Idea Support System and Idea Support Method

Abstract

To support a user to develop ideation based on analogy.SOLUTION: In an ideation support system 100, conversion with an idea encoder 105 has such features that smallness of a distance between multivariate items and similarity of thinking data before conversion have a correlation to each other in multivariate items obtained by converting a plurality of pieces of thinking data. A data selection unit (analogy reminiscent data selection unit 104) analyzes, for a first multivariate item and a second multivariate item converted by the idea encoder 105, relationships between multivariate components and field codes to remove components affected by the field codes from the multivariate items and selects analogy reminiscent information capable of reminding a user of analogy from the thinking data recorded in a thinking database on the basis of similarity indicated by a distance between the first multivariate item and the second multivariate item from which the components have been removed.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an idea support system and an idea support method, and is suitable for application to an idea support system and an idea support method for presenting information for supporting human ideas.

Background Art

[0002] Those engaged in research and development or planning are required to come up with new ideas different from known ones. Usually, in such cases, the person concerned often deepens their own thinking and collects relevant information in order to obtain new ideas.

[0003] Generally, as a method of idea generation, a strategy of expanding ideas by repeating the "divergence" and "convergence" of ideas is often adopted. That is, through information collection, etc., the thinking currently being considered is once expanded regardless of feasibility, etc., and then, by repeating the process of settling down into a realistic form, it can be expanded into better ideas. In such a thinking process, the "divergence" part is particularly difficult. As a commonly used method, there is a method called brainstorming, in which multiple people come up with ideas one after another and expand ideas based on them. This is a method of expanding ideas by borrowing the ideas of others because the scope of ideas cannot be expanded by one person alone.

[0004] As described above, since it is difficult to expand ideas alone, a form of cooperation among multiple people is often used. However, since multiple people need to spend time, there are cases where cooperation cannot be obtained or there are differences in prior knowledge, and there are also many situations where one has to expand ideas alone. In this case, it is assumed that reference examples are searched by keyword searching on the Web or in documents. However, when keywords for searching are input, there is a problem that only the same kind of information can be obtained and it is difficult to obtain information that can be a hint for expanding ideas.

[0005] Therefore, a system has been developed that promotes the spread of accidental ideas by presenting information randomly, etc. However, in random information presentation, a lot of information that is far from the current idea is also displayed. From the above, a mechanism for efficiently presenting useful information is required.

[0006] For example, Patent Document 1 discloses a method of extracting an object from an image, encoding it into multiple variables, converting it into another object by adjusting the components of the multiple variables, and presenting the image to expand ideas. By this method, an unknown image can be created by rewriting a part of the image, and new ideas based on it can be expected.

[0007] Also, for example, Patent Document 2 discloses a method of determining the similarity of text from syntactic dependencies in text analysis of documents and determining the text to be presented to the user. According to this method, another text similar to the text expressing the current idea can be retrieved.

Prior Art Documents

Patent Documents

[0008]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0009] However, in the method disclosed in Patent Document 1, if the object for image replacement is determined in advance, accidental ideas cannot be generated, and if the image is randomly replaced, only information that is too random and has low usefulness can be obtained, and this problem remains. Also, the method disclosed in Patent Document 2 is a method of retrieving similar texts, and since it does not separate the encoded multiple variables, there is a problem that the serendipity in expanding ideas is not high.

[0010] As described above, among the idea support systems based on the prior art, there are some that attempt to stimulate new ideas by presenting new information to people. However, the information presented is highly random, and it is difficult for the presented information to lead to ideas. The wider the range of expanding ideas, the more exponentially the information that does not lead to ideas increases. Therefore, if random information is presented, it becomes difficult to select the information that leads to new ideas, which is time-consuming for the users of the system. On the other hand, if the range of expanding ideas is narrowed, information that greatly exceeds the original idea category cannot be presented, and the value of using the system decreases.

[0011] The present invention has been made in consideration of the above points, and proposes an idea support system and an idea support method capable of presenting information similar to the current ideas even from fields different from the fields in which the user wants to have ideas, and supporting the user to expand ideas based on analogies.

Means for Solving the Problems

[0012] In order to solve such problems, in the present invention, a thought database that records thought data obtained by informatizing existing thoughts in association with a field code indicating the corresponding field of the thought, an input unit that receives user thought data which is thought data obtained by informatizing the user's thoughts, an idea encoder that converts the thought data and the user thought data into multivariate data respectively, a first multivariate data obtained by converting the user thought data by the idea encoder, and a second multivariate data obtained by converting the thought data recorded in the thought database by the idea encoder, and a data selection unit that selects analogy evoking information capable of evoking an analogy for the user based on the above. The conversion by the idea encoder has a feature that in the multivariate data obtained by converting a plurality of thought data, the distance between the multivariate data is correlated with the similarity of the thought data before conversion. The data selection unit analyzes the relevance between the components of the multivariate data and the field code for each of the first multivariate data and the second multivariate data converted by the idea encoder, removes the components affected by the field code from the multivariate data, and selects the analogy evoking information from the thought data recorded in the thought database based on the similarity indicated by the distance between the first multivariate data and the second multivariate data after the removal. A creative support system is provided, which is characterized by the above.

[0013] In addition, in order to solve such problems, the present invention provides a method for assisting idea generation by an idea generation support system that supports the development of a user's thinking. The idea generation support system includes a thinking database that records thinking data obtained by digitizing existing thinking, associating it with a field code indicating the corresponding field of the thinking; an input unit that receives user thinking data, which is thinking data obtained by digitizing a user's thinking; an idea encoder that converts the thinking data and the user thinking data into multiple variables; a data selection unit that selects analogy reminder information that can cause a user to recall an analogy based on a first multiple variable obtained by converting the user thinking data by the idea encoder and a second multiple variable obtained by converting the thinking data recorded in the thinking database by the idea encoder. The conversion by the idea encoder has a feature that in the multiple variables obtained by converting a plurality of pieces of thinking data, the distance between the multiple variables is correlated with the similarity of the thinking data before conversion. The data selection unit analyzes the relevance between the components of the multiple variables and the field code for each of the first multiple variable and the second multiple variable converted by the idea encoder, removes the components affected by the field code from the multiple variables, and selects the analogy reminder information from the thinking data recorded in the thinking database based on the similarity indicated by the distance between the first multiple variable and the second multiple variable after the removal.

Effect of the Invention

[0014] According to the present invention, it is possible to support a user in expanding an idea based on an analogy.

Brief Description of the Drawings

[0015]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Figure 23

Figure 24

Embodiments for Carrying Out the Invention

[0016] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[0017] Note that the following description and drawings are examples for explaining the present invention, and for clarification of the explanation, omissions and simplifications are made as appropriate. Also, not all combinations of features described in the embodiments are essential for the solution means of the invention. The present invention is not limited to the embodiments, and all application examples that conform to the idea of the present invention are included in the technical scope of the present invention. Those skilled in the art can make various additions and changes within the scope of the present invention. The present invention can also be implemented in various other forms. Unless otherwise particularly limited, each component may be plural or singular.

[0018] In the following description, various information may be described using expressions such as "table", "sheet", "list", "queue", etc., but the various information may be expressed in other data structures. In order to indicate independence from the data structure, "XX table", "XX list", etc. may be referred to as "XX information". When explaining the content of each information, expressions such as "identification information", "identifier", "name", "ID", "number", etc. are used, and these can be replaced with each other.

[0019] Also, in the following description, when explaining without distinguishing between elements of the same kind, reference signs or common numbers in the reference signs are used, and when explaining by distinguishing between elements of the same kind, the reference signs of those elements may be used or the ID assigned to those elements may be used instead of the reference signs.

