Information processing device, information processing system, information processing method, and recording medium

Abstract

The present invention helps a user understand a relationship between designated elements. This information processing device includes an acquisition unit, a reception unit, an instruction unit, and an output unit. The acquisition unit acquires information indicating the strength of a relationship between a plurality of elements. The reception unit receives designation of one or more elements among the plurality of elements. The instruction unit issues an instruction to analyze, using a generation model, a relationship between designated one or more elements, on the basis of the information indicating the strength of the relationship between the plurality of elements. The output unit outputs information expressing the relationship between the one or more elements in a language, on the basis of the result of the analysis by the generation model.

Description

Information processing device, information processing system, information processing method, and recording medium

[0001] The present disclosure relates to an information processing device, an information processing system, an information processing method, and a recording medium, and more particularly to an information processing device, an information processing system, an information processing method, and a recording medium that provide a user with an analysis result of information.

[0002] In technical fields such as behavior prediction and anomaly detection, related techniques for analyzing causal relationships between elements are known.

[0003] Patent Literature 1 discloses a technology for visualizing how a change in a certain KPI (Key Performance Indicator) in business data has affected a KPI used for business evaluation. The related technology described in Patent Literature 1 expresses a KPI causal model using a graph structure that shows the causal relationships between KPIs.

[0004] International Publication No. 2015 / 193983

[0005] In the related technology, the graph structure becomes more complex as the number of KPIs increases, which results in a problem in that it becomes difficult for users to interpret the graph structure and understand the information that is necessary for them.

[0006] The present disclosure has been made in view of the above-mentioned problems, and its purpose is to help a user understand the relationships between elements that they specify.

[0007] An information processing device according to one aspect of the present disclosure includes an acquisition means for acquiring information representing the strength of the relationship between multiple elements, a reception means for receiving the designation of one or more elements from the multiple elements, an instruction means for instructing the analysis of the relationship between the one or more designated elements and other elements using a generative model based on the information representing the strength of the relationship between the multiple elements, and an output means for outputting information expressing in language the relationship between the one or more designated elements and other elements based on the analysis results using the generative model.

[0008] An information processing system according to one aspect of the present disclosure comprises an information processing device, one or more user terminals used by a user to input information and instructions to the information processing device, and an information analysis execution device trained to contextualize the input data and output output data according to the context of the input data, wherein the information processing device acquires information representing the strength of the relationship between multiple elements, accepts designation of one or more elements among the multiple elements, instructs the use of a generative model to analyze the relationship between the designated one or more elements and other elements based on the information representing the strength of the relationship between the multiple elements, and outputs information expressing in language the relationship between the one or more elements and other elements based on the analysis results by the generative model.

[0009] In an information processing method according to one aspect of the present disclosure, a computer acquires information representing the strength of the relationship between multiple elements, accepts the designation of one or more elements among the multiple elements, instructs the computer to use a generative model to analyze the relationship between the specified one or more elements and other elements based on the information representing the strength of the relationship between the multiple elements, and outputs information expressing in language the relationship between the one or more elements and other elements based on the analysis results using the generative model.

[0010] A recording medium according to one embodiment of the present disclosure stores a program for causing a computer to execute the following processes: acquiring information representing the strength of the relationship between multiple elements; accepting the designation of one or more elements from the multiple elements; instructing the use of a generative model to analyze the relationship between the one or more specified elements and other elements based on the information representing the strength of the relationship between the multiple elements; and outputting information expressing in language the relationship between the one or more elements and other elements based on the analysis results using the generative model.

[0011] According to one aspect of the present disclosure, it is possible to help a user understand the relationship between elements that they specify.

