Production assistance device, production assistance method, and recording medium

Abstract

The present invention provides a production assistance device, a production assistance method, and a recording medium, with the aim of improving the production efficiency of scenarios. The production assistance device can access a database that stores a set of relational knowledge data composed of two nodes of specified knowledge and edges connecting the two nodes to define the relationship between the two nodes; obtain an inference route that constitutes a hypothesis and assigns an order to multiple pieces of knowledge; update the scenario when a second node corresponding to the first node in the inference route is added to the scenario that concretizes the hypothesis; generate a first search query for searching for a second connection target node in the direction from the second node based on the first node in the inference route, the first connection target node in the direction from the first node, and the first edge connecting the first node and the first connection target node; search for specific first relational knowledge data that matches the first search query in the database; and output the specific first relational knowledge data.

Description

Technical Field

[0001] The present invention relates to a production assistance device, a production assistance method, and a recording medium storing a production assistance program for producing auxiliary data. Background Art

[0002] Patent Document 1 discloses a scenario generation device that collects elements that are the basis for generating a social scenario useful for people to make a balanced and appropriate decision. The scenario generation device includes: a causal relationship phrase pair DB that stores causal relationship phrase pairs; a synonym relationship generation unit that retrieves causal relationship phrase pairs having a causal consistency with the result phrase for each causal relationship phrase pair and generates connection information of the causal relationship phrase pairs; a connection relationship DB that stores the connection information; and a causal relationship connection unit that uses the connection information to connect causal relationships by linking causal relationship phrase pairs having a causal consistency with the result phrase of the causal relationship phrase pair to the cause phrase.

[0003] Prior Art Documents

[0004] Patent Documents

[0005] Patent Document 1: Japanese Unexamined Patent Application Publication No. 2015-121897 Summary of the Invention

[0006] Problems to be Solved by the Invention

[0007] However, the scenario generation device of the above Patent Document 1 cannot specify the relationship between knowledge, the end point of the scenario, and the intermediate conditions. Therefore, there is a problem that more scenarios other than the desired scenarios are generated. In addition, not only social scenarios for the purpose of decision-making as described above, but also the generation of cross-branched scenarios such as hypothetical scenarios or cause scenarios of abnormal condition phenomena in technological development is the same.

[0008] An object of the present invention is to improve the production efficiency of scenarios.

[0009] Means for Solving the Problems

[0010] As a production assistance device for a technical solution of the invention disclosed in the present application, the production assistance device having a processor that executes a program and a storage device that stores the above program, is characterized in that it can access a database that stores a set of relational knowledge data, the relational knowledge data being composed of two nodes of prescribed knowledge and an edge that prescribes the relationship between the two nodes and connects the two nodes; the processor executes: an acquisition process of acquiring an inference route in which an order is assigned to a plurality of the above knowledge, the inference route constituting a hypothesis; an update process of updating the scenario when a second node corresponding to the first node in the inference route acquired by the acquisition process is added to the scenario that concretizes the hypothesis; a generation process of generating a first search query for searching for a second connection target node in the direction from the second node, based on the first node in the inference route, a first connection target node in the direction from the first node, and a first edge that connects the first node and the first connection target node; a search process of searching for specific first relational knowledge data that matches the first search query generated by the generation process from the database; and an output process of outputting the specific first relational knowledge data retrieved by the search process.

[0011] Advantages of the Invention

[0012] According to a representative embodiment of the present invention, it is possible to improve the production efficiency of scenarios. Other problems, structures, and effects will become clearer through the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is an explanatory diagram showing an example of the system configuration of a production assistance system.

[0014] Figure 2 It is a block diagram showing an example of the hardware configuration of a computer.

[0015] Figure 3 It is an explanatory diagram showing an example of scenario generation using a production assistance device.

[0016] Figure 4 It shows Figure 1 An explanatory diagram of an example of the relational knowledge DB shown.

[0017] Figure 5 It is an explanatory diagram showing an example of the structure of a search query.

[0018] Figure 6 It is a flowchart showing an example of the order of scenario production assistance processing performed by the production assistance device.

[0019] Figure 7 It shows Figure 6Flowchart of a detailed processing sequence example of the retrieval process (step S605) shown

[0020] Figure 8 represents Figure 7 Explanatory diagram of a phrase extraction example shown in steps S706 - S710

[0021] Figure 9 Explanatory diagram of scenario creation example 1 performed by user operation

[0022] Figure 10 Explanatory diagram of scenario creation example 2 performed by user operation

[0023] Figure 11 Explanatory diagram of scenario creation example 3 performed by user operation

[0024] Figure 12 Explanatory diagram of scenario creation example 4 performed by user operation

[0025] Figure 13 Explanatory diagram of scenario creation example 5 performed by user operation

[0026] Figure 14 Explanatory diagram of scenario creation example 6 performed by user operation

[0027] Figure 15 Explanatory diagram of scenario creation example 7 performed by user operation

[0028] Figure 16 Explanatory diagram of scenario creation example 8 performed by user operation

[0029] Figure 17 Explanatory diagram of scenario creation example 9 performed by user operation

[0030] Figure 18 Explanatory diagram of scenario creation example 10 performed by user operation

[0031] Figure 19 Explanatory diagram of scenario creation example 11 performed by user operation

[0032] Figure 20 Explanatory diagram of scenario creation example 12 performed by user operation

[0033] Reference numeral description

[0034] 100 Production assistance system

[0035] 101 Server

[0036] 102 Terminal

[0037] 103 Network

[0038] 104 Relationship Knowledge DB

[0039] 104-i Relationship Knowledge Data

[0040] 201 Processor

[0041] 202 Storage Device

[0042] 301 Deduction Route

[0043] 302 Scenario

[0044] 310 Deduction Route

[0045] 410 Source Sentence

[0046] 510 Retrieval Query

[0047] 810 Statistical Data

[0048] 2000 Registered Object Scenario

[0049] E, e Edge

[0050] N, n Node

[0051] C, c Connection Point Detailed Implementation Manner

[0052] Hereinafter, the production assistance for the scenario of this embodiment will be described. In addition, the phrase is a string of consecutive words that represents a general meaning, but in this embodiment, even a single word is treated as a phrase.

[0053] <Example Structure of Production Assistance System>

[0054] Figure 1 It is an explanatory diagram showing an example of the system structure of the production assistance system. The production assistance system 100 includes a server 101 and a terminal 102. The server 101 and the terminal 102 are communicably connected via a network 103 such as the Internet, a LAN (Local Area Network), or a WAN (Wide Area Network).

[0055] Server 101 is a computer that assists in the generation of scenarios. Server 101 has a relational knowledge database (DB) 104. The relational knowledge DB 104 is a database that contains information that has been relationally extracted from a group of documents in advance and, in addition, has been schema-transformed from existing knowledge DB groups. It may also include a link as access information to existing knowledge DBs such as the Unified Medical Language System (UMLS). The relational knowledge DB 104 may also be in a computer that can communicate with Server 101 via network 103. Terminal 102 is a computer that inputs and outputs data to and from Server 101. Specifically, for example, Terminal 102 remotely inputs data to Server 101 or displays data from Server 101 through user operations. Additionally, in Figure 1 the client-server type production assistance system 100 has been described, but it may also be a stand-alone type.