[0020] In the following description, there may be cases where the processes performed by executing a program are described. However, the program is executed by at least one or more processors (e.g., CPU), and in order to perform the defined processes while appropriately using storage resources (e.g., memory) and / or interface devices (e.g., communication ports), etc., the processor may be regarded as the main body of the process. Similarly, the main body of the process performed by executing the program may be a controller, a device, a system, a computer, a node, a storage system, a storage device, a server, a management computer, a client, or a host having a processor. The main body of the process performed by executing the program (e.g., a processor) may include a hardware circuit that performs part or all of the process. For example, the main body of the process performed by executing the program may include a hardware circuit that performs encryption and decryption, or compression and decompression. The processor operates as a functional unit that realizes a predetermined function by operating according to the program. The devices and systems including the processor are devices and systems including these functional units.

[0021] The program may be installed from a program source into a device such as a computer. The program source may be, for example, a program distribution server or a computer-readable storage medium. When the program source is a program distribution server, the program distribution server includes a processor (e.g., CPU) and storage resources, and the storage resources may further store a distribution program and the program to be distributed. Then, by the processor of the program distribution server executing the distribution program, the processor of the program distribution server may distribute the program to be distributed to other computers. Also, in the following description, two or more programs may be realized as one program, or one program may be realized as two or more programs.

[0022] (1) First Embodiment (1-1) Configuration FIG. 1 is a block diagram showing a configuration example of an idea support system 100 according to a first embodiment of the present invention. The idea support system 100 includes a user idea reception unit 101 that receives an input from a user (user) of the system, an idea support information presentation unit 102 that presents information to the user, a screen transition control unit 103 that controls the flow of screen transitions on the display screen presented to the user, an analogy recollection data selection unit 104 having a function of selecting information that can cause the user to recollect an analogy, an idea encoder 105 having a function of converting information expressing an idea into a multi-variable (for example, vector data), a thought database 106 that stores data of cases (thought case data) expressing existing ideas in various fields, a history storage database 107 that stores the usage history of the system by the user, and a user authentication unit 108 having a function of authenticating the user. Details of each unit will be described later with reference to the drawings as appropriate.

[0023] FIG. 2 is a block diagram showing a hardware configuration example of the idea support system 100. The idea support system 100 can be implemented using a general computer. That is, as illustrated in FIG. 2, the idea support system 100 includes a processor 11 having computing performance, a DRAM (Dynamic Random Access Memory) 12 that is a volatile temporary storage area capable of high-speed reading and writing, a storage device 13 that is a persistent storage area using an HDD (Hard Disk Drive), a flash memory, etc., an input device 14 such as a mouse and a keyboard for performing operations, a monitor 15 for indicating an operation to the user, and a communication interface (I / F) 16 such as a serial port for communicating with the outside, and is an information processing device configured to include these components.

[0024] The hardware shown in FIG. 2 realizes each component of the idea support system 100 shown in FIG. 1 as follows, for example. The user idea reception unit 101, the idea support information presentation unit 102, the screen transition control unit 103, the analogy recollection data selection unit 104, the idea encoder 105, and the user authentication unit 108 are realized by the processor 11 executing a program recorded in the storage device 13. Also, the thinking database 106 and the history storage database 107 can be implemented by the processor 11 executing a program that stores data in the storage device 13.

[0025] (1-2) Idea Support Processing FIG. 3 is a diagram schematically showing the overall operation procedure in the first embodiment. FIG. 3 shows the interaction between the user 200 and the idea support system 100 and the operation content within each element of the idea support system 100, centered around the idea support processing executed by the idea support system 100 when supporting the idea of the user 200.

[0026] According to FIG. 3, first, when the user 200 activates the idea support system 100 and performs a login operation to the system, the user authentication unit 108 of the idea support system 100 performs user authentication to identify the user 200 who logs in to the system using a known user authentication method (step S101).

[0027] Next, the idea support system 100 presents a topic specification screen and requests the user to specify a "topic" indicating the field in which the user wants to generate a new idea. Then, when the user performs an input operation to specify a topic for the topic specification screen 310, the idea support system 100 receives information (topic information) related to the specified topic (step S102). A specific example of the topic specification screen will be shown in FIG. 4 described later, and a specific example of the topic information will be shown in FIG. 5 described later.

[0028] Next, the idea support system 100 generates a screen (initial idea input screen) for inputting an initial text by executing an initial text reception screen presentation process, and presents it to the user 200 (step S103). Details of the initial text reception screen presentation process will be described later with reference to FIG. 6. Also, a specific example of the initial idea input screen is shown in FIG. 7 described later.

[0029] When the user 200 inputs an initial text on the initial idea input screen presented in step S103, the idea support system 100 executes an analogy recall information presentation process while repeatedly receiving input of updated text from the user 200, and presents information (analogy recall screen) for prompting the user 200 to make a leap in ideas, thereby guiding the thinking of the user 200 to a new area (step S104). Details of the analogy recall information presentation process will be described later with reference to FIG. 9. Also, a specific example of the analogy recall screen is shown in FIG. 11 described later.

[0030] After going through the process of step S104, when the user 200 finally determines that the idea conversion has been sufficiently achieved, the idea support system 100 determines the final consideration result in response to this (step S105), and ends the idea support work.

[0031] Note that in the process of the above series of procedures, when the user 200 gives an input (for example, pressing a predetermined button on the display screen, etc.) instructing a change to the "history display mode" for displaying the history of past ideas, the idea support system 100 can be switched to the display of the history display mode at any time (step S106).

[0032] The above is the outline of the idea support process executed by the idea support system 100. Below, the procedure of the idea support process will be described in more detail.

[0033] (1-2-1) Topic specification Regarding the process of step S102 in FIG. 3, a topic designation screen 310 for designating a topic and topic information generated based on the topic designation will be described in detail.

[0034] FIG. 4 is a diagram showing an example of a topic designation screen. The topic designation screen is a screen for allowing the user 200 to designate a field (topic) in which a new idea is desired. Topic information is generated by associating the designated topic with the user's information.

[0035] Note that the topic information may be generated by the process of the program that provides the topic designation screen 310 (i.e., the idea support information presentation unit 102) when the determination button 313 is pressed on the topic designation screen 310, or the information itself input on the topic designation screen 310 may be transmitted to the user idea reception unit 101, and the topic information may be generated in the user idea reception unit 101.

[0036] In the case of FIG. 4, the topic designation screen 310 includes a text box 311 for inputting the name of the topic (topic name), a drop-down box 312 for designating the field of the topic (target field), a determination button 313 to be pressed when the input of these is completed, and a transition button 314 to be pressed when switching to the history display mode. When the user 200 operates the drop-down box 312, the names of predetermined fields are displayed in a list as candidates, and by selecting one from the displayed candidates, the user 200 can designate the field of the topic. Then, when the user 200 operates the determination button 313 after designating the topic name and its target field, the topic information based on the designated content is transmitted to the idea support system 100 and received by the user idea reception unit 101.

[0037] FIG. 5 is a diagram showing a configuration example of topic information. The topic information 410 shown in FIG. 5 includes items such as a user ID 411 which is an identifier uniquely identifying a user, a topic ID 412 which is an identifier of a topic, a topic name 413 indicating the topic name input by the user on the topic designation screen 310, and a field code 414 indicating an identification code of the field in which the user wants to have ideas.

[0038] The user ID 411 is information identifying the user 200 specified in the user authentication in step S101 of FIG. 3. When information capable of specifying the user ID in the user authentication is obtained, that information can be used. Also, the information recorded in the user ID 411 is not limited to the information specified in the user authentication. For example, any information that can uniquely identify the user, such as employee information, can be used if it is separately specified.

[0039] The topic ID 412 is generated in a way that can uniquely identify the topic, such as being assigned a serial number when the topic information 410 is recorded.

[0040] The topic name 413 stores the information input in the text box 311 on the topic designation screen 310.