[0012] 1 is a conceptual diagram showing an example of the operation of a system including an information processing device according to an embodiment. FIG. 1 is a diagram showing an example of input data (information representing the strength of relationships between multiple elements) input to a generative model in a system including an information processing device according to an embodiment. FIG. 2 is a diagram showing an example of output data (information based on the analysis results by the generative model) output from the system. FIG. 3 is a diagram showing an example of output data (discussion screen) output by an information processing device according to an embodiment. FIG. 4 is a block diagram showing the configuration of an information processing device according to an embodiment. FIG. 5 is a flowchart showing the operation of an information processing device according to an embodiment. FIG. 6 is a diagram showing an example of a reception screen for accepting the designation of one or more elements from multiple elements. FIG. 7 is a diagram showing another example of a reception screen for accepting the designation of one or more elements from multiple elements. FIG. 8 is a diagram showing an example of a causal graph representing a causal relationship between one designated (selected) element and another element from multiple elements. FIG. 9 is a diagram showing an example of a causal graph representing a causal relationship between two designated (selected) elements and another element from multiple elements. FIG. 10 is a block diagram showing the configuration of an information processing device according to an embodiment. FIG. 11 is a flowchart showing the operation of an information processing device according to an embodiment. FIG. 12 is a conceptual diagram showing an example of the operation of a system including an information processing device according to an embodiment. FIG. 13 is a block diagram showing the configuration of an information processing device according to an embodiment. FIG. 14 is a flowchart showing the operation of an information processing device according to an embodiment. FIG. 15 is a diagram showing an example of an information processing system according to an embodiment.

[0013] Some embodiments of the present disclosure will be described with reference to the drawings.

[0014] First Embodiment A first embodiment of the present disclosure will be described with reference to FIGS. 1 to 6. FIG.

[0015] (Example of System) Fig. 1 is a conceptual diagram showing an example of the operation of the system according to embodiment 1. As shown in Fig. 1, the system acquires information indicating the strength of the relationship between a plurality of elements (S1).

[0016] For example, the information representing the strength of the relationships between the multiple elements is a correlation coefficient matrix composed of correlation coefficients representing the correlations between the multiple elements. In another example, the information representing the strength of the relationships between the multiple elements is a causal graph ( FIG. 8 ) representing the causal relationships between the multiple elements. The causal graph ( FIG. 8 ) is an example of chart data.

[0017] As shown in FIG. 1, the user inputs instructions to the user terminal 100 to select which of the multiple elements the relationship between which should be analyzed by the generative model (e.g., selecting elements A and B, and selecting elements E and F) (S2).

[0018] An information processing device 10 (FIG. 5) described below inputs information indicating the strength of the relationship between elements designated by the user into the generative model in accordance with instructions from the user terminal 100 (S3).

[0019] A generative model is an information analysis program that is trained, typically through machine learning or deep learning, to contextualize input data and output data that corresponds to the context of the input data. A generative model analyzes information that represents the strength of relationships between multiple elements and outputs the analysis results.

[0020] More specifically, a generative model is a model that learns the relationships between words in a sentence (including letters, symbols, and numbers) and the relationships between elements shown in a diagram, and generates related strings related to a target string from the target string. By using a generative model that has been trained on sentences and paragraphs from various contexts, it is possible to generate related strings with appropriate content related to the target string.

[0021] For example, a case where a generative model is used in question answering will be described. The generative model receives an input question such as "What kind of country is Japan?" as a target string. The generative model generates a string such as "Japan is an island country in the Northern Hemisphere..." as an answer to the question.

[0022] The method of training the generative model is not particularly limited, but as an example, the generative model may be trained to output at least one sentence that includes an input string.

[0023] As a specific example, a generative model is a Generative Pre-trained Transformer (GPT) that outputs a sentence including an input string by predicting a string that is likely to follow the input string. Other generative models include a Text-to-Text Transfer Transformer (T5), a Bidirectional Encoder Representations from Transformers (BERT), a Robustly optimized BERT approach (RoBERTa), and an Efficiently Learning an Encoder that Classifies Token Replacements Accurately (ELECTRA).