[0056] <Hardware Structure of Computer>

[0057] Figure 2 is a block diagram showing an example of the hardware structure of a computer. Computer 200 has a processor 201, a storage device 202, an input device 203, an output device 204, and a communication interface (communication IF) 205. The processor 201, the storage device 202, the input device 203, the output device 204, and the communication IF 205 are connected by a bus 206. The processor 201 controls the computer 200. The storage device 202 is the working area for the processor 201. In addition, the storage device 202 is a non-temporary or temporary recording medium that stores various programs and data. Examples of the storage device 202 include a ROM (Read Only Memory), a RAM (Random Access Memory), an HDD (Hard Disk Drive), and a flash memory. The input device 203 inputs data. Examples of the input device 203 include a keyboard, a mouse, a touch panel, a numeric keypad, and a scanner. The output device 204 outputs data. Examples of the output device 204 include a display, a printer, and a speaker. The communication IF 205 is connected to the network 103 and transmits and receives data.

[0058] <Example of Scenario Production Assistance>

[0059] Figure 3It is an explanatory diagram showing an example of scenario generation using the server 101. The server 101 enables a user to set up an inference route 301 and use the inference route 301 to generate a scenario 302. The inference route 301 is a hypothesis, that is, the order of establishment of unknown information predicted based on known information. The scenario 302 is new knowledge that connects multiple pieces of knowledge obtained from multiple data sources according to the hypothesis, that is, the inference route 301.

[0060] The inference route 301 has a node group consisting of two or more nodes and an edge group consisting of one or more edges that connect between the nodes. In Figure 3 the example, the node group consists of nodes N1 to N4 (simply referred to as node N when not distinguished), and the edge group consists of edges E1 to E3 (simply referred to as edge E when not distinguished). The circular or triangular figures at both ends of the edge E are connection points C12, C21, C31, C32, C41 (simply referred to as connection point C when not distinguished) that are connected to the node N. The edge E1 connects between nodes N1 and N2 with connection points C12 and C21, the edge E2 connects between nodes N2 and N3 with connection points C22 and C31, and the edge E3 connects between nodes N3 and N4 with connection points C32 and C41.

[0061] The node N contains a string through user input. The string is defined as a concept class or a phrase. A string enclosed in curly braces {} is a concept class, and a string not enclosed in curly braces {} is a phrase. The concept class represents the upper concept of the phrase. The phrase is a specific example of the concept class. In addition, "*" indicates that neither a concept class nor a phrase is specified.

[0062] For the edge E, relationship data representing the relationship between the nodes N at both ends is specified. For example, "mutual" is specified for the edge E1, "evaluation index" is specified for the edge E2, and "increase or decrease" is specified for the edge E3. The shape of the connection point C determines the direction of the edge E. A connection point C (for example, connection point C31) where one vertex of the triangle is inside the node N determines the direction from this edge E (in this case, edge E2) towards this node N (in this case, node N3). A connection point C where two vertices of the triangle are inside the node N determines the direction from this node N towards this edge E. The circular connection point C does not determine a direction.

[0063] Similar to the inference route 301, the scenario 302 also has a node group consisting of two or more nodes and an edge group consisting of one or more edges that connect the nodes N. However, to distinguish it from the inference route 301, in the scenario 302, the nodes are denoted as n (n1, n2, n31, n32, n41, n42, n43), the edges as e (e11, e21, e22, e312, e312, e313), and the connection points as c (c12, c21, c22, c311, c312, c411, c412, c413). The nodes n, edges e, and connection points c are copied from the inference route 301 or the retrieval result of a retrieval query (described later in Figure 5 ).

[0064] Specifically, for example, the nodes n within the scenario 302 are copied from the nodes N of the inference route 301 through a user operation. In this case, a correspondence is established between the source node N for copying and the target node n for copying. In addition, as described later, the nodes n within the scenario 302 can be copied from the retrieval result through a user operation. In the case of copying from the retrieval result, a correspondence is established between the target node n for copying and the node N within the inference route 301 corresponding to that node n. Within the scenario 302, through a user operation, the nodes n can be connected by the edges e. A correspondence is established between the edge e and the edge E within the inference route 301 corresponding to that edge e.

[0065] <Relational Knowledge DB104>

[0066] Figure 4 is an explanatory diagram showing an example of the relational knowledge DB104 as shown in Figure 1 . The relational knowledge DB104 stores, for example, n (n is an integer greater than or equal to the above) relational knowledge data 104-1 to K104-n. The relational knowledge data 104-i (i is an integer satisfying 1 ≤ i ≤ n) is composed of two nodes ns and nd connected to both ends of one edge esd. It may also include a source sentence 410.

[0067] The edge esd has relational data 401 (in this example, it means an evaluation index). The nodes ns and nd each have a concept class 402 and a phrase 403. In addition, the nodes ns and nd each have a connection point c at both of their ends. Here, the connection points c at both ends of the edge e are respectively set as connection points cs and cd.

[0068] The source sentence 410 is a string of an article within the source that forms the basis of the relational knowledge data 104-i. The phrase 403 within the nodes ns and nd and the relational data 401 of the edge esd are included in the source sentence 410.

[0069] <Structure of Retrieval Query>

[0070] Figure 5 This is an explanatory diagram showing a construction example of a retrieval query. The retrieval query 510 is generated by the server 101 or the terminal 102 on the occasion of a user operation using the input device 203 of the terminal 102. In Figure 5 this example, it is assumed that the terminal 102 has generated the state of the node n2 as the scenario 302 from the node N2 of the inference route 301 by means of drag-and-drop (D&D) performed by the user.

[0071] If the user clicks on the connection point c22 of the node n2 with the cursor 500, the server 101 or the terminal 102 generates the retrieval query 510. The retrieval query 510 has a key 511, a data source 512, and a value 513. The key 511 is an item used to retrieve the relational knowledge DB 104. The data source is the location where the key 511 exists, and the value is the item value represented by the key 511.

[0072] The key 511 includes, for example, the type of edge, the phrase of the starting point, the phrase of the connection target, the concept class of the starting point, the concept class of the connection target, the phrase of the edge, the phrase within the source sentence, the phrase of the source title, the type of source, the document ID of the source, and the context.

[0073] The type of edge is an item that determines the type of the edge e connected to the connection point c within the inference route 301 corresponding to the clicked connection point. In Figure 5 this example, the clicked connection point c is the connection point c22 within the scenario 302, and the connection point C within the inference route 301 corresponding to the connection point c22 is the connection point C22. The data source 512 is the inference route 301 connected to the connection point C22 within the inference route 301 corresponding to the connection point c22. In Figure 5 this example, the inference route 301 where the edge E2 exists is the data source 512. The value 513 of the type of edge is the relational data 401 defined by the edge E within the inference route 301 connected to the connection point C corresponding to the clicked connection point c. In Figure 5 this example, "evaluation index" is the value 513 of the type of edge.