[0041] The field code 414 stores a code value uniquely corresponding to the field specified in the drop-down box 312 on the topic designation screen 310. This field code (field symbol) is an identifier uniquely identifying the field and is assumed to be numbered using a predetermined system. Specifically, for example, when the field is "database system", it is "1", and when it is "artificial intelligence system", it is "2", etc. A system in which the code and the field name correspond in a defined form is used. Note that as the system of the field code, a known classification code such as a patent classification code may be used.

[0042] (1-2-2) Initial document reception screen presentation process The initial copy reception screen presentation process in step S103 of FIG. 3 will be described in detail. As described above, when a topic is specified by the user 200 on the topic specification screen and topic information 410 is generated based on the specified content, the idea support system 100 starts the initial copy reception screen presentation process.

[0043] FIG. 6 is a diagram schematically showing an example of the processing procedure of the initial copy reception screen presentation process. As shown in FIG. 6, the initial copy reception screen presentation process is executed by the user idea reception unit 101, the screen transition control unit 103, and the idea support information presentation unit 102.

[0044] According to FIG. 6, first, the user idea reception unit 101 receives the topic information 410 generated based on the topic specification on the topic specification screen (step S201). Then, the user idea reception unit 101 sends the received topic information 410 to the screen transition control unit 103.

[0045] Next, the screen transition control unit 103 stores the topic information 410 in the history storage database 107 (step S202), and further sends the topic information 410 to the idea support information presentation unit 102.

[0046] Then, the idea support information presentation unit 102 generates an initial idea input screen and presents it to the user 200 to inquire about the initial situation of the idea (step S203), and ends the initial copy reception screen presentation process.

[0047] FIG. 7 is a diagram showing an example of the initial idea input screen. The initial idea input screen 320 shown in FIG. 7 includes a text box 321 for inputting the content of the user's own initial idea, a decision button 322 to be pressed when determining the input content, a text box 323, an import button 324, and a transition button 325 to be pressed when switching to the history display mode.

[0048] On the initial idea input screen 320, in the text box 321, the user 200 can input the content of their idea in text (a sentence). After the input, by pressing the OK button 322, information indicating the user's idea content (user idea data) is generated based on the input information and sent to the user idea reception unit 101. A specific example of the user idea data will be shown in FIG. 8 described later.

[0049] Note that the user idea data may be generated by the process of the program that provides the initial idea input screen 320 (i.e., the idea support information presentation unit 102) when the OK button 322 is pressed on the initial idea input screen 320. Or, the information itself input on the initial idea input screen 320 may be sent to the user idea reception unit 101, and the user idea data may be generated in the user idea reception unit 101.

[0050] Also, the text box 323 and the import button 324 are a mechanism for acquiring data from a system (hereinafter referred to as a document creation support device), not shown, that supports the user in creating a sentence, which is separate from the idea support system 100. Specifically, for example, by inputting the URL indicating the storage destination of the sentence created using the document creation support device into the text box 323 and pressing the import button 324, the text obtained from the URL is automatically input into the text box 321.

[0051] FIG. 8 is a diagram showing a configuration example of user idea data. The user idea data 420 shown in FIG. 8 is an example of user idea data generated based on the input to the initial idea input screen 320, or user idea data generated when idea modification is performed on the analogy recall screen 330 of FIG. 11 described later. The user idea data 420 includes items of a user ID 421, a topic ID 422, a field code 423, an idea ID 424, an idea date and time 425, and an idea content expression 426. Among these, the user ID 421, the topic ID 422, and the field code 423 are stored by replicating the contents of the items with the same name from the topic information 410 illustrated in FIG. 5. Although details will be described later in the description of FIG. 14, depending on the method of removing the field component from the vector data (user idea vector data) converted from the idea content expression 426 of the user idea data 420, the user idea data 420 may not include the field code 423.

[0052] The idea ID 424 uses a value generated in a way that can uniquely identify the corresponding topic, such as numbering in sequence when the topic information 410 is recorded. The idea date and time 425 records a time stamp of the time when the user idea data 420 is generated. The idea content expression 426 stores the character string of the text input by the user 200 as his or her own idea in the text box 321 on the initial idea input screen 320 (or the character string of the text representing the idea modified by the user 200 on the analogy recall screen 330 described later).

[0053] (1-2-3) Analogy Recall Information Presentation Process The analogy recall information presentation process in step S104 of FIG. 3 will be described in detail. As described above, when the user 200 inputs his or her own initial idea in text on the initial idea input screen and the user idea data 420 is generated based on the input content, the analogy recall information presentation process starts.

[0054] FIG. 9 is a diagram schematically showing an example of a processing procedure of analog recall information presentation processing. As shown in FIG. 9, the analog recall information presentation processing is executed by a user idea reception unit 101, a screen transition control unit 103, an analog recall data selection unit 104, and an idea support information presentation unit 102.

[0055] According to FIG. 9, first, the user idea reception unit 101 of the idea support system 100 receives user idea data 420 generated based on the input on the initial idea input screen (step S301). Then, the user idea reception unit 101 sends the received user idea data 420 to the screen transition control unit 103.

[0056] Next, the screen transition control unit 103 stores the user idea data 420 in the history storage database 107 (step S302), and further sends the user idea data 420 to the analog recall data selection unit 104.

[0057] Then, the analog recall data selection unit 104 executes an analog recall data selection process of examining the thinking database 106 based on the user idea data 420 and selecting thinking case data contributing to the analogy (step S303). Specific examples of the thinking case data will be shown in FIG. 10 described later. Also, the details of the analog recall data selection process will be described later with reference to FIG. 12. As a result of this process, a set of thinking case data is selected, and the analog recall data selection unit 104 sends the selected set of thinking case data to the screen transition control unit 103.

[0058] FIG. 10 is a diagram showing a configuration example of thought example data. The thought example data 430 shown in FIG. 10 includes a case ID 431 which is an identifier for identifying a case, case attribute information 432 for recording supplementary information regarding the case such as the author, a field code 433 which stores a value of the same system as the field code 414 (see FIG. 5) of the topic information 410, which is a code value indicating the field of the case, thought expression data 434 which is a character string of a sentence expressing the idea of the case, and thought expression vector data 435 which stores vector data obtained by inputting the character string of the thought expression data 434 into the idea encoder 105. The operation of the idea encoder 105 will be described in detail later with reference to FIG. 13.

[0059] In the thought database 106, before the idea support system 100 starts the idea support process, that is, before the system is operational, the thought example data 430 is prepared and stored by processing data that can be obtained, for example, by searching patent gazettes. Note that for the thought database 106, data may be periodically crawled or the like, and the thought example data 430 to be stored may be added at any time.

[0060] Returning to the description of FIG. 9, as described above, a set of thought example data 430 is selected by the analogy recall data selection process in step S303. Next, the screen transition control unit 103 generates an analogy recall screen based on this set of thought example data 430 sent from the analogy recall data selection unit 104 (step S304). Then, the generated analogy recall screen 330 is presented to the user 200 through the idea support information presentation unit 102 (step S305).

[0061] FIG. 11 is a diagram showing an example of an analogy recall screen. The analogy recall screen 330 shown in FIG. 11 includes a current idea display box 331, an idea correction box 332, an analogy recall information display box 333, a decision button 334, an end button 335, and a transition button 336 which is pressed when switching to the history display mode.

[0062] The current idea display box 331 is a text box that displays the current idea (thought), and the thought content expression 426 of the user thought data 420 received in the immediately preceding step S301 is displayed.

[0063] The idea correction box 332 is a text box for correcting the current idea shown in the current idea display box 331, and is an interface capable of receiving correction of the text by the user 200. The idea correction box 332 initially displays the same text as the current idea display box 331, and the user 200 can rewrite the content of the idea by correcting the text according to his / her own thought.