[0024] Furthermore, the character strings generated by the generative model are not limited to natural languages. For example, the generative model may output an artificial language (such as program source code) in response to a character string input in a natural language. For example, the generative model may receive an input question such as, "How do I retrieve data containing a specific character string from a database?" as a target character string. The generative model may output program source code for performing database processing. Alternatively, the generative model may output a natural language corresponding to a character string input in an artificial language.

[0025] Furthermore, the content generated by a generative model is not limited to character strings. For example, the generative model may generate image data, video data, audio data, or other data formats corresponding to the input data.

[0026] In one example, the information input to the generative model is a correlation coefficient matrix that includes only elements designated (selected) by the user, or a correlation coefficient matrix in which all elements not designated (selected) by the user are set to 0.

[0027] The generative model analyzes information indicating the strength of the relationships between multiple elements specified by the user, and outputs the analysis results (S4).

[0028] An information processing device 10 (FIG. 5) described below creates information based on the analysis results of the input data, based on the analysis results generated by the generative model.

[0029] The system outputs information based on the analysis results of the generative model (S5).

[0030] For example, the output data may be comments about the relationships between the elements (FIG. 3). In one example, the comments may include at least one of facts, observations, and suggestions (e.g., policy proposals) derived from the relationships between the elements.

[0031] (Example of Input Data) Figure 2 shows an example of input data input to a generative model. In the example shown in Figure 2, the strength of the relationship between multiple elements (e.g., "livability" and "medical and nursing care system") is expressed as a numerical value. Specifically, these numerical values ​​are correlation coefficients between one element (variable) and another element.

[0032] (Example of Output Data) FIG. 3 shows an example of output data created based on the analysis results of the generative model. In the example shown in FIG. 3, comments about the relationship between multiple elements are shown. For example, the example shown in FIG. 3 states that "the size of a house itself is not likely to be directly related to the motivation for purchasing a house." This comment explains (considers) the magnitude of the influence that one element (variable) called "house size" has on another element called "motivation for purchasing a house."

[0033] (Another example of output data: consideration screen) Figure 4 shows an example of information based on the analysis results using a generative model. This example is a consideration screen that allows the user to view the analysis results using a generative model. The consideration screen is displayed on the user terminal 100 (Figure 1) or the like.

[0034] As shown in FIG. 4, the consideration screen includes a causal graph (an example of information indicating the strength of the causal relationships between multiple elements) and analysis results (e.g., facts, considerations, and conclusions) from the generative model. Furthermore, a comment field is also shown below these. In addition to inputting the analysis results from the generative model into the comment field, the user can also enter their own opinions in the comment field using the user terminal 100.

[0035] By viewing the consideration screen shown in Figure 4, the user can confirm the analysis results of the generative model and can also directly understand information representing the strength of the causal relationships between multiple elements from the causal graph.

[0036] (Configuration of Information Processing Apparatus 10) The configuration of the information processing apparatus 10 according to the first embodiment will be described with reference to Fig. 5. Fig. 5 is a block diagram showing the configuration of the information processing apparatus 10.

[0037] 5, the information processing device 10 includes an acquisition unit 11, a reception unit 12, an instruction unit 13, and an output unit 14. Note that an information processing device 20 according to a second embodiment, which will be described later, also has the same configuration as the information processing device 10 described here.

[0038] The acquisition unit 11 acquires information representing the strength of the relationship between a plurality of elements. The acquisition unit 11 is an example of an acquisition means. The information representing the strength of the relationship between a plurality of elements may be numerical data such as a vector or a matrix, or may be diagram data such as a graph or a diagram.

[0039] For example, the acquisition unit 11 acquires the results of a causal analysis performed on information about a plurality of elements from a causal analysis device (not shown). The results of the causal analysis indicate whether or not one or more other elements (causes) have an effect on one element (result), and the magnitude of the effect. In this example, the results of the causal analysis performed by the causal analysis device are information representing the strength of the causal relationship between the plurality of elements (for example, a coefficient or matrix representing the strength of the causal relationship between the plurality of elements).

