the input definition information 113, the definition information creation unit 130 advances the process to step S25. If the selected item does not exist in the input item of the input definition information 113, the definition information creation unit 130 proceeds to the process in step S26.

(S25) The definition information creation unit 130 adds an item to the end of the line of the same item. For example, when the unstructured document read in step S21 is the unstructured document 403a and the item selected in step S22 is the "customer ID", the definition information creation unit 130 determines the "customer ID, customer information / customer" of the input definition information 113. Add "Inquiry history / Customer ID" after "ID".

The definition information creation unit 130 advances the process to step S31.
(S26) The definition information creation unit 130 registers an input item and a search target document name / item in the input definition information 113. For example, when the unstructured document read in step S21 is the unstructured document 400a and the item selected in step S22 is the "customer name", the definition information creation unit 130 adds "customer name, customer information / customer" to the input definition information 113. Register the "name".

The definition information creation unit 130 advances the process to step S31.
FIG. 12 is a flowchart (No. 2) showing an example of the definition information creation process. Hereinafter, the process shown in FIG. 12 will be described along with the step numbers.

(S31) The definition information creation unit 130 determines whether or not the selected item and the document name of the non-structural document read in step S21 exist in the document-related definition information 114. If the determination condition is satisfied, the definition information creation unit 130 advances the process to step S35. If the determination condition is not satisfied, the definition information creation unit 130 advances the process to step S32.

(S32) The definition information creation unit 130 determines whether or not the item of the non-structural document saved in the save area provided in the storage unit 110 and the item selected in step S22 are the same. In the same case, the definition information creation unit 130 advances the process to step S33. If they are different, the definition information creation unit 130 advances the process to step S34.

(S33) The definition information creation unit 130 associates the items of the non-structured document saved in the save area with the items selected in step S22, and registers them in the document-related definition information 114. For example, it is assumed that the item of the non-structured document saved in the save area is the "customer ID" of the "customer information". When the unstructured document read in step S21 is the unstructured document 403a and the item selected in step S22 is the "customer ID", the definition information creation unit 130 documents "customer information / customer ID = inquiry history / customer ID". Register in the relationship definition information 114.

The definition information creation unit 130 advances the process to step S35.
(S34) The definition information creation unit 130 stores the document name of the unstructured document read in step S21 and the item selected in step S22 in association with each other in the save area. Then, the definition information creation unit 130 advances the process to step S35.

(S35) The definition information creation unit 130 determines whether or not all the items have been selected. When all the items are selected, the definition information creation unit 130 ends the process. If all the items are not selected, the definition information creation unit 130 advances the process to step S22.

In this way, the definition information creation unit 130 automatically creates the input definition information 113 and the document relationship definition information 114 based on the non-structural document, so that the user creates the input definition information 113 and the document relationship definition information 114. You can save the trouble of doing it.

The definition information creation unit 130 executes the processes shown in FIGS. 11 and 12 each time the unstructured processing unit 120 creates an unstructured document. The registration unit 140 registers the unstructured document for which the definition information creation process by the definition information creation unit 130 has been completed in the unstructured database 112.

FIG. 13 is a flowchart showing an example of the search process. When the search unit 150 receives the search request from the terminal device 200, the search unit 150 executes the process shown in FIG. Hereinafter, the process shown in FIG. 13 will be described along with the step numbers.

(S41) The search unit 150 receives a search request from the terminal device 200. The search request includes input items and input contents.
(S42) The search unit 150 refers to the input definition information 113 and specifies the search target document name corresponding to the input item.

(S43) The search unit 150 refers to the unstructured database 112 and searches for a document in which the search target document name, the input item, and the input content match the document name, the item, and the item content.
(S44) The search unit 150 identifies a document satisfying the conditions in the search in step S43. The search unit 150 acquires the information contained in the specified document.

(S45) The search unit 150 refers to the document-related definition information 114, excludes the search target document name searched once, and the search target document name (related document name) related to the previous search target document name remains. Determine if it is. If it remains, the search unit 150 proceeds to step S46. If not, the search unit 150 proceeds to step S51.