[0064] The analogy recall information display box 333 is an area for displaying a set of thought example data 430 obtained by the analogy recall data selection process for reference of the thought of the user 200 when rewriting the idea with respect to the idea correction box 332. The set of thought example data 430 displayed in the analogy recall information display box 333 is selected by the analogy recall data selection process, and the details will be described later with reference to FIG. 12. The display mode in the analogy recall information display box 333 is not limited to a specific one, but for example, it is preferable to display the information of the case in an easy-to-understand manner using the case attribute information 432 and the thought expression data 434 of the thought example data 430. When the case attribute information 432 includes a URL or the like, corresponding information may be acquired from outside the idea support system 100 based on this URL or the like and displayed in the analogy recall information display box 333.

[0065] The decision button 334 is a button pressed by the user 200 when the input of the user 200 in the analogy recall screen 330 (specifically, the correction of the idea in the idea correction box 332) is completed.

[0066] The end button 335 is a button that is pressed by the user 200 when the expansion of ideas using the analogy recall screen 330 is completed, that is, when the consideration of the topic is completed.

[0067] In the analogy recall screen 330, after the text in the idea correction box 332 is corrected and the decision button 334 is pressed, new user idea data 420 is generated based on the information input on the screen (the content of the idea correction box 332) and sent to the user idea reception unit 101. The method of generating the user idea data 420 may be considered in the same way as the generation of the user idea data 420 when the decision button 322 is pressed in the initial idea input screen 320 shown in FIG. 7. Specifically, in the generation of the user idea data 420, the corresponding data of the topic information 410 is copied to the user ID 421, the topic ID 422, and the field code 423, the time stamp of the time when this data is generated is recorded in the idea date and time 425, and the character string of the text in the idea correction box 332 corrected by the user 200 is stored in the idea content expression 426.

[0068] As described above, when the decision button 334 is pressed after the text in the idea correction box 332 is corrected, new user idea data 420 storing a new idea is generated and sent to the user idea reception unit 101. Therefore, in FIG. 9, based on the new user idea data 420, the processes of steps S301 to S305 can be executed again. By repeating this procedure many times, the ideas of the user 200 will spread with reference to the analogy recall information. Then, when the development of the ideas of the user 200 is exhausted, the end button 335 is pressed on the analogy recall screen 330. At this time, the repetition of the above procedure ends, and the analogy recall information presentation process ends. That is, the idea by the user 200 is determined by the content of the final idea correction box 332.

[0069] (1-2-3-1) Analogy Recall Data Selection Process Hereinafter, the analog recall data selection process executed in step S303 of FIG. 9 will be described in detail.

[0070] FIG. 12 is a diagram schematically showing an example of the processing procedure of the analog recall data selection process. In FIG. 9, it was described that the analog recall data selection unit 104 executes the analog recall data selection process. However, strictly speaking, as shown in FIG. 12, the process of step S402 by the idea encoder 105 is included in the middle.

[0071] According to FIG. 12, first, the analog recall data selection unit 104 extracts the character string of the idea expression 426 from the user idea data 420 received from the screen transition control unit 103 (step S401), and passes it to the idea encoder 105.

[0072] Next, the idea encoder 105 converts the received character string of the idea expression 426 into corresponding vector data (step S402). This vector data is a sequence of numerical values having a predetermined length. Specifically, for example, it is a numerical sequence such as (0.1, 2.1, 1.2, 0.1, 0.6). The conversion to vector data in step S402 is performed so that when the contents of the idea expression 426 are similar, the vector data is also similar, that is, the distance between the vector data becomes small.

[0073] FIG. 13 is a diagram schematically explaining the operation of converting a character string into vector data by the idea encoder 105.

[0074] For the conversion from a character string to vector data by the idea encoder 105, known embedding processing can be applied. Generally, embedding refers to a process of converting data into vector data such that the mathematical structure of the original data is maintained as much as possible. In the present invention, a method is adopted in which the similarity between the sentences (character strings) of the idea content expressions 426 in the plurality of user idea data 420 is reflected in the distance between the vector data. As a method of embedding a character string, a method using a neural network such as Seq2Seq (Sequence To Sequence) or Word2Vec (Word to Vector) is known.

[0075] Referring to FIG. 13, in this example, the idea encoder 105 first splits the input character string 511 into words and converts it into a column of vectors by a method called One-Hot encoding. In One-Hot encoding, each component of the numerical sequence corresponds to a word, and for each word, only the corresponding component is converted to "1" and the others are converted to "0". As a result, the character string 511 is converted into a vector column 512 consisting of a plurality of vectors in which only one of the components is "1" and the rest are "0". The length of this vector column 512 varies depending on the input character string 511.

[0076] Therefore, next, the idea encoder 105 inputs the vector column 512 into a neural network 513 incorporating a recurrent neural network or the like that can receive variable-length data. As a result, a vector column 514 is obtained as the output of the neural network 513, and this vector column 514 becomes the vector data obtained by converting the character string 511.

[0077] Note that the above neural network 513 is trained in advance using a large number of texts. This training may be the training of a neural network having an embedding function, and known methods can be used. Specifically, for example, methods such as adversarial generative networks, variational autoencoders, and GPT using an attention mechanism can be applied. In these embeddings, the similarity of the input text (character string) correlates with the distance between the vector data (vector sequence 514) output from the neural network 513. However, depending on the embedding method, the perspective on the similarity of texts is different. Therefore, the idea support system 100 may be configured to select, according to the purpose of idea support by the idea support system 100 and the like, what kind of embedding the idea encoder 105 uses.

[0078] Returning to the description of FIG. 12. The analogical recall data selection unit 104 that has received the vector data (hereinafter also referred to as user idea vector data) converted from the character string of the idea content expression 426 by the idea encoder 105 in step S402 uses this vector data to obtain a plurality of thought example data 430 (see FIG. 10) from the thought database 106 (step S403).

[0079] The thought example data 430 obtained in step S403 includes a field code 433 and thought expression vector data 435. Therefore, the analogical recall data selection unit 104 identifies a component correlated with the field (hereinafter referred to as the field component) in the thought expression vector data 435 based on the relationship between the two (step S404), and performs a process of removing the field component from the thought expression vector data 435 of the thought example data 430 obtained in step S403 (step S405). Note that in step S405, the analogical recall data selection unit 104 also performs a process of removing the field component on the user idea vector data converted in step S402 in the same manner.

[0080] FIG. 14 is a diagram schematically explaining a method of removing field components from vector data. Hereinafter, with reference to FIG. 14, a method of removing field components from the thought expression vector data 435 of the thought example data 430 will be described.

[0081] The scatter diagram 520 shown in (A) in FIG. 14 is an example in which the thought expression vector data 435 is plotted. For simplicity, the thought expression vector data 435 is represented in a two-dimensional xy plane. In the scatter diagram 520, the black circle 521 and the white circle 522 each correspond to one case. Among these, the black circle 521 means a case in "a certain field", and the white circle 522 means a case in "another field" other than "a certain field".

[0082] Next, the analogy recall data selection unit 104 performs rotation, parallel translation, or a combination thereof on the scatter diagram 520 to convert the vector basis 531 to x'y' as in the scatter diagram 530 shown in (B) in FIG. 14. In the scatter diagram 530, for reference, the basis 532 based on xy before conversion is shown by a broken line. In the example of the scatter diagram 530, the coordinate transformation parameters are optimized so that the black circle 521 is arranged as much as possible in the negative region of x', and the white circle 522 is arranged as much as possible in the positive region of x'. This optimization calculation can be realized by determining a boundary line by known linear discriminant analysis and rotating it so that the line is parallel to the x' axis. This optimization is for identifying the component most affected by the field. In the scatter diagram 530 with the transformed basis, the greater the distance in the x' axis direction between the plots, the more different the fields are. Also, since the y' axis direction corresponds to a component not affected by the field, the greater the distance in the y' axis direction between the plots, the lower the similarity. Therefore, the analogy recall data selection unit 104 can identify the field components correlated with the field in the thought expression vector data 435 by performing a basis transformation from the scatter diagram 520 to the scatter diagram 530 (step S404 in FIG. 12).