[0040] Alternatively, the acquisition unit 11 may acquire information that indicates the strength of the relationship between a plurality of elements, which information is input to the information processing device 10 by the user.

[0041] The acquisition unit 11 outputs information indicating the strength of the relationship between a plurality of elements to the instruction unit 13 .

[0042] The receiving unit 12 receives the designation of one or more elements from among the plurality of elements. The receiving unit 12 is an example of a receiving means.

[0043] For example, the reception unit 12 receives an instruction from the user terminal 100 ( FIG. 1 ) to specify (select) elements to be analyzed by the generative model. In one example, the reception unit 12 accepts the specification of one or more elements on a reception screen ( FIGS. 7 and 8 ) that displays information indicating the strength of the relationships between multiple elements. A specific example of the reception screen will be described in embodiment 2.

[0044] In addition, after information based on the analysis results using the generative model is output, the reception unit 12 may accept the designation of one or more specified elements to be targeted for reanalysis using the generative model (embodiment 2).

[0045] The receiving unit 12 outputs information indicating the element designated by the user to the instruction unit 13 .

[0046] The instruction unit 13 instructs the system to analyze the relationships between one or more specified elements and other elements using a generative model, based on information indicating the strength of the relationships between the elements. The instruction unit 13 is an example of an instruction means.

[0047] For example, the instruction unit 13 receives information indicating the strength of the relationship between multiple elements from the acquisition unit 11. The instruction unit 13 also receives information indicating one or multiple elements designated by the user from the reception unit 12.

[0048] The instruction unit 13 inputs information representing the strength of relationships between multiple elements to the generative model ( FIG. 1 ). In one example, the instruction unit 13 inputs the information representing the strength of relationships between multiple elements to the generative model in the form of numerical values ​​such as vectors or matrices. In another example, the instruction unit 13 inputs the information representing the strength of relationships between multiple elements to the generative model in the form of images such as graphs or diagrams.

[0049] The instruction unit 13 then instructs the system to use a generative model to analyze the relationships between one or more elements specified by the user and other elements, based on information indicating the strength of the relationships between the multiple elements. At this time, the instruction unit 13 may accept input of a prompt (instruction statement) from the user. The generative model analyzes the relationships between the specified one or more elements and other elements in accordance with the instructions from the instruction unit 13 and the prompt, and outputs the analysis results.

[0050] Thereafter, the instruction unit 13 notifies the output unit 14 that the generative model has been caused to analyze the information.

[0051] The output unit 14 outputs information that expresses in language the relationship between one or more specified elements and other elements based on the analysis results (e.g., facts, considerations, and conclusions) by the generative model. The output unit 14 is an example of an output means.

[0052] For example, after receiving a notification from the instruction unit 13, the output unit 14 acquires an analysis result of the relationship between one or more elements and other elements from the generative model. The output unit 14 outputs information based on the acquired analysis result.

[0053] In one example, the output unit 14 displays, on the user terminal 100 (FIG. 1), a discussion screen (FIG. 4) on which the user can view the analysis results obtained by the generative model. The discussion screen may include input data (e.g., information indicating the strength of the relationship between multiple elements, or information indicating one or multiple elements specified by the user) along with the analysis results obtained by the generative model. The discussion screen may also include comments entered by the user (e.g., the user's opinion on the analysis results obtained by the generative model).

[0054] (Operation of Information Processing Apparatus 10) The operation of the information processing apparatus 10 according to the first embodiment will be described with reference to Fig. 6. Fig. 6 is a flowchart showing the operation of the information processing apparatus 10.

[0055] As shown in FIG. 6, first, the acquisition unit 11 acquires information indicating the strength of the relationship between a plurality of elements (S101).

[0056] The acquisition unit 11 outputs information indicating the strength of the relationship between a plurality of elements to the instruction unit 13 .

[0057] Next, the receiving unit 12 receives designation of one or more elements from the plurality of elements from the user terminal 100 (FIG. 1) (S102).