[0074] The phrase of the starting point is a phrase within the node n that is the starting point. The starting point is an item that determines the node n having the clicked connection point c. In Figure 5 this example, since the clicked connection point c is the connection point c22, the node n2 is the starting point. Since the starting point exists within the scenario 302, the data source 512 of the phrase of the starting point is also the scenario 302. The value 513 of the phrase of the starting point is a string representing the phrase of the starting point. In Figure 5 this example, the value 513 of the phrase of the starting point is "UGT1A1" representing the phrase within the node n2 that is the starting point.

[0075] The phrase of the connection target is an item that determines the phrase within the node of the connection target. The connection target described above is a node N within the inference route 301 connected via an edge E from a connection point C within the inference route 301 corresponding to the clicked connection point c. In Figure 5 the example, since the clicked connection point c is the connection point c22, the node N3 within the inference route 301 connected via the edge E2 from the connection point C22 within the inference route 301 corresponding to the connection point c22 is the connection target. Since the connection target exists within the inference route 301, the data source 512 of the phrase of the connection target is also the inference route 301. The value 513 of the phrase of the connection target is a string representing the phrase of the connection target. In Figure 5 the example, the value 513 of the phrase of the connection target is "*" (no phrase is specified. It can also be a blank) representing the phrase within the node N3 that is the connection target.

[0076] The concept class of the starting point is an item that determines the concept class within the node n that is the starting point. In Figure 5 the example, as described above, the node n2 is the starting point. Since the starting point exists within the scenario 302, the data source 512 of the concept class of the starting point is also the scenario 302. The value 513 of the concept class of the starting point is a string enclosed in curly braces {} representing the concept class of the starting point. In Figure 5 the example, since there is no string enclosed in curly braces {} in the node n2, the value 513 of the concept class of the starting point is "*" (no concept class is specified. It can also be a blank).

[0077] The concept class of the connection target is an item that determines the concept class within the node N that is the connection target. In Figure 5 the example, as described above, the node N3 is the connection target. Since the connection target exists within the inference route 301, the data source 512 of the concept class of the connection target is also the inference route 301. The value 513 of the concept class of the connection target is a string enclosed in curly braces {} representing the concept class of the connection target. In Figure 5 the example, since there is no string enclosed in curly braces {} in the node N3, the value 513 of the concept class of the connection target is "*" (no concept class is specified. It can also be a blank).

[0078] The phrase of the edge is an item that determines the phrase assigned to the edge E within the inference route 301 connected to the connection point C corresponding to the clicked connection point c. In Figure 5 the example, the phrase of the edge represents the phrase of the edge E2. The data source 512 of the phrase of the edge is a reduced box that displays the search results of the search query 510 (to be described later in Figure 13 etc.). The value 513 of the phrase of the edge is the phrase displayed in the reduced box that is the data source 512 of the phrase of the edge. InFigure 5 In the example, the value 513 of the edge phrase is "*" (the phrase is not specified).

[0079] The phrase within the source sentence is an item that determines the phrase existing within the source sentence 410 selected from the source. The data source 512 of the phrase within the source sentence is the same reduced frame as the data source 512 of the edge phrase. The value 513 of the phrase within the source sentence is the phrase displayed in the reduced frame that is the data source 512 of the phrase within the source sentence. For example, when the phrase "generation" is included in the source sentence 410, the value 513 of the phrase within the source sentence is "generation".

[0080] The phrase of the title of the source is an item that determines the phrase representing the title of the source. The data source 512 of the phrase of the title of the source is the same reduced frame as the data source 512 of the edge phrase. The value 513 of the phrase of the title of the source is the phrase displayed in the reduced frame that is the data source 512 of the phrase of the title of the source. For example, when the title of the source is the phrase "UGT1A1", the value 513 of the phrase within the source sentence is "UGT1A1".

[0081] The phrase of the type of the source is an item that determines the phrase representing the type of the source. The type of the source described is the classification by which the source is classified and the attribution target of the source. The data source 512 of the phrase of the type of the source is the same reduced frame as the data source 512 of the edge phrase. The value 513 of the phrase of the type of the source is the phrase displayed in the reduced frame that is the data source 512 of the phrase of the type of the source. For example, when the type of the source is the phrase "PubMed", the value 513 of the phrase within the source sentence is "PubMed".

[0082] The document ID of the source is an item that determines the document ID of the source. The document ID of the source described is the identification information that uniquely determines the document of the source. The data source 512 of the document ID of the source is the same reduced frame as the data source 512 of the edge phrase. The value 513 of the document ID of the source is the string representing the document ID displayed in the reduced frame that is the data source 512 of the phrase of the type of the source. For example, when the document ID of the source is the phrase "xxx - 12", the value 513 of the phrase within the source sentence is "xxx - 12".

[0083] The context is the text of the article, that is, the continuous state of the meaning of the words in the text, and is specified in order to obtain the search results associated with the current scenario 302. The data source 512 of the context is the scenario. For example, in the scenario 302, the search result of the node n1 connected to the connection point c21 on the opposite side of the connection point c22 of the clicked node n2 is the data source 512 of the context. In addition, in the source with the document ID of the source specified by the edge e1 as "Jia-Long+2004", the source sentence 410 containing the phrase 403 representing the node n1, "Carci no gen Detoxification Phe no type" and the phrase 403 representing the node n2, "UGT1A1" is the context value 513.

[0084] In addition, the server 101 may use, for example, cosine similarity and Doc2Vec to calculate the similarity between the context value 513 and the source sentence 410 included in the search result of the search query 510. In this case, the server 101 may be set in advance by a setting instruction from the user to adopt the search result of the search query 510 if the similarity is above a predetermined threshold, or to adopt the search result of the search query 510 if the similarity is below the predetermined threshold.

[0085] <Scene Production Auxiliary Processing Order>

[0086] Figure 6 1 is a flowchart showing an example of a scenario creation support process sequence performed by the server 101. The server 101 waits for a screen request from the terminal 102 (step S601: No). If the screen request is accepted (step S601: Yes), the server 101 transmits screen data to the terminal 102 (step S602). In addition, if there is a scenario 302 generated by the user of the terminal 102 and saved (step S610), the scenario 302 is read from the storage device 202 using the user's ID as the key 511, and the screen data including the scenario 302 is transmitted.

[0087] Next, the server 101 waits for receiving the inference route 301 (step S603: No). When receiving the inference route 301 (step S603: Yes), the inference route 301 is associated with the ID of the user of the terminal 102 and stored in the storage device 202 (step S604). The inference route 301 is used to create the search query 510.

[0088] Next, the server 101 waits for the reception of the save button 920 (step S605: No). When the save button is not pressed (step S605: No), the server 101 determines whether there is an edit of the scenario 302 (step S606). Specifically, for example, when a node n is added or deleted in the scenario 302, or an edge e is added or deleted, or a string in the node n is added, changed, or deleted, due to the operation of the terminal 102, the server 101 determines that there is an edit of the scenario 302.

[0089] When there is no edit of the scenario 302 (step S606: No), it returns to step S605. On the other hand, when there is an edit of the scenario 302 (step S606: Yes), the server 101 updates the scenario 302 with the edited content (step S607). For example, if the node n in the scenario 302 is copied from the node N of the inference route 301 by a user operation, the server 101 establishes a correspondence between the source node N of the copy and the target node n of the copy. In addition, if the node n in the scenario 302 is copied from the search result by a user operation, the server 101 establishes a correspondence between the target node n of the copy and the node N in the inference route 301 corresponding to the node n. In addition, if the nodes n are connected by the edge e by a user operation in the scenario 302, the server 101 establishes a correspondence between the edge e and the edge E in the inference route 301 corresponding to the edge e.