[0083] Next, the analogy recall data selection unit 104 subtracts the value in the x'-axis direction (x' value) of each plot in the scatter diagram 530 and fixes it to "0". As a result, as shown in the scatter diagram 540 shown in (C) in FIG. 14, for all plots (black circles 521, white circles 522), only the components independent of the field remain on the y'-axis. Therefore, the analogy recall data selection unit 104 can remove the field components from the thought expression vector data 435 by performing the conversion to the scatter diagram 540 (step S405 in FIG. 12).

[0084] Note that in reality, the thought expression vector data 435 is not limited to two dimensions, but the same treatment can be performed even in the case of multiple dimensions.

[0085] Also, when there are three or more fields, an axis that can discriminate between "a certain field" and other fields is obtained, subtracted along that axis, and the same process is repeated by obtaining an axis that can make the same discrimination with another field as "a certain field" and subtracting it. Alternatively, for example, appropriate numerical values such as "1" for the computer field, "2" for the materials science field, and "3" for the mechanical field are assigned to the fields, and an axis that can predict the value by known regression analysis is obtained. Even when dealing with a large number of fields more than the number of dimensions of the thought expression vector data 435, the analogy recall data selection unit 104 can remove the field components.

[0086] And the method of removing the field components from the vector data described above can be similarly applied when removing the field components from the user idea vector data.

[0087] For example, when applying the above-described method of removing the field components to the user idea vector data (broadly speaking, the user idea data 420), similar to the image in FIG. 14, a vector transformation is obtained using the dataset of the thought example data 430 and its field code 423, and the field components can be removed from the thought example data 430 by performing coordinate rotation and translation according to the vector transformation.

[0088] Note that, as a variant method at this time, when obtaining the vector transformation, the user's idea data 420 may be included by a process of removing the field component without using the user's idea data 420. When using this method, the process can be executed even if the field code 423 is not present in the user's idea data 420. Therefore, when adopting this method, the field code 423 does not have to be input when inputting the user's idea data 420 on the initial idea input screen 320 or the analogy recall screen 330. Alternatively, when the field code 423 is not input, a function such as estimating the field code 423 of the user's idea data 420 using known discriminant analysis may be provided to the user's idea reception unit 101 or the like.

[0089] Returning again to the description of FIG. 12. After removing the field component from the thought expression vector data 435 and the user's idea vector data in step S405, the analogy recall data selection unit 104 calculates the distance between the user's idea vector data and each thought expression vector data 435 in the state where the field component is removed (step S406). In step S406, the smaller the distance, the higher the similarity.

[0090] Next, based on the calculation result of step S406, the analogy recall data selection unit 104 selects a predetermined number of thought example data 430 having the corresponding thought expression vector data 435 in ascending order of the distance from the user's idea vector data (vector data of the idea content expression 426) (step S407). For example, the Euclidean distance (square root of the sum of the squares of the differences of each component) between the two vectors can be calculated, and those with a value equal to or less than a predetermined threshold can be selected. Any distance can be used as long as it can evaluate the similarity. Then, the analogy recall data selection unit 104 sends the set of the selected thought example data 430 to the screen transition control unit 103 and ends the analogy recall data selection process.

[0091] By performing the analogical recollection data selection process as described above, it is possible to select, as a group of thought example data 430, information that has a high similarity to the current idea of user 200 after removing the influencing components by field, and by displaying these in the analogical recollection information display box 333 of the analogical recollection screen 330, it is possible to present information that stimulates the user's creativity beyond the field to which the topic they are trying to come up with belongs.

[0092] (1-2-4) History display mode Below, the history display mode will be described. The history display mode is a mode used when user 200 wants to review past topics and records of ideas (user idea data 420), etc. When a predetermined operation (pressing the transition buttons 314, 325, 336) for specifying a transition to the history display mode is performed on each of the various display screens described above, a thought history display screen showing the history of past ideas is displayed.

[0093] FIG. 15 is a diagram showing an example of a thought history display screen.

[0094] In the thought history display screen 340 shown in FIG. 15, a list of past thought histories is displayed in the text box 341 at the upper part of the screen based on the user idea data 420 stored in the history storage database 107. More specifically, this list display obtains all of the user idea data 420 having a combination of the user ID 411 and topic ID 412 of user 200 that has been acquired by the topic information 410 from the history storage database 107 (that is, the combination of user ID 421 and topic ID 422 matches), sorts these user idea data 420 in the order of the thought date and time 425, and sequentially displays predetermined information (for example, the thought date and time, the beginning part of the thought content expression, etc.) stored in each of these user idea data 420 in a plurality of text boxes 341. Also, in the thought history display screen 340 of FIG. 15, an interface is provided such that the plurality of text boxes 341 can be scrolled, and it is possible to select one from a large number of cases (thought histories).

[0095] When any one of the multiple text boxes 341 is selected, detailed information about the corresponding case (for example, the full text of the idea date and time and the idea content expression) is displayed in the text box 342.

[0096] Also, at the lower left of the screen, there is a model display box 343 that displays an "Idea History Correlation Diagram" that schematically depicts the relevance between the currently selected idea history and other idea histories. In the Idea History Correlation Diagram displayed in the model display box 343, the star mark represents the position of the idea history (user idea data 420) currently selected in the text box 341. The white rectangle represents the position of the unselected idea history (user idea data 420). And the curved line with arrows drawn in the form of connecting the star mark and the multiple white rectangles is the connection of the user idea data 420 related to the target topic of the user 200 in chronological order, representing the thinking process of the user 200. Also, the white circle represents the thinking case data 430 selected from the thinking database 106. When any one of the white circles is selected, detailed information about the thinking case data 430 is displayed in the text box 344.

[0097] In addition, the circle 345 drawn by a broken line with the star mark as the center represents the selection criteria of the analogy recall information presented in the analogy recall information display box 333 on the analogy recall screen 330 around the currently selected idea history (user idea data 420). In other words, the circle 345 drawn by the above broken line represents the threshold value that serves as the selection criteria when selecting a set of thought example data 430 in step S407 of FIG. 12 (specifically, for example, the thought example data 430 corresponding to the thought expression vector data 435 whose distance from the user idea vector data is below the threshold value is selected as the analogy recall information), and this threshold value may be displayed numerically. In addition, an editing function such as changing the radius (threshold value) of the circle 345 may be added to the idea history display screen 340. The change of the threshold value is performed, for example, according to the wish of the user 200, and the user idea reception unit 101 may receive this and notify the analogy recall data selection unit 104. By adding a function to change the threshold value, the presentation range of the analogy recall information can be adjusted according to the requirements of the user 200. By changing the criteria for selecting the thought example data 430, the similarity of the thought example data 430 presented to the user 200 also changes, so effects such as changing the direction in which the user 200 advances the idea can be expected.

[0098] FIG. 16 is a diagram schematically showing an example of the processing procedure of the idea history correlation diagram generation process. The idea history correlation diagram generation process shown in FIG. 16 is a process for creating the idea history correlation diagram displayed in the model display box 343 of FIG. 15, and is executed by the idea support information presentation unit 102 and the analogy recall data selection unit 104.

[0099] According to FIG. 16, first, the idea support information presentation unit 102 acquires all the past user idea data 420 having the combination of the same user ID 421 and topic ID 422 based on the combination of the user ID 411 and topic ID 412 of the user 200 that has been acquired as the topic information 410 from the history storage database 107 (step S501).

[0100] Next, the idea support information presentation unit 102 selects one (for example, the one with the oldest idea date and time 425) from the user idea data 420 acquired in step S501 (step S502), and sends the selected user idea data 420 to the analog recall data selection unit 104.