[0058] The receiving unit 12 outputs information indicating the element designated by the user to the instruction unit 13 .

[0059] Next, the instruction unit 13 instructs the generative model (FIG. 1) to analyze one or more elements specified by the user using the generative model (S103).

[0060] Thereafter, the instruction unit 13 notifies the output unit 14 that the generative model has been caused to analyze the information.

[0061] Next, the output unit 14 outputs information that expresses in language the relationship between the one or more elements specified by the user and other elements based on the results of the generative model's analysis of the relationship between the one or more elements specified by the user and other elements (S104).

[0062] For example, the output unit 14 displays on the user terminal 100 a consideration screen (FIG. 4) on which the user can view the analysis results based on the generative model.

[0063] This completes the operation of the information processing device 10 according to the first embodiment.

[0064] (Effects of this embodiment) According to the configuration of this embodiment, the acquisition unit 11 acquires information representing the strength of relationships between multiple elements. The reception unit 12 receives designation of one or more elements from the multiple elements. The instruction unit 13 instructs, based on the information representing the strength of relationships between the multiple elements, to analyze the relationships between the designated one or more elements and other elements using a generative model. The output unit 14 outputs information expressing, in language, the relationships between the one or more elements and other elements based on the analysis results using the generative model.

[0065] In this way, information that expresses in words the relationship between one or more elements specified by the user and other elements is output, rather than information that directly indicates the strength of the relationship between multiple elements, thereby helping the user to understand the relationship between the elements specified.

[0066] [Embodiment 2] Embodiment 2 will be described with reference to Figures 7 to 10. In this embodiment 2, a specific example of a reception screen that receives, from a user, the designation (selection) of one or more elements to be analyzed by the generative model (Figure 1) will be described.

[0067] The configuration of the information processing device 20 according to the second embodiment is the same as that of the information processing device 10 ( FIG. 5 ) described in the first embodiment. As in the first embodiment, the reception unit 12 ( FIG. 5 ) of the information processing device 20 receives designation of one or more elements from among the multiple elements. At this time, the reception unit 12 presents a reception screen exemplified below to receive, from the user, designation (selection) of one or more elements to be analyzed by the generative model ( FIG. 1 ).

[0068] 7 is a diagram showing an example of a reception screen for receiving designation of one or more elements among the plurality of elements. In one example, the reception screen is displayed on the user terminal 100 (FIG. 1).

[0069] In the example shown in FIG. 7 , one element, "Number of Children," and another element, "Size of House," are specified (selected) from among multiple elements in a pull-down format. Below the pull-down, information based on the results of a generative model's analysis of the relationship between the elements specified by the user (here, "Number of Children" and "Size of House") is displayed. Alternatively, the user may be able to manually input one or more elements to specify into the reception screen.

[0070] For example, based on the analysis results of the generative model, a comment is written that "As the family size increases, the living space required increases, so the size of the house may also increase." This comment explains the relationship between one element, "the number of children," and another element, "the size of the house."

[0071] (Another Example of Reception Screen) Fig. 8 shows another example of the reception screen. The reception screen shown in Fig. 8 shows a causal graph. The causal graph is made up of a plurality of nodes and arrows connecting the nodes.

[0072] As shown in Figure 8, in a causal graph, one node is connected to one or more other nodes by arrows. Each node represents a result or a cause in a causal relationship. Arrows connect nodes in a causal relationship. The node at the end of an arrow is the result, and the node at the start of an arrow is the cause.

[0073] In this example, the node "Motivation for Purchasing a Home" is the final result, that is, the "Objective Variable." The nodes other than the node "Motivation for Purchasing a Home" are the "Explanatory Variables."

[0074] The numbers attached to the lines connecting nodes are correlation coefficients that represent the strength of the relationship between the nodes. If the number between nodes is "0", there is no relationship between those nodes. On the other hand, if the number between nodes is not "0", there is a relationship between those nodes. The strength of the relationship between nodes is compared based on the size of the numbers.