[0090] And, the server 101 waits for an instruction to generate the search query 510 from the terminal 102 (step S608: No). The instruction to generate the search query 510 includes information on the correspondence between the node n and the node N, the correspondence between the edge e and the edge E, and the clicked connection point c in the latest scenario 302 after the update (step S607).

[0091] When an instruction to generate the search query 510 is received (step S608: Yes), the server 101 Figure 5 generates the search query 510 as shown, and performs a search process using the generated search query 510 (step S610). In addition, the generation of the search query 510 (step S610) may also be executed by the terminal 102.

[0092] Further, the server 101 sends the retrieval result of the retrieval process (step S608) to the terminal 102 (step S611), and returns to step S605. In step S605, when it is detected that the save button 920 is pressed (step S605: Yes), the server 101 takes the scenario composed of the node n selected by the user operation and the edge e connecting the nodes in scenario 302 as the registration target scenario (if none is selected, it is scenario 302 itself), associates it with the ID of the user of this terminal 102, and saves it to the storage device 202 (step S612), thus ending a series of processes.

[0093] <Retrieval process (step S608)>

[0094] Figure 7 represents Figure 6 The flowchart shown is an example of the detailed processing order of the retrieval process (step S608). First, the server 101 attempts to perform a retrieval in the relational knowledge DB 104 using the received retrieval query 510 (hereinafter referred to as the first retrieval query) (step S701). Next, the server 101 determines whether there is relational knowledge data 104-i in the relational knowledge DB 104 that matches the first retrieval query 510 (step S702). When there is relational knowledge data 104-i that matches the first retrieval query 510 (step S702: Yes), the server 101 obtains the relational knowledge data 104-i that matches the first retrieval query 510 (step S703).

[0095] Further, the server 101 determines whether there is unselected acquired relational knowledge data 104-i in the group of relational knowledge data (acquired relational knowledge data) obtained in step S703 (step S704). When there is unselected acquired relational knowledge data 104-i (step S704: Yes), the server 101 selects one unselected acquired relational knowledge data 104-i (step S705).

[0096] Further, the server 101 extracts the execution phrases 403 of the connection target of the acquired relational knowledge data 104-i (step S706). Specifically, for example, the server 101 extracts one or more phrases 403 from the phrases 403 within the node N that is the connection target of the acquired relational knowledge data 104-i. The extracted phrases 403 are different phrases 403 from each other, but there may be duplicate words.

[0097] Next, the server 101 generates a retrieval query (hereinafter referred to as the second retrieval query) 510 starting from the node N including the link target of the extraction phrase 403 for each extraction phrase 403 (step S707). Specifically, for example, similar to the first retrieval query 510, the server 101 generates the second retrieval query 510 starting from the node N including the link target of the extraction phrase 403 for each extraction phrase 403.

[0098] Moreover, similar to step S701, the server 101 attempts to perform a retrieval in the relational knowledge DB 104 with the second retrieval query 510 (step S708). The server 101 obtains the number of pieces of relational knowledge data 104-i that match the second retrieval query 510 for each extraction phrase 403 (step S709). And the server 101 calculates statistical data for the number of pieces of relational knowledge data 104-i that match the second retrieval query 510 for each extraction phrase 403 (step S710), and returns it to step S704.

[0099] The statistical data described above is statistical data related to the number of pieces of relational knowledge data 104-i for each extraction phrase 403. For example, it can be a combination of the maximum value and the minimum value of each number of pieces, or it can be the average value or the median of each number of pieces. In addition, it can also include a combination of the maximum value and the minimum value of each number of pieces and the average value or the median value of each number of pieces.

[0100] In step S704, when there is no unselected relational knowledge data 104-i to be obtained (step S704: No), it returns to step S702. In step S702, when there is no relational knowledge data 104-i that matches the first retrieval query 510 (step S702: No), the retrieval process (step S608) ends and it transfers to step S609.

[0101] <Example of phrase extraction>

[0102] Figure 8 It is an explanatory diagram showing an example of phrase extraction shown in steps S706 to S710 that Figure 7 is. In Figure 8 it, the phrase 403 in the node n that is the extraction object of the phrase in the relational knowledge data 104-i is set to "detecting the elimination of bilirubin substrate or the generation of bilirubin glucuronides.". The node n that is the extraction object is the link target nd when viewed from the first retrieval query 510, and is the starting point ns when viewed from the second retrieval query 510. In Figure 8In this case, before the execution of phrase extraction (step S706), the node n to be extracted is expressed as a connection target nd, and after the execution of phrase extraction (step S706), it is expressed as a starting point ns.

[0103] The extracted phrases 801 to 804 are partial phrases extracted from phrase 403. The extracted phrase 805 is the phrase extracted from phrase 403 as it is. The number of extracted phrases is at least 1 piece and is determined by a prior setting. In addition, the extraction method can be an existing grammatical extraction method such as obtaining a noun phrase or splitting with a noun phrase, or a method of extracting a string that matches a part of phrase 403 from an external dictionary accessible to the server 101, or a method of extraction using machine learning. In the machine learning, by inputting phrase 403, or phrase 403 and the scenario 302 being produced, the phrase output from the learning model is used as the extracted phrase.

[0104] In Figure 8 the example, the number of pieces of relationship knowledge data that match the extracted phrase 801 is 6, the extracted phrase 802 is 18, the extracted phrase 803 is 7, the extracted phrase 804 is 21, and the extracted phrase 805 is 3.

[0105] In addition, in Figure 8 the example, the statistical data 810 includes the average number of pieces "11" of the number of pieces of the extracted phrases 801 to 805, and the number range "3 to 21" composed of the minimum number of pieces 3 and the maximum number of pieces 21.

[0106] <Scenario production example>

[0107] Next, Figures 9 to 20 an example of scenario production by user operation will be described. In addition, even if there is only one node n in the scenario production area 902 described later, it is the scenario 302 even if it is not saved to the server 101 (step S610).

[0108] Figure 9 is an explanatory diagram showing an example of scenario production example 1 by user operation. Figure 9 shows an example of the display screen 900 of the terminal 102 before the production of the inference route 301 and the scenario 302, where the screen data has been sent to the terminal 102 through step S602. The display screen 900 has an inference route production area 901, a scenario production area 902, and a search result display area 903. In the inference route production area 901, the inference route 301 is displayed. In the inference route production area 901, the node N1 and the connection point C12 are pre-displayed. However, for the node N1, neither a concept class nor a phrase is specified.

[0109] The scenario 302 can be displayed in the scenario production area 902. In Figure 9 , since the scenario 302 has not been produced yet, the scenario 302 is not displayed. Below the scenario production area 902, a save button 920 is displayed. If the save button 920 is pressed, the latest scenario 302 displayed in the scenario production area 902 is sent from the terminal 102 to the server 101.