(S46) The search unit 150 refers to the document-related definition information 114 and specifies the search target document name (related document name) related to the previous search target document name.
(S47) The search unit 150 identifies other items other than the item searched in step S43 and the item contents corresponding to the other items among the information acquired in step S44. When the step S47 is executed two or more times, the search unit 150 identifies other items other than the item searched in step S49 and the item contents corresponding to the other items among the information acquired in step S50.

(S48) The search unit 150 refers to the document-related definition information 114, and among the search target document names (related document names) specified in step S46, the related document name / item that matches the other items specified in step S47. To identify.

(S49) The search unit 150 refers to the unstructured database 112, the search target document name (related document name), the item specified in step S48, the item content specified in step S47 corresponding to the item, and the document name. , Search for documents that match the item and item content.

(S50) The search unit 150 identifies a document satisfying the conditions in the search in step S49. The search unit 150 acquires the information contained in the specified document. Then, the search unit 150 advances the process to step S45.

(S51) The search unit 150 creates a search result by combining the information of the documents acquired in steps S44 and 50. The search unit 150 outputs the created search result to the terminal device 200.
Next, the search process will be described with reference to specific examples.

FIG. 14 is a diagram showing a specific example of the search process. The search unit 150 receives the search request 500 from the terminal device 200 (step ST1). The search request 500 indicates the input item "customer name" and the input content "company A".

The search unit 150 refers to the input definition information 113 and specifies the search target document name “customer information” corresponding to the input item “customer name” (step ST2). The search unit 150 refers to the unstructured database 112, and a document in which the document name, the item, and the item content match the search target document name “customer information”, the input item “customer name”, and the input content “Company A”. To search for. The search unit 150 identifies a document satisfying the conditions (step ST3). The search unit 150 acquires the information contained in the specified document. The information is customer information, customer ID "C001", customer name "Company A", and order number "ORDER001".

The search unit 150 refers to the document-related definition information 114 and specifies the search target document name (related document name) “shipping information” and “inquiry history” related to the search target document name “customer information” (step ST4). ).

The search unit 150 identifies other items other than the searched item "customer name" and the item contents corresponding to the other items in the document including the document name "customer information". That is, the search unit 150 specifies the customer ID "C001" and the order number "ORDER001".

The search unit 150 refers to the document-related definition information 114 and identifies the inquiry history / customer ID from "customer information / customer ID = inquiry history / customer ID". The search unit 150 refers to the document-related definition information 114 and specifies the shipping information / order number from "customer information / order number = shipping information / order number". The search unit 150 refers to the document-related definition information 114 and specifies the inquiry history / order number from "customer information / order number = inquiry history / order number".

The search unit 150 refers to the unstructured database 112, and finds the search target document name (related document name) "inquiry history", the item "customer ID", the item content "C001", and the document name, item, and item content. Search for matching documents. The search unit 150 refers to the unstructured database 112, and finds the document name (related document name) "shipping information", the item "order number", the item content "ORDER001", and the document name, item, and item content. Search for matching documents. Further, the search unit 150 refers to the unstructured database 112, and refers to the search target document name (related document name) "inquiry history", the item "order number" and the item content "ORDER001", and the document name, the item and the item content. You may search for documents that match. Further, the search unit 150 simultaneously executes a search for a document including "shipping information" and the like and a document including "inquiry history" and the like.

The search unit 150 acquires document information including shipping information, an order number “ORDER001”, and a product ID “PRD001” (step ST5). Further, the search unit 150 acquires document information including the inquiry history, the inquiry ID "QA001", the customer ID "C001", the order number "ORDER001", the product ID "PRD001", and the inquiry content "CT1" (step ST5). ).

The search unit 150 refers to the document-related definition information 114 and specifies the search target document name “shipping information” or the related document name “product information” related to the search target document name “inquiry history” (step ST6). ..