[0075] For example, the node "Number of Children" has a relationship strength of 0.28 with the node "House Size." On the other hand, the node "Salary" has a relationship strength of 0.29 with the node "House Size." Therefore, it can be seen that "House Size" is more influenced by "Salary" than "Number of Children."

[0076] Note that solid arrows indicate a positive effect on "Motivation for Purchasing a Home." On the other hand, dashed arrows indicate a negative effect on "Motivation for Purchasing a Home." For example, the node "Number of Children" has a strength relationship with "Motivation for Purchasing a Home" of plus "0.28." On the other hand, the node "Price (of Real Estate)" has a strength relationship with "Motivation for Purchasing a Home" of minus "0.4."

[0077] The generative model (FIG. 1) may be input with the causal graph itself, as in this example (FIG. 8), or with a matrix whose elements are the numerical values ​​(coefficients) between nodes in the causal graph.

[0078] The user performs a user operation on the reception screen shown in Fig. 8 to specify (select) one or more elements. Specifically, the user selects one or more nodes on the reception screen. Alternatively, the user can select an arrow line connecting nodes on the reception screen. In the latter case, the two nodes connected by the selected arrow line are selected.

[0079] FIG. 9 illustrates a case where a user selects one node, "house size," on the reception screen illustrated in FIG. 8 . In this case, as illustrated in FIG. 9 , only information (solid or dashed arrows) indicating the strength of the relationship between "house size" and the objective variable, "house purchase motivation," is displayed or highlighted. In this example, along with "house size" and "house purchase motivation," the node "price," which is related to both "house size" and "house purchase motivation," is also displayed. The correlation coefficient between the node "house size" and the node "price" is plus "0.28." The correlation coefficient between the node "price" and the node "house purchase motivation" is minus "0.4."

[0080] Fig. 10 shows a case where a user selects two nodes, "house size" and "salary," on the reception screen shown in Fig. 8. In this case, as shown in Fig. 10, in addition to information (solid or dashed arrow lines) indicating the strength of the relationship between "house size" or "salary" and the objective variable, "motive for purchasing a house," the node "price" related to "house size" and "motive for purchasing a house" is displayed or highlighted.

[0081] The correlation coefficient between "salary" and "price" is plus "0.37", and the correlation coefficient between "salary" and "house size" is plus "0.29".

[0082] Alternatively, if a user selects two nodes, "house size" and "salary," on the reception screen shown in FIG. 8, the objective variable, "reason for purchasing a house," may not be displayed, and only information related to at least one of "house size" and "salary" may be displayed.

[0083] (Effects of this embodiment) According to the configuration of this embodiment, the acquisition unit 11 acquires information representing the strength of relationships between multiple elements. The reception unit 12 receives designation of one or more elements from the multiple elements. The instruction unit 13 instructs, based on the information representing the strength of relationships between the multiple elements, to analyze the relationships between the designated one or more elements and other elements using a generative model. The output unit 14 outputs information expressing, in language, the relationships between the one or more elements and other elements based on the analysis results using the generative model.

[0084] In this way, information that expresses in words the relationship between one or more elements specified by the user and other elements is output, rather than information that directly indicates the strength of the relationship between multiple elements, thereby helping the user to understand the relationship between the elements specified.

[0085] 11 and 12, a third embodiment will be described. In the third embodiment, a configuration will be described in which, after information based on the analysis results using a generative model is output, a user operation for narrowing down the target of analysis using the generative model is accepted.

[0086] In the third embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, and the description thereof will be omitted.

[0087] (Configuration of Information Processing Device 30) The configuration of the information processing device 30 according to the third embodiment will be described with reference to Fig. 11. Fig. 11 is a block diagram showing the configuration of the information processing device 30.

[0088] As shown in FIG. 11, the information processing device 30 includes an acquisition unit 11, a reception unit 12′, an instruction unit 13, and an output unit 14.