[0101] Based on the received user idea data 420, the analog recall data selection unit 104 executes an analog recall data selection process (step S503). As detailed with reference to FIG. 12, by executing the analog recall data selection process, a set of thought example data 430 related to the user idea data 420 and thought expression vector data obtained by removing the field components from each of these thought example data 430 are obtained from the thought database 106.

[0102] Next, the idea support information presentation unit 102 checks whether the process of step S503 has been executed for all of the user idea data 420 acquired in step S501 (step S504). If there is unprocessed user idea data 420 remaining (NO in step S504), it returns to step S502, selects one from the unprocessed user idea data 420, and executes the process of step S503 again. If all of the user idea data 420 has been processed in step S504 (YES in step S504), it proceeds to step S505.

[0103] In step S505, the idea support information presentation unit 102 executes a dimensionality reduction process on the thought expression vector data and the user idea vector data after the field components have been removed in the analog recall data selection process of step S503. The dimensionality reduction process refers to a process of compressing multi-dimensional data into data with a smaller number of dimensions, and known methods such as the principal component analysis method and the self-organizing map method can be used.

[0104] Finally, the idea support information presentation unit 102 creates an idea history correlation diagram by projecting the data reduced to two-dimensional values in step S505 onto a graph (step S506), and ends the idea history correlation diagram generation process.

[0105] As described above, by displaying the idea history display screen 340 in the history display mode, the idea support system 100 can visualize the past idea history in a form with high listability. Therefore, the user 200 can review the ideas and consider new ideas.

[0106] As described above, according to the idea support system 100 according to the present embodiment, when a user (a user of this system) has an initial idea (idea), when collecting information for the purpose of developing the idea, information on cases having similarity with the idea can be presented across the field to which the idea belongs (spanning multiple fields). Specifically, the idea support system 100 obtains and presents information in a field different from the field that the user wants to come up with ideas across fields, so as to broaden the user's ideas. In addition, information that can be assumed to cause an analogy of an existing example referred to when the user obtains a new idea from information in other fields is selected and presented. That is, the idea support system 100 according to the present embodiment can present information focused on useful information that contributes to new ideas from the perspective of having an analogy while ensuring the spread of new ideas. Therefore, it can generate an analogy and richly expand the user's ideas.

[0107] (2) Second Embodiment In the second embodiment of the present invention, an idea support system that supports the expansion of ideas by a plurality of users based on drawings will be described by diverting the configuration of the idea support system 100 described in the first embodiment. Further, as one of the differences between the first embodiment and the second embodiment, the information expressing thinking (idea content expression 426) is an image instead of a text.

[0108] FIG. 17 is a diagram schematically showing the overall operation procedure in the second embodiment. Comparing the operation procedure of FIG. 17 with the operation procedure of FIG. 3 shown in the first embodiment, the processes of steps S601 to S604 and S606 in FIG. 17 are the same as steps S101 to S104 and S106 in FIG. 3, and the description thereof is omitted. In terms of the details of each procedure, as described above, the overall operation procedure in the second embodiment is different from the first embodiment in that an image is used as information expressing thoughts. Such detailed differences will be described later.

[0109] In step S105 of FIG. 3, after the repetition of the analogical reminder information presentation process in step S104 is completed, the idea support system 100 determines the final consideration result. That is, the operation procedure shown in FIG. 3 was to support the expansion of ideas by a single user 200. In contrast, in the second embodiment, in order to expand ideas through the cooperation of multiple people, the process of step S605 in FIG. 17 is executed.

[0110] According to FIG. 17, after the repetition of the analogical recollection information presentation process in step S605 is completed, the idea support system 100 (for example, the screen transition control unit 103) adds the user ID of another user (the user selected on the user selection screen 380 in FIG. 22 described later, hereinafter also referred to as the "next user") to the topic information 410 received in step S602, and then stores it in the history storage database 107 (step S605). Alternatively, the idea support system 100 (for example, the screen transition control unit 103) may copy the topic information 410 received in step S602, change the user ID 411 in the copied topic information 410 from the user ID of the current user 200 to the user ID of the next user, and then store it in the history storage database 107. By performing any of the above processes, when the next user logs in to the system, the topic under consideration can be carried over. Furthermore, when the next user logs in to the system and conducts idea consideration, by reading the final user idea data 420 of the current user using the updated topic information 410, the final idea of the current user can be displayed as the current idea in the analogical recollection screen (the current idea display box 361 of the analogical recollection screen 360 in FIG. 19 described later).

[0111] By performing the process of step S605, the idea support system 100 according to this embodiment can entrust the expansion of ideas to other users (next users), and by repeating this, multiple users can cooperate to expand ideas. Specifically, after the ideas of the current user are completed, when another user (next user) logs in to the system, based on the topic information 410 updated in step S605, the topic under consideration can be taken over. Also, when taking over the consideration of ideas, the initial document reception screen presentation process of step S603 (see FIG. 6) may be omitted. In that case, in the first analogy recall information presentation process (step S604, see FIG. 9), by using the updated topic information 410 to read the final user idea data 420 of the previous user, the final idea of the previous user can be displayed as the current idea in the analogy recall screen (the current idea display box 361 of the analogy recall screen 360 in FIG. 19 described later). As a result, the next user can smoothly take over the ideas and perform further expansion. When multiple users conduct idea consideration, by adding a known access right management function to the idea support system 100, the disclosure range of ideas can be appropriately controlled.

[0112] Next, the topic specification screen displayed in step S602 in the second embodiment will be described in detail.

[0113] FIG. 18 is a diagram showing an example of the topic specification screen in the second embodiment. The topic specification screen 350 shown in FIG. 18 has a different display mode from the topic specification screen 310 described with reference to FIG. 4 in the first embodiment, but the generation subject, display timing, etc. are the same as those in the first embodiment.

[0114] In the case of FIG. 18, the topic designation screen 350 includes a drop-down box 351 for designating an existing topic, a decision button 352 to be pressed when the designation in the drop-down box 351 is completed, a text box 353 for entering a topic name when newly designating a topic, drop-down boxes 354 and 355 for designating the field (major classification and minor classification) of the topic, a decision button 356 to be pressed when the designation of the new topic is completed, and a transition button 314 to be pressed when switching to the history display mode.

[0115] The drop-down box 351 and the decision button 352 are interfaces for resuming the consideration of a topic from the middle. Specifically, the user 200 designates an existing topic to be considered in the drop-down box 351 and presses the decision button 352. The topics that can be designated in the drop-down box 351 (i.e., the topics for resuming the consideration from the middle) include not only the topics that the user himself / herself has considered up to the middle in the past, but also the topics under consideration sent from other users.

[0116] The text box 353, the drop-down boxes 354 and 355, and the decision button 356 are interfaces for starting a new consideration of a topic. In the topic designation screen 350 shown in FIG. 18, the drop-down boxes 354 and 355 are provided so that the target field of the topic can be designated step by step by major classification and minor classification. However, the configuration of the topic designation screen according to this embodiment is not limited to this. For example, it may have the same configuration as the topic designation screen 310 shown in FIG. 4 in the first embodiment.

[0117] Also, regarding the major classification, when obtaining a plurality of thought case data 430 from the thought database 106 in the analog recall data selection process (step S403 in FIG. 12), only the field codes corresponding to the major classification that are common to the current topic may be targeted for acquisition. In this case, while restricting the range of the field in the form of the major classification, various thought case data 430 are obtained across a plurality of minor classifications, and a set of thought case data 430 that evokes analogies is selected and can be presented on the analog recall screen. Therefore, it is possible to restrict the selection of thought case data 430 with extremely different fields so that they are not selected.