[0110] The search result display area 903 is a reduced box with a node label 931 and a note label 932 (refer to Figure 5 ). In the node label 931, the node candidates NC1 to NC6 required for the production of the scenario 302 are displayed as available for scenario production, or the search results of the first search query 510 are displayed as being available for the production of the scenario 302. In the note label 932, notes associated with the search results of the first search query 510 are displayed. In Figure 9 , it indicates a state where the node label 931 is selected and the node candidates NC1 to NC6 are displayed as available. The node candidates NC1 to NC6 can be copied to the inference route production area 901 by drag and drop.

[0111] Figure 10 It is an explanatory diagram showing an example of scenario production 2 by user operation. In Figure 10 , it indicates a state where the node candidate NC is copied to the inference route production area 901. In Figure 10 , as a result of the user dragging and dropping the node candidate NC1 to the inference route production area 901, the edge E1 and the node N2 are copied and connected to the node N1. Next, as a result of the user dragging and dropping the node candidate NC2 to the inference route production area 901, the edge E2 and the node N3 are copied and connected to the node N2. Next, as a result of the user dragging and dropping the node candidate NC4 to the inference route production area 901, the edge E3 and the node N4 are copied and connected to the node N3.

[0112] Figure 11 It is an explanatory diagram showing an example of scenario production 3 by user operation. Figure 11 Indicates Figure 10 From the state of Figure 11 , a state where the concept class 402 or the phrase 403 as a condition is input to the nodes N1 to N4 in the inference route production area 901 through user operation. In

[0113] Figure 12 It is an explanatory diagram showing an example of scenario production 4 by user operation. Figure 12 Indicates fromFigure 11 Starting from the state where the node N2 of the inference route 301 is copied as the node n2 to the scenario creation area by drag and drop D&D. The server 101 maintains the correspondence between the node n2 and the node N2. In addition, the connection points c21 and c22 of the node n2 are the connection points c copied from the connection points C21 and C22 of the node N2. The connection points c21 and c22 can be pressed by a user operation.

[0114] In addition, after the completion of the inference route 301, the node label 931 is switched from the display of node candidates to the display of search results by a user operation. The node label 931 has a starting point condition input field 1201, a connection target condition input field 1202, a source sentence condition input field 1203, and a metadata condition input field 1204. For the starting point condition input field 1201, a string representing a concept class or phrase as a condition for retrieving the starting point is input. For the connection target condition input field 1202, a string representing a concept class 402 or phrase 403 as a condition for retrieving the connection target is input. For the source sentence condition input field 1203, a string as a condition for retrieving the source sentence 410 is input. For the metadata condition input field 1204, a string as a condition for retrieving the metadata included in the relational knowledge data 104-i is input.

[0115] In addition, in Figure 12 , for example, if the user presses the connection point c22, as shown in Figure 5 , the terminal 102 generates a first search query 510 and sends it to the server 101. As a result of the sending, the terminal 102 receives search results from the server 101 and displays a search result group 1210 on the node label 931. In addition, in the case of pressing the connection point c, the concept class 402 or phrase 403 (in this example, "UGT1A1" representing the phrase 403) as a condition for the starting point including the connection point c is automatically set in the starting point condition input field 1201.

[0116] The search result group 1210 includes zero or more search results. In Figure 12 , as an example, the search result group 1210 includes three search results 1211 to 1213. The search results 1211 to 1213 are relational knowledge data 104-i including the concept class 402 or phrase 403 as a condition for the starting point of the first search query 510. In addition, the relational knowledge data 104-i having a starting point with a phrase 403 that is the same or partially the same as the starting point of a certain search result 1211 to 1213 is called the partially matching relational knowledge data 104-i. For the search results 1211 to 1213 where the partially matching relational knowledge data 104-i exists, a plus button 1214 is displayed. The partially matching relational knowledge data 104-i is folded into the plus button 1214.

[0117] Statistical data 810 is displayed in search results 1211 to 1213. In addition, source sentence 410 is included in search results 1211 and 1213. In source sentence 410, for example, the value 513 of the conditions (starting concept class 402 or phrase 403, connecting target concept class 402 or phrase 403, type of edge) specified by the first search query 510 is highlighted. In addition, as shown in search result 1212, instead of source sentence 410, a link 1215 can be displayed as access information for accessing a knowledge DB such as the Unified Medical Language System (UMLS).

[0118] Figure 13 It is an explanatory diagram showing an example 5 of scenario creation by user operation. Figure 13 Indicates from Figure 12 The state where the plus button 1214 of search results 1211 and 1213 is pressed from the state. By pressing the plus button 1214 of search results 1211 and 1213, search results 1221 and 1223 showing the relationship knowledge data 104 - i with partial consistency are displayed.

[0119] Figure 14 It is an explanatory diagram showing an example 6 of scenario creation by user operation. Through user operation, "bilirubin AND {indicator}" is input in the connecting target condition input field 1202, which is used to search for the condition of the connecting target where the phrase is "bilirubin" and the concept class 402 is "{indicator}". In addition, "calculate NOT pcr" is input in the source sentence condition input field 1203, which is used to search for source sentence 410 that contains "calculate" but does not contain "pcr". By obtaining these conditions from the terminal 102, the server 101 can further reduce the search result group 1210.

[0120] In addition, the terminal 102 can also display a box 1400 near the node (for example, below) in the scenario creation area 902. In the box, through user operation, a string such as a synonym of the concept class 402 or phrase 403 of node n2 can be freely input. In addition, a synonym of the concept class 402 or phrase 403 of node n2 can be automatically input in the box. Specifically, for example, the server 101 retrieves a synonym of the concept class 402 or phrase 403 of node n2 from the relationship knowledge DB 104 or an external knowledge DB and sends it back to the terminal 102 as the result of sending the first search query 510. Thereby, the terminal 102 can display the retrieved synonym of the concept class 402 or phrase 403 of node n2 in the box.

[0121] Figure 15It is an explanatory diagram showing Scenario Production Example 7 performed by a user operation. Figure 15 Indicates the state starting from Figure 14 where, by drag and drop, the search result 1211 is copied into the scenario production area 902 and connected to the node n2. Specifically, for example, the terminal 102, through a user operation, draws an edge e21 between the node n2 and the copied node n31 to connect the node n2 and the node n31. Thereby, the server 101 establishes a correspondence between the node n31 that is the connection target of the node n2 and the node N3 that is the connection target of the node N2 of the inference route 310, and establishes a correspondence between the edge e21 and the edge E2 between the nodes N2 and N3. The dragged-and-dropped search result 1211 is displayed, for example, lighter than the other search results 1212 and 1213 so that it can be recognized that it has been dragged and dropped.

[0122] In addition, in the scenario production area 902, [QI+2015] displayed below the edge e21 is the document ID 1500 of the source of the source sentence of the search result 1211. In the document ID 1500, a link to the document of the source can also be implanted.

[0123] Figure 16 It is an explanatory diagram showing Scenario Production Example 8 performed by a user operation. Figure 16 Indicates the state starting from Figure 15 where, through a user operation, the note label 932 is selected. By selecting the note label 932, the document 1600 of the source determined by the document ID restricted within the scenario production area 902 is displayed. In the document 1600 of the source, for example, the value 513 of the condition (starting concept class or phrase, connecting target concept class or phrase, edge type) specified by the first search query 510 is highlighted.