The search unit 150 identifies other items other than the searched item "order number" and the item contents corresponding to the other items in the document including the document name "shipping information". That is, the search unit 150 identifies the product ID "PRD001". The search unit 150 identifies other items other than the searched items "customer ID" and "order number" and the item contents corresponding to the other items in the document including the document name "inquiry history". That is, the search unit 150 specifies the inquiry ID "QA001", the product ID "PRD001", and the inquiry content "CT1".

The search unit 150 refers to the document-related definition information 114 and identifies the product information / product ID from "shipping information / product ID = product information / product ID" or "inquiry history / product ID = product information / product ID". .. The search unit 150 excludes the items "inquiry ID" and "inquiry content" that are not related to the search target document name (related document name) "product information".

The search unit 150 refers to the unstructured database 112 and searches for a document in which the related document name “product information”, item “product ID” and item content “PRD001” match the document name, item and item content. .. As a result, the search unit 150 acquires the information of the document including the product information, the product ID "PRD001", and the product name "Product A" (step ST7).

The search unit 150 excludes the search target document name that has been searched once, and determines that the search target document name (related document name) related to the search target document name “product information” does not remain. The search unit 150 creates a search result by combining the document information acquired in steps ST3, ST5, and ST7. The search unit 150 outputs the created search result to the terminal device 200.

FIG. 15 is a diagram showing a specific example of the search result. FIG. 15 is a search result 600 output by the search process of FIG. The search result 600 is displayed on the terminal device 200.
According to the second embodiment, the unstructured database 112 includes a plurality of documents having the same normalization level as the tables included in the relational database group 111. Therefore, the unstructured database 112 has less redundancy than the unstructured database having a low normalization level. That is, the unstructured database 112 has a smaller amount of data than the unstructured database having a low normalization level. The server 100 can speed up the search using the unstructured database by searching using the unstructured database 112 having a small amount of data.

The server 100 refers to the input definition information 113 and searches for a document in which the search target document name, the input item, and the input content match the document name, the item, and the item content. In this way, the server 100 can obtain accurate search results by increasing the number of search conditions for the unstructured database 112.

Further, when a plurality of search target document names corresponding to the input items included in the search request are registered in the input definition information 113, the server 100 simultaneously searches for the document names having the search target document names. For example, when the search target document names "customer information" and "inquiry history" corresponding to the input item "customer ID" included in the search request are registered in the input definition information 113, the server 100 uses the document name "customer information". The document having "" and the document having the document name "inquiry history" are searched at the same time. In this way, the server 100 can shorten the search time by simultaneously searching for the document name having the search target document name.

Further, the server 100 refers to the document-related definition information 114 and identifies an item to be searched and an item not to be searched. For example, after step ST6 in FIG. 14, the server 100 identifies the "product ID" as a search target item. Further, the server 100 excludes the items "inquiry ID" and "inquiry content" that are not related to the search target document name (related document name) "product information" from the search target. As a result, the server 100 does not have to search the unstructured database 112 with the "query ID" and "query content" as search target items. That is, since the server 100 has the document-related definition information 114, it is not necessary to perform an extra search process.

Further, the document-related definition information 114 is information in which documents having different document names are associated with a document having at least one item in common and a document other than the document. That is, documents having different document names and common items have a relationship with each other. As a result, the server 100 can output the contents set for each document having a different document name. For example, the server 100 can output the contents set in each of the different document names "customer information", "shipping information", "product information", and "inquiry history".

Further, the information processing of the first embodiment can be realized by causing the processor used in the search device 1 to execute the program. The information processing of the second embodiment can be realized by causing the processor 101 to execute the program. The program can be recorded on a computer-readable recording medium.

For example, the recording medium on which the program is recorded can be distributed by distributing the program. Further, the programs that realize the functions corresponding to the unstructured processing unit 120, the definition information creation unit 130, the registration unit 140, and the search unit 150 can be distributed separately as separate programs. The functions of the unstructured processing unit 120, the definition information creation unit 130, the registration unit 140, and the search unit 150 may be realized by separate computers. For example, the computer may store the program recorded on the recording medium in the RAM 102 and the HDD 103, read the program, and execute the program.