[0118] FIG. 19 is a diagram showing an example of an analog recall screen in the second embodiment. The analog recall screen 360 shown in FIG. 19 includes a current idea display box 361, an idea correction box 362, an analog recall information display box 363, a decision button 364, and a transition button 365 that is pressed when switching to the history display mode. The functions of these respective components are the same as those of the respective components with the same names in the analog recall screen 330 shown in FIG. 11, except for the points described below.

[0119] In the second embodiment, the image of the drawing is used for the idea expression 426 included in the user idea data 420. Therefore, in the analog recall screen 360, the current idea display box 361 that displays the current idea and the analog recall information display box 363 that displays the set of thought case data 430 obtained in the analog recall data selection process are configured to display images. Also, the idea correction box 362 for correcting the idea has an interface that can be input in the form of an image.

[0120] Also, in the analogy recall screen 360 that is first displayed after the user 200 is changed, the final idea of the previous user (user 200 who passed on the consideration of ideas to the current user) is displayed in the current idea display box 361. Therefore, the current user can carry on the ideas of the previous user and expand new ideas. When a new idea is conceived, information (in this example, a graphic) representing the idea can be input into the idea correction box 362.

[0121] Also, the analogy recall screen 360 is provided with a transition button 365 for the user 200 to shift to the history display mode when desired. However, in the idea support system 100 according to the present embodiment, apart from such manual operation by the user 200, when a predetermined condition is satisfied, such as when the vector data obtained by converting the expression of the idea content of the corrected idea changes significantly from the initial value, the screen transition control unit 103 is configured to automatically switch to the history display mode.

[0122] In the second embodiment, similar to the first embodiment, in the analogy recall data selection process and the idea history correlation diagram generation process, the analogy recall data selection process is executed. The processing procedure of this analogy recall data selection process is the same as the procedure shown in FIG. 12 in the first embodiment. In step S402, the idea encoder 105 performs an embedding process to convert the content of the idea content expression 426 into vector data. Here, in the second embodiment, an image is used for the idea content expression 426. However, as shown in FIG. 20, even for an image, the idea encoder 105 can perform the conversion into vector data.

[0123] FIG. 20 is a diagram schematically explaining the operation of converting an image into vector data by the idea encoder 105.

[0124] As described with reference to FIG. 13, in the first embodiment, the input data to the neural network 513 used the result of decomposing the character string 511 into words and then performing One-Hot encoding. However, in the second embodiment, the idea encoder 105 directly inputs the pixel values of the image 551 stored in the idea content expression 426, and implements embedding using a neural network 552 such as a known CNN (Convolutional Neural Network). As a result, as shown in FIG. 20, a vector sequence 553 is obtained as the output of the neural network 552.

[0125] Next, the history display mode in the second embodiment will be described.

[0126] FIG. 21 is a diagram showing an example of an idea history display screen in the second embodiment. Regarding the configuration common to the idea history display screen 340 shown in FIG. 15 in the first embodiment among the idea history display screen 370 shown in FIG. 21, the description will be omitted.

[0127] In the idea history display screen 370, different from the idea history display screen 340, the displayed idea history (idea content expression) is an image instead of a character string. Also, although the idea history is displayed in chronological order, a button 371 is provided next to the latest idea history for the current user to finish considering the idea and entrust the consideration to the next user. When the button 371 is pressed by the user 200, the idea support information presentation unit 102 displays a user selection screen for designating the user to whom the consideration will be entrusted next.

[0128] FIG. 22 is a diagram showing an example of a user selection screen. The user selection screen 380 shown in FIG. 22 is configured to include an arrangement diagram 381 and a button 382.

[0129] In the layout diagram 381, the relative positional relationship between users is displayed based on the relevance of the past ideas of multiple users using this system. In the layout diagram 381, the star mark represents the position of the current user, and the white circles each represent the positions of other users. These star marks and white circles are displays that can be generated based on the user idea data 420. Specifically, for example, by collecting the idea content expressions 426 of the user idea data 420 for each user and displaying the average value of the vector data obtained by inputting these into the idea encoder 105, the position display of each user can be generated. At this time, by removing the field components in the same way as the above-described analogical recollection data selection process, it is also possible to commission the expansion of ideas to users in other fields.

[0130] When the current user selects one white circle corresponding to the user (next user) to whom consideration will be entrusted next in the layout diagram 381 and presses the button 382, for example, the user idea reception unit 101 duplicates the topic information 410 and the user idea data 420 currently being handled, replaces the user ID in each duplicated data with the user ID of the next user selected in the layout diagram 381, and stores it in the history storage database 107. Thereafter, the idea support system 100 can present the thought example data 430 that supports the expansion of ideas by the next user by performing the operation shown in FIG. 17 with the next user.

[0131] According to the idea support system 100 according to the second embodiment as described above, even when the idea of the user 200 is made in a drawing (image), the contents of the user idea data 420 and the thought example data 430 can be converted into multivariate (vector data) in the same manner as in the first embodiment. Therefore, in order to expand the idea of the user 200, information capable of evoking an analogy can be selected and presented. Furthermore, according to the idea support system 100 according to the second embodiment, the result of the expansion of the idea by the user based on the presentation of information capable of evoking an analogy can be passed on to other users selected by the user. Therefore, it is possible to support the expansion of ideas while multiple people cooperate.

[0132] (3) Third Embodiment In the third embodiment of the present invention, regarding an idea support system that enables setting of idea branches on an idea history display screen, the configuration of the idea support system 100 described in the first or second embodiment will be reused for explanation. In this embodiment, although a mathematical formula is used as an example of information expressing thoughts (idea content expression 426), a sentence (character string) may be used as in the first embodiment, or an image (figure) may be used as in the second embodiment.

[0133] Although a mathematical formula has a different form from a sentence, by using known expression forms such as TeX or MathML (Mathematical Markup Language), it can be treated as a character string, and thus can be treated in the same way as in the first embodiment. In the case of a mathematical formula, the notation of symbols and characters may vary depending on the field. For example, the current may be denoted as "i" in some cases or "j" in others. Or, when similar phenomena occur in different objects, although the variable names and the like are different, there are cases where the equations themselves are similar. Therefore, it can be expected that the idea support system 100 according to this embodiment will also present, on the analogy recall screen, as analogy recall information, mathematical formulas that handle similar phenomena but have different notations according to the conventions for each field as described above.

[0134] Also, the idea support system 100 according to this embodiment can apply such correspondence not only to mathematical formulas but also to various types of information that can be expressed as character strings, such as chemical formulas, program codes, markup data such as HTML, a series of drawing commands for a figure, a network structure such as a neural network, and diagrams such as UML class diagrams, and can be treated as information expressing thoughts (idea content expression).

[0135] Next, the ability to set idea branches on the idea history display screen will be explained in detail.

[0136] FIG. 23 is a diagram showing an example of an idea history display screen in the third embodiment. Regarding the configuration common to the idea history display screen 340 shown in FIG. 15 in the first embodiment among the idea history display screens 390 shown in FIG. 23, the description thereof will be omitted.

[0137] In the idea history display screen 390, as a difference from the idea history display screen 340, the text box 391 at the upper part of the screen for displaying the past idea history is configured to be branchable during the display along the time series. Further, a button 392 labeled "Resume from Here" is provided for each text box 391 indicating a past idea. By pressing the button 392 corresponding to any one of the text boxes 391, the user idea data corresponding to the idea of the selected text box 391 can be used as the immediately preceding idea (original idea), and the consideration of the next idea can be performed in a form of branching from the said idea.

[0138] Here, in order to implement the function of branching ideas as described above in the idea history display screen 390, it is necessary to make the user idea data hold information indicating the idea (original idea) that is the origin of the said idea.