[0124] Figure 17 It is an explanatory diagram showing Scenario Production Example 9 performed by a user operation. Figure 17 Indicates the state where the user edits the scenario 302 starting from Figure 16 the state. Specifically, for example, the phrase 403 within the node n31 is corrected. In this way, the conditions within the node n displayed in the scenario production area 902 can be edited by a user operation.

[0125] Figure 18 It is an explanatory diagram showing Scenario Production Example 10 performed by a user operation. Figure 18 Indicates the state starting from Figure 17From the state where..., the state where the search result 1213 is copied to the scenario creation area 902 by drag-and-drop and linked to the node n2. Specifically, for example, the terminal 102, through a user operation, draws an edge e22 between the node n2 and the copied node n32 to link the node n2 and the node n32. Thereby, the server 101 establishes a correspondence between the node n32 that is the connection target of the node n2 and the node N3 that is the connection target of the node N2 of the inference route 310, and establishes a correspondence between the edge e22 and the edge E2 between the nodes N2 and N3. The dragged-and-dropped search result 1213 is displayed, for example, lighter than other search results 1212 so that it can be recognized that it has been dragged and dropped.

[0126] In addition, in the scenario creation area 902, [Sara + 2010] displayed below the edge e21 is the document ID 1800 of the source of the source sentence of the search result 1213. In the document ID 1800, a link to the document of the source can also be implanted. In this way, multiple search results can be connected to one connection point c22.

[0127] Figure 19 It is an explanatory diagram showing an example 11 of scenario creation by a user operation. Figure 19 Indicates from Figure 18 the state where..., the state where the search result 1901 is copied to the scenario creation area 902 by drag-and-drop and linked to the node n2. In Figure 19 , through a user operation, "diabetes" is input in the metadata condition input field 1204. By obtaining the conditions specified by the start condition input fields 1201 to the metadata condition input field 1204 from the terminal 102, the server 101 reduces the search result group 1210 again and sends the search result group 1900 to the terminal 102. Thereby, the search result group 1900 is displayed in the node label 931 of the terminal 102.

[0128] The user drags and drops the search result 1901 from the search result group 1900 and copies it to the scenario creation area 902. Specifically, for example, the terminal 102, through a user operation, draws an edge e1 between the node n2 and the copied node n1 to link the node n2 and the node n1. Thereby, the server 101 establishes a correspondence between the node n1 that is the connection target of the node n2 and the node N1 that is the connection target of the node N2 of the inference route 310, and establishes a correspondence between the edge e1 and the edge E1 between the nodes N2 and N1. In this way, the scenario 302 is updated to the nodes n1, e1, n2, e21, n31, e22, n32.

[0129] Figure 20 It is an explanatory diagram showing an example 12 of scenario creation by a user operation. Figure 20 Indicates from Figure 19Starting from the state, nodes n41 to n43 and edges e321 to e323 are added to scenario 302. Specifically, for example, through a user operation on terminal 102, an edge e311 is drawn between node n31 and the copied node n41 to connect node n31 and node n41. Thereby, server 101 establishes a correspondence between node n41, which is the connection target of node n31, and node N4, which is the connection target of node N3 of inference route 310, and establishes a correspondence between edge e311 and edge E3 between nodes N3 and N4.

[0130] Similarly, through a user operation on terminal 102, an edge e312 is drawn between node n32 and the copied node n42 to connect node n32 and node n42. Thereby, server 101 establishes a correspondence between node n42, which is the connection target of node n32, and node N4, which is the connection target of node N3 of inference route 310, and establishes a correspondence between edge e312 and edge E3 between nodes N3 and N4.

[0131] Similarly, through a user operation on terminal 102, an edge e313 is drawn between node n32 and the copied node n43 to connect node n32 and node n43. Thereby, server 101 establishes a correspondence between node n43, which is the connection target of node n32, and node N4, which is the connection target of node N3 of inference route 310, and establishes a correspondence between edge e313 and edge E3 between nodes N3 and N4.

[0132] In addition, the user can select scenario 2000 to be registered. Specifically, for example, the user selects nodes n1, n2, n31, n32, and n42 from scenario 302 through input device 203 ( Figure 20 shown in black). Server 101 sets the selected nodes n1, n2, n31, n32, and n42, and the edges e1, e22, and e312 between them, to the registration target scenario 2000. And if the user presses the save button 920 through input device 203, terminal 102 associates the registration target scenario 2000 with the ID of the user of this terminal 102 and saves it to storage device 202 (step S610).

[0133] In this way, in the above embodiment, it is possible to improve the production efficiency of scenario 302 and the quality of the produced scenario 302.

[0134] In addition, in the above-described embodiment, in the client-server type production assistance system 100, it has been described that the server 101 is a production assistance device that assists in the production of the scenario 302 in accordance with operations from the terminal 102. In contrast, the terminal 102 may be a production assistance device that generates a retrieval query 510 and causes the server to perform a retrieval. Further, in the above-described embodiment, the client-server type production assistance system 100 has been described, but it may also be implemented by a stand-alone server 101.

[0135] In addition, the production assistance devices related to the above-described Embodiment 1 and Embodiment 2 may be configured as follows in (1) to (13).

[0136] (1) A production assistance device (server 101, terminal 102) having a processor 201 that executes a program and a storage device 202 that stores the program can access a relation knowledge DB 104 that stores a set of relation knowledge data 104-i, the relation knowledge data 104-i being composed of two nodes ns and nd that define knowledge, and an edge esd that defines the relationship between the two nodes ns and nd and connects the two nodes ns and nd. The processor 201 executes: an acquisition process (steps S603, S604) for acquiring an inference route 301 in which an order is given to a plurality of pieces of knowledge, the inference route 301 constituting a hypothesis; an update process (step S607) for updating the scenario 302 that concretizes the hypothesis when a second node n2 corresponding to the first node N2 in the inference route 301 acquired by the acquisition process is added to the scenario 302; a generation process (step S609) for generating a first retrieval query 510 for retrieving a second connection target node n31 from the second node n2 based on the first node N2 in the inference route 301, the first connection target node N3 facing from the first node N2, and the first edge E2 that connects the first node N2 and the first connection target node N3; a retrieval process (step S610) for retrieving a specific first relation knowledge data 104-i that matches the first retrieval query 510 generated by the generation process from the relation knowledge DB 104; and an output process (step S611) for outputting the specific first relation knowledge data 104-i retrieved by the retrieval process.

[0137] Accordingly, the user can connect nodes n representing a plurality of pieces of knowledge with edges e, and can create a scenario 302. Specifically, for example, since a connection target node n is selected from specific first relationship knowledge data 104-i listed as the retrieval result of the first retrieval query 510, a scenario 302 that is not interesting to the user is not generated. That is, by retrieving specific first relationship knowledge data 104-i with the first retrieval query 510, a scenario 302 is created that satisfies the conditions imposed on each node n or each edge e of the scenario 302. In this way, it is possible to improve the production efficiency of the scenario 302 and the quality of the created scenario 302.