1 Search device 1a Storage unit 1aa Unstructured database 1a1,1a2 Parent element 1b Search unit 2 Search information 3 Search results S1, S2, S3, S4, S5 Step

Claims (7)

On the computer
It corresponds to a record in any of the multiple tables included in the relational database, and has a plurality of child elements whose child element name is the name of the item included in the record and whose child element content is the value of the item. , The parent element having the name of the table as the parent element name acquires the search information indicating the element name and the element content in the unstructured database in which a plurality of stored elements are stored.
Refer to the first definition in which the parent element name of each of the parent elements having the child element of the child element name is listed in association with the child element name, and the element name that matches the element name indicated by the search information. Identify the parent element name of the parent element having the child element of the child element name,
Wherein among the plurality of the parent element of the unstructured database, from among the parent element of the same the parent element name and the parent element name identified, the said element contents and the element name the search information indicating Search for the first parent element having the first child element including the matching child element name and the child element content, and search for the first parent element.
Among the child elements possessed by the first parent element, a second child element other than the first child element is acquired.
From the unstructured database, a second parent element other than the first parent element having the child element including the child element name and the child element content matching the second child element is searched for.
Outputs the child element of the first parent element or the second parent element.
A search program that executes processing. On the computer,
On the basis of the relational database, to execute processing for creating the unstructured database with the same normalized level and the relational database, the search program according to claim 1. In the generation of the unstructured database, the parent element corresponding to the record in the table included in the relational database is generated, the child element corresponding to the value set in the record is created, and the created child is created. The search program according to claim 2, wherein the element is included in the parent element corresponding to the record. In the search for the second parent element,
Refer to the second definition relating the parent element names of the plurality of parent elements each having the child element of the same child element name.
The parent element name of the parent element having a relationship with the parent element name of the first parent element and having the child element name matching the element name indicated by the second child element is specified.
The parent element name from the parent element with, searches the second parent element search program according to claim 1 that match specified the parent element name. The search program according to claim 4 , wherein the first definition and the second definition are created based on the parent element name and the child element included in the plurality of parent elements. It corresponds to a record in any of the multiple tables included in the relational database, and has a plurality of child elements whose child element name is the name of the item included in the record and whose child element content is the value of the item. , A storage unit that stores a plurality of unstructured databases in which a plurality of parent elements having the name of the table as a parent element name are stored.
Refer to the first definition in which the search information indicating the element name and the element content is acquired, and the parent element name of each of the parent elements having the child element of the child element name is listed in association with the child element name. Then, the parent element name of the parent element having the child element of the child element name matching the element name indicated by the search information is specified, and among the plurality of parent elements in the unstructured database. Among the parent elements having the same parent element name as the specified parent element name, the child element name and the child element content matching the element name and the element content indicated by the search information are included. The first parent element having one child element is searched, the second child element other than the first child element among the child elements possessed by the first parent element is acquired, and the first child element is obtained from the unstructured database. A second parent element other than the first parent element having the child element including the child element name and the child element contents matching the two child elements is searched, and the first parent element or the second parent element is searched. A search unit that outputs the child elements of
Search device to have. The computer
It corresponds to a record in any of the multiple tables included in the relational database, and has a plurality of child elements whose child element name is the name of the item included in the record and whose child element content is the value of the item. , The parent element having the name of the table as the parent element name acquires the search information indicating the element name and the element content in the unstructured database in which a plurality of stored elements are stored.
Refer to the first definition in which the parent element name of each of the parent elements having the child element of the child element name is listed in association with the child element name, and the element name that matches the element name indicated by the search information. Identify the parent element name of the parent element having the child element of the child element name,
Wherein among the plurality of the parent element of the unstructured database, from among the parent element of the same the parent element name and the parent element name identified, the said element contents and the element name the search information indicating Search for the first parent element having the first child element including the matching child element name and the child element content, and search for the first parent element.
Among the child elements possessed by the first parent element, a second child element other than the first child element is acquired.
From the unstructured database, a second parent element other than the first parent element having the child element including the child element name and the child element content matching the second child element is searched for.
Outputs the child element of the first parent element or the second parent element.
retrieval method.

Images