[0139] FIG. 24 is a diagram showing a configuration example of user idea data in the third embodiment. The user idea data 440 shown in FIG. 24 is configured to include items of user ID 421, topic ID 422, field code 423, idea ID 424, and idea date and time 425, similar to the user idea data 420 shown in FIG. 8. The description of each of these items will be omitted. Further, the user idea data 440 has an idea content expression 442 instead of the idea content expression 426 of the user idea data 420. The idea content expression 442 stores a mathematical formula representing the idea input on the initial idea input screen 320 or the analogy recall screen 330.

[0140] Furthermore, the user idea data 440 has an original idea ID 441 that stores the idea ID of the idea that was the basis of this idea, as a configuration different from the user idea data 420. The idea ID stored in this original idea ID 441 is the idea ID 424 in the user idea data 440 of the original idea. More specifically, when the button 392 is pressed on the idea history display screen 390 described above, the value of the idea ID 424 can be obtained from the user idea data 440 corresponding to the idea in the selected text box 391.

[0141] By using the user idea data 440 having such an original idea ID 441, the idea support information presentation unit 102 can, on the idea history display screen 390 shown in FIG. 23, not only simply arrange the past idea history (user idea data 440) in chronological order, but also add and draw it as a branch when there is another user idea data 440 that can be referenced by the original idea ID.

[0142] Also, such a branch of the idea can be reflected in the idea history correlation diagram displayed in the model display box 393 on the idea history display screen 390. For example, in the idea history correlation diagram of FIG. 23, the arrowed curve connecting the user idea data 420 related to the target topic in chronological order branches in two directions starting from the star mark, which represents that the idea has branched into two from the currently selected idea history (user idea data 440). By performing such a display, it is possible to easily recognize the transition and branch of the idea, and thus realize highly listable information presentation.

[0143] As described above, according to the idea support system 100 according to the present embodiment, even if the idea content expression is a mathematical formula, it is possible to present an example that guides the user to expand the idea based on analogy, similar to the first or second embodiment. Furthermore, according to the idea support system 100 according to the present embodiment, by visualizing the branch of the idea, it becomes easier for the user to look back on their own ideas and, in some cases, to retry the ideas from a certain point in the past, so that it is possible to support the user in expanding their ideas more flexibly.

[0144] Note that the present invention is not limited to the above-described embodiments and includes various modifications. It is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. Further, it is possible to add, delete, or replace a part of the configuration of each embodiment with another configuration. For example, regarding the "expression of the idea content" of the user idea data that holds the content of the user's idea, it is represented as text (character string) in the first embodiment, as an image (graphic) in the second embodiment, and as a mathematical formula (or other notations that can be treated as a character string by using a predetermined expression format) in the third embodiment. However, these are not limited to each embodiment and can be replaced or combined.

Explanation of Signs

[0145] 11 Processor 12 DRAM 13 Storage device 14 Input device 15 Monitor 16 Communication interface 100 Idea support system 101 User idea reception unit 102 Idea support information presentation unit 103 Screen transition control unit 104 Analogy recollection data selection unit 105 Idea encoder 106 Thought database 107 History storage database 108 User authentication unit 200 User 310, 350 Topic specification screen 320 Initial idea input screen 330, 360 Analogy recollection screen 340, 370, 390 Idea history display screen 380 User selection screen 410 Topic information 420, 420 User idea data 430 Thinking example data

Claims

1. A thought database that records thought data obtained by digitizing existing thoughts, in association with a field code indicating the corresponding field of the thought; An input unit that receives user thought data, which is thought data obtained by digitizing the user's thoughts; An idea encoder that converts the thought data and the user thought data into multivariate data respectively; A data selection unit that selects analogy evoking information capable of evoking an analogy for the user, based on a first multivariate data obtained by converting the user thought data by the idea encoder and a second multivariate data obtained by converting the thought data recorded in the thought database by the idea encoder; Comprising: The conversion by the idea encoder has a feature that in the multivariate data obtained by converting a plurality of thought data, the distance between the multivariate data correlates with the similarity of the thought data before conversion; The data selection unit analyzes the relevance between the components of the multivariate data and the field code for each of the first multivariate data and the second multivariate data converted by the idea encoder, removes the components affected by the field code from the multivariate data, and selects the analogy evoking information from the thought data recorded in the thought database based on the similarity indicated by the distance between the first multivariate data and the second multivariate data after the removal; An idea support system characterized by the above.

2. An information presentation unit that presents the analogy evoking information selected by the data selection unit to the user; A screen transition control unit that controls the switching between information input from the user and information presentation to the user; Further comprising: When the input unit receives the user thought data representing the changed thought input by the user after the user's thought has changed in response to the presentation of the analogy evoking information by the information presentation unit; The screen transition control unit controls to repeatedly execute the information presentation by the information presentation unit and the reception of the user thought data by the input unit. The idea support system according to Claim 1, characterized by the above.

3. Further comprising a history storage database that stores the records of the information presentation by the information presentation unit and the reception of the user thought data by the input unit as history information; The information presentation unit displays the history information stored in the history storage database in chronological order on a history display screen. The idea support system according to claim 2, characterized in that

4. configured to be capable of repeatedly executing information presentation by the information presentation unit and reception of user thinking data by the input unit for each user while taking over the user, the screen transition control unit controls to perform information presentation by the information presentation unit for the user after takeover using the history information by the user stored in the history storage database The idea support system according to claim 3, characterized in that

5. the information presentation unit generates and presents a correlation diagram in which the positional relationship between the user thinking data and the analogical recall information can be visually recognized based on the first multivariate and the second multivariate converted by the idea encoder The idea support system according to claim 3, characterized in that

6. as a result of calculating the distance between the first multivariate and the second multivariate after removing the components affected by the field code by the data selection unit, the thinking data corresponding to the second multivariate whose distance from the first multivariate is equal to or less than a predetermined threshold is selected as the analogical recall information, the information presentation unit displays the predetermined threshold on the correlation diagram The idea support system according to claim 5, characterized in that

7. the input unit receives an input for changing the predetermined threshold from the user The idea support system according to claim 6, characterized in that

8. the user thinking data is associated with the user thinking data that is the source of the thinking of the user thinking data, the information presentation unit displays the context of thinking for each of the user thinking data shown as the history information on the history display screen, the input unit enables the user to receive user thinking data representing thinking branched and developed from any of the user thinking data displayed on the history display screen The idea support system according to claim 3, characterized in that

9. the idea encoder performs conversion to the multivariate using a pre-trained machine learning model The idea support system according to claim 1, characterized in that

10. the thinking data is data in which thinking is informatized using a character string, the idea encoder has a function of converting a character string into a multivariate The idea support system according to claim 1, characterized in that

11. the character string is a mathematical formula The idea support system according to claim 10, characterized in that...

12. The thinking data is data in which thinking is informatized using images, The idea encoder has a function of converting an image into multiple variables The idea support system according to claim 1, characterized in that...

13. An idea support method by an idea support system for supporting the development of a user's thinking, wherein the idea support system includes a thinking database that records thinking data obtained by informatizing existing thinking, in association with a field code indicating the corresponding field of the thinking, an input unit that receives user thinking data, which is thinking data obtained by informatizing a user's thinking, an idea encoder that converts the thinking data and the user thinking data into multiple variables, a data selection unit that selects analogy reminder information that can cause a user to recall an analogy based on a first multiple variable obtained by converting the user thinking data by the idea encoder and a second multiple variable obtained by converting the thinking data recorded in the thinking database by the idea encoder, and has the conversion by the idea encoder has a feature that in the multiple variables obtained by converting a plurality of thinking data, the distance between the multiple variables is correlated with the similarity of the thinking data before conversion, the data selection unit analyzes the relevance between the components of the multiple variables and the field code for each of the first multiple variable and the second multiple variable converted by the idea encoder, removes the components affected by the field code from the multiple variables, and selects the analogy reminder information from the thinking data recorded in the thinking database based on the similarity indicated by the distance between the first multiple variable and the second multiple variable after removal The idea support method is characterized in that...

Images