[0138] (2) In the production assistance device in the above (1), in the generation process, the processor 201 generates a second retrieval query 510 for retrieving a fourth connection target node n41 facing from the second connection target node n31 based on the first connection target node N3 in the inference route 301, the third connection target node N4 facing from the first connection target node N3, and the third edge E3 connecting the first connection target node N3 and the third connection target node N4; in the retrieval process, the processor 201 retrieves specific second relationship knowledge data 104-i that conforms to the second retrieval query 510 generated by the generation process from the relationship knowledge DB104; in the output process, the processor 201 outputs the number of pieces of the specific first relationship knowledge data 104-i and the specific second relationship knowledge data 104-i.

[0139] Accordingly, the user can confirm the existence of the fourth connection target node n41 when retrieving the second connection target node n31. Therefore, it is possible to avoid adding the second connection target node n31 for which the fourth connection target node n41 does not exist to the scenario 302.

[0140] (3) In the production assistance device in the above (2), the processor 201 executes an extraction process (step S706) of extracting a plurality of partial strings each including a part of the string from the string of knowledge defined in the second connection target node n3; in the generation process, the processor 201 uses each string in the strings extracted by the extraction process as the second connection target node n3 to generate the second retrieval query 510; in the retrieval process, the processor 201 retrieves specific second relationship knowledge data 104-i from the relationship knowledge DB104 for each second retrieval query 510; the processor 201 executes a calculation process (step S710) of calculating statistical data 810 regarding the number of pieces of the specific second relationship knowledge data 104-i retrieved from each second retrieval query 510; in the output process, the processor 201 outputs the specific first relationship knowledge data 104-i and the statistical data 810 calculated by the calculation process.

[0141] Through the extraction process (step S706), it is possible to comprehensively improve the retrieval pattern of the second retrieval query 510 starting from the second connection target node n31. As a result, the user can inclusively confirm the existence of the fourth connection target node n41 when retrieving the second connection target node n31. Therefore, it is possible to avoid adding the second connection target node n31 where the fourth connection target node n41 does not exist to scenario 302.

[0142] (4) In the production assistance device in (1) above, the relationship knowledge data 104-i has the association information of the relationship knowledge data 104-i. In the output process, the processor 201 outputs the specific second relationship knowledge data 104-i including the association information.

[0143] Thus, when the user selects the connection target node n from the specific first relationship knowledge data 104-i listed as the retrieval result of the first retrieval query 510, the user can refer to the association information as a judgment index when adding to scenario 302.

[0144] (5) In the production assistance device in (4) above, the association information is the source sentence 410 that is the basis of the relationship knowledge data 104-i.

[0145] Thus, when the user selects the connection target node n from the specific first relationship knowledge data 104-i listed as the retrieval result of the first retrieval query 510, the user can refer to the source sentence 410 as a judgment index when adding to scenario 302.

[0146] (6) In the production assistance device in (4) above, the association information is the link 1215 to the source of the relationship knowledge data 104-i.

[0147] Thus, when the user selects the connection target node n from the specific first relationship knowledge data 104-i listed as the retrieval result of the first retrieval query 510, the user can refer to the web page of the link target of the link 1215 as a judgment index when adding to scenario 302.

[0148] (7) In the production assistance device in (5) above, in the output process, the processor 201 can output the string representing the knowledge specified for each of the second node n2 and the second connection target node n31 and the string representing the association specified by the first edge E2 in an emphatically displayable manner in the source sentence 410.

[0149] Thus, when the user selects the connection target node n from the specific first relationship knowledge data 104-i listed as the retrieval result of the first retrieval query 510, the user can refer to the emphatically displayed part of the source sentence 410 as a judgment index when adding to scenario 302.

[0150] (8) In the production assistance device described in (1) above, in the update process, when, in scenario 302, the second connection target node n31 in the second node n2 and the specific first relation knowledge data 104-i is connected by the second edge e21 corresponding to the first edge E2, the processor 201 updates scenario 302.

[0151] Thereby, it is possible to continue the production of scenario 302 in the latest state.

[0152] (9) In the production assistance device described in (8) above, in the generation process, the processor 201 generates a third search query 510 for retrieving the fourth connection target node n41 leading from the second connection target node n31 based on the first connection target node N3 in the inference route 301, the third connection target node N4 facing from the first connection target node N3, and the third edge E3 connecting the first connection target node N3 and the third connection target node N4; in the search process, the processor 201 searches the relation knowledge DB 104 for the specific second relation knowledge data 104-i that matches the third search query 510 generated by the generation process, and in the output process, the processor 201 outputs the specific second relation knowledge data 104-i.

[0153] Thereby, the production assistance device can generate a search query with scenario 302 in the latest state.

[0154] (10) In the production assistance device described in (9) above, the relation knowledge data 104-i has the association information of the relation knowledge data 104-i; in the generation process, the processor 201 generates the third search query 510 based on the first connection target node N3, the third connection target node N4, the third edge E3 in the inference route 301, and the association information included in the specific first relation knowledge data 104-i, and in the output process, the processor 201 outputs the specific second relation knowledge data 104-i and the association information.

[0155] Thereby, the production assistance device can perform a search using the association information included in the specific first relation knowledge data 104-i as the previous search result.

[0156] (11) In the production assistance device described in (10) above, the association information is the source sentence 410 that is the basis of the relation knowledge data 104-i.

[0157] (12) In the production assistance device of the above (10), during the generation process, the processor 201 determines the relationship knowledge data 104-i that conforms to the third search query 510 as the specific second relationship knowledge data 104-i based on the similarity between the association information of the third search query 510 and the association information of the relationship knowledge data 104-i that conforms to the third search query 510.

[0158] Thereby, it is possible to perform a search such as determining the relationship knowledge data 104-i that conforms to the third search query 510 as the specific second relationship knowledge data 104-i when the similarity is above the threshold, or determining the relationship knowledge data 104-i that conforms to the third search query 510 as the specific second relationship knowledge data 104-i when the similarity is below the threshold. Thus, it is possible to efficiently reduce the specific second relationship knowledge data 104-i.

[0159] (13) In the production assistance device of the above (1), the processor 201 executes a saving process (step S612) of determining the registration target scenario 2000 by the node group selected from the scenario 302 and the edge group connecting the node group, and saving it to the storage device 202.

[0160] Thereby, the production assistance device can easily execute the saving of the required scenario 2000 within the scenario 302 only by the selection of the nodes performed by the user operation.

[0161] In addition, the present invention is not limited to the above-described embodiments, and includes various modification examples and equivalent structures within the gist of the appended claims. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and the present invention is not limited to necessarily having all the structures described. In addition, a part of the structure of a certain embodiment can be replaced with the structure of another embodiment. In addition, the structure of another embodiment can be added to the structure of a certain embodiment. In addition, addition, deletion, or replacement of other structures can be performed on a part of the structure of each embodiment.

[0162] In addition, for the above-described respective structures, functions, processing units, processing mechanisms, etc., a part or all of them can be implemented by hardware, for example, by integrated circuit design, or can be implemented by software by the processor 201 interpreting and executing a program for implementing each function.

[0163] The information of programs, tables, files, etc. that implement various functions can be saved to a storage device such as a memory, a hard disk, an SSD (Solid State Drive), or a recording medium such as an IC (Integrated Circuit) card, an SD card, or a DVD (Digital Versatile Disc).

[0164] In addition, the control lines and information lines represent those considered necessary for explanation, and do not necessarily represent all the control lines or information lines required for installation. In fact, it can be considered that almost all structures are interconnected.

Claims

1. A production assistance device having a processor that executes a program and a storage device that stores the above program, characterized in that, it can access a database that stores a set of relational knowledge data, the above relational knowledge data being composed of two nodes of specified knowledge and an edge that specifies the relationship between the above two nodes and connects the above two nodes; the above processor executes: an acquisition process for acquiring an inference route in which an order is given to a plurality of the above knowledge, the above inference route constituting a hypothesis; an update process for updating the above scenario when a second node obtained by copying the first node in the inference route obtained by the above acquisition process into the above scenario is added to the scenario that concretizes the above hypothesis; a generation process for generating a first search query for searching for a second connection target node in the direction from the above second node, based on the above first node in the inference route, a first connection target node in the direction from the above first node, and a first edge that connects the above first node and the above first connection target node; a search process for searching the above database for specific first relational knowledge data that matches the first search query generated by the above generation process; and an output process for outputting the specific first relational knowledge data retrieved by the above search process, the above update process further includes updating the above scenario when the above second connection target node in the above specific first relational knowledge data is copied into the above scenario and the above second node and the above second connection target node are connected by a second edge corresponding to the above first edge.

2. The production assistance device according to claim 1, characterized in that, in the above generation process, the above processor generates a second search query for searching for a fourth connection target node in the direction from the above second connection target node, based on the above first connection target node in the inference route, a third connection target node in the direction from the above first connection target node, and a third edge that connects the above first connection target node and the above third connection target node; in the above search process, the above processor searches the above database for specific second relational knowledge data that matches the second search query generated by the above generation process; in the above output process, the above processor outputs the number of pieces of the above specific second relational knowledge data and the above specific first relational knowledge data.

3. The production assistance device according to claim 2, characterized in that, the above processor executes an extraction process for extracting a plurality of partial strings from the string of knowledge specified in the above second connection target node, the above partial strings including a part of the string; in the above generation process, the above processor generates the above second search query with each of the strings extracted by the above extraction process as the above second connection target node; in the above search process, the above processor searches the above database for the above specific second relational knowledge data for each of the above second search queries; the above processor executes a calculation process for calculating statistical data on the number of pieces of the above specific second relational knowledge data retrieved from each of the above second search queries. In the above output process, the above processor outputs the above specific first relationship knowledge data and the statistical data calculated through the above calculation process.

4. The production assistance device according to claim 2, wherein, the above relationship knowledge data has associated information of the above relationship knowledge data; in the above output process, the above processor outputs the above specific second relationship knowledge data including the above associated information.

5. The production assistance device according to claim 4, wherein, the above associated information is document data regarding the source as the basis of the above relationship knowledge data.

6. The production assistance device according to claim 4, wherein, the above associated information is access information to the source as the basis of the above relationship knowledge data.

7. The production assistance device according to claim 5, wherein, in the above output process, the above processor outputs, in a manner capable of highlighting display, the string representing the knowledge specified in each of the above second node and the above second connection target node in the above document data, and the string representing the relevance specified by the above first edge.

8. The production assistance device according to claim 1, wherein, in the above generation process, the above processor generates a third search query for retrieving a fourth connection target node directed from the above second connection target node based on the above first connection target node within the above inference route, the third connection target node directed from the above first connection target node, and the third edge connecting the above first connection target node and the above third connection target node; in the above search process, the above processor searches the above database for specific second relationship knowledge data that conforms to the third search query generated through the above generation process; in the above output process, the above processor outputs the above specific second relationship knowledge data.

9. The production assistance device according to claim 8, wherein, the above relationship knowledge data has associated information of the above relationship knowledge data; in the above generation process, the above processor generates the above third search query based on the above first connection target node, the above third connection target node, the above third edge, and the associated information included in the above specific first relationship knowledge data within the above inference route; in the above output process, the above processor outputs the above specific second relationship knowledge data and the above associated information.

10. The production assistance device according to claim 9, wherein, the above associated information is document data regarding the source as the basis of the above relationship knowledge data.

11. The production assistance device according to claim 9, wherein, in the above generation process, the above processor determines the relationship knowledge data that conforms to the above third search query as the above specific second relationship knowledge data based on the similarity between the associated information of the above third search query and the associated information of the relationship knowledge data that conforms to the above third search query.

12. The production assistance device according to claim 1, wherein, The above-mentioned processor executes a saving process of determining a registration target scenario based on a group of nodes selected from the above-mentioned scenarios and a group of edges connecting the above-mentioned group of nodes, and saving the same to the above-mentioned storage device.

13. An authoring assistance method, which is executed by an authoring assistance device having a processor that executes a program and a storage device that stores the above-mentioned program. The authoring assistance method is characterized in that it can access a database storing a set of relational knowledge data, the relational knowledge data being composed of two nodes defining knowledge and an edge defining the relationship between the two nodes and connecting the two nodes; the above-mentioned processor executes: an acquisition process of acquiring an inference route in which an order is assigned to a plurality of the above-mentioned knowledge, the inference route constituting a hypothesis; an update process of updating the scenario when a second node obtained by copying the first node in the inference route obtained by the above-mentioned acquisition process into the scenario is added to the scenario in which the above-mentioned hypothesis is materialized; a generation process of generating a first search query for searching for a second connection target node in the direction from the above-mentioned second node, based on the above-mentioned first node in the inference route, a first connection target node in the direction from the above-mentioned first node, and a first edge connecting the above-mentioned first node and the above-mentioned first connection target node; a search process of searching for specific first relational knowledge data that matches the first search query generated by the above-mentioned generation process from the above-mentioned database; and an output process of outputting the specific first relational knowledge data retrieved by the above-mentioned search process, the above-mentioned update process further includes updating the scenario when the second connection target node in the specific first relational knowledge data is copied into the scenario and the second node and the second connection target node are connected by a second edge corresponding to the above-mentioned first edge.

14. A recording medium, characterized in that it stores an authoring assistance program that causes a processor capable of accessing a database storing a set of relational knowledge data composed of two nodes defining knowledge and an edge defining the relationship between the two nodes and connecting the two nodes to execute: an acquisition process of acquiring an inference route in which an order is assigned to a plurality of the above-mentioned knowledge, the inference route constituting a hypothesis; an update process of updating the scenario when a second node obtained by copying the first node in the inference route obtained by the above-mentioned acquisition process into the scenario is added to the scenario in which the above-mentioned hypothesis is materialized; a generation process of generating a first search query for searching for a second connection target node in the direction from the above-mentioned second node, based on the above-mentioned first node in the inference route, a first connection target node in the direction from the above-mentioned first node, and a first edge connecting the above-mentioned first node and the above-mentioned first connection target node; a search process of searching for specific first relational knowledge data that matches the first search query generated by the above-mentioned generation process from the above-mentioned database; and an output process of outputting the specific first relational knowledge data retrieved by the above-mentioned search process, The above update process further includes updating the above scenario when the above second connection target node in the above specific first relationship knowledge data is copied into the above scenario and the above second node and the above second connection target node are connected by a second edge corresponding to the above first edge.

Images