[0089] The acquisition unit 11 acquires information indicating the strength of the relationship between a plurality of elements, and outputs the information indicating the strength of the relationship between the plurality of elements to the instruction unit 13.

[0090] The receiving unit 12′ receives the designation of one or more elements from among the plurality of elements, and outputs information indicating the elements designated by the user to the instruction unit 13.

[0091] The instruction unit 13 instructs the generative model to analyze the relationships between one or more specified elements and other elements based on information indicating the strength of the relationships between the multiple elements. The instruction unit 13 notifies the output unit 14 that the generative model has executed the information analysis.

[0092] The output unit 14 outputs information expressing in language the relationship between one or more elements and other elements based on the analysis result of the generative model. In one example, the output unit 14 displays the consideration screen (FIG. 4) shown in the first embodiment on the user terminal 100 (FIG. 1).

[0093] After the output unit 14 outputs information based on the analysis results using the generative model, the reception unit 12′ receives a user operation for narrowing down the target of analysis using the generative model. In one example, the reception unit 12′ displays the reception screen (FIGS. 7 and 8) described in the second embodiment on the user terminal 100.

[0094] When a user performs an operation to narrow down the target of analysis using a generative model, the receiving unit 12 ′ outputs, to the instruction unit 13 , information indicating the elements narrowed down by the user.

[0095] (Operation of Information Processing Device 30) The operation of the information processing device 30 according to the third embodiment will be described with reference to Fig. 12. Fig. 12 is a flowchart showing the operation of the information processing device 30.

[0096] As shown in FIG. 12, first, the acquisition unit 11 acquires information indicating the strength of the relationship between a plurality of elements (S301).

[0097] The acquisition unit 11 outputs information indicating the strength of the relationship between a plurality of elements to the instruction unit 13 .

[0098] Next, the receiving unit 12' receives the designation of one or more elements from the plurality of elements from the user terminal 100 (FIG. 1) (S302).

[0099] The receiving unit 12 ′ outputs information indicating the element designated by the user to the instruction unit 13 .

[0100] Next, the instruction unit 13 instructs the generative model (FIG. 1) to analyze one or more elements specified by the user using the generative model (S303).

[0101] Thereafter, the instruction unit 13 notifies the output unit 14 that the generative model has been caused to analyze the information.

[0102] Next, the output unit 14 outputs information that expresses in language the relationship between the one or more elements specified by the user and other elements based on the results of the generative model's analysis of the relationship between the one or more elements specified by the user and other elements (S304).

[0103] For example, the output unit 14 displays on the user terminal 100 a consideration screen (FIG. 4) on which the user can view the analysis results based on the generative model.

[0104] Thereafter, the reception unit 12′ receives a user operation for narrowing down the target of analysis using the generative model (S305). For example, the reception unit 12′ displays a reception screen (FIGS. 7 and 8) on the user terminal 100 (FIG. 1) for receiving designation of elements to be reanalyzed.

[0105] If a user operation has been performed to narrow down the target of analysis using the generative model (Yes in S306), the receiving unit 12′ outputs information indicating the elements narrowed down by the user to the instruction unit 13. Then, the flow returns to step S303.

[0106] On the other hand, if the user does not perform an operation to narrow down the target of analysis using the generative model (No in S306), the operation of the information processing device 30 according to the third embodiment ends.

[0107] (Effects of this embodiment) According to the configuration of this embodiment, the acquisition unit 11 acquires information representing the strength of relationships between multiple elements. The reception unit 12′ receives designation of one or more elements from the multiple elements. The instruction unit 13 instructs, based on the information representing the strength of relationships between the multiple elements, to analyze the relationships between the designated one or more elements and other elements using a generative model. The output unit 14 outputs information expressing, in language, the relationships between the one or more elements and other elements based on the analysis results using the generative model.

[0108] In this way, information that expresses in words the relationship between one or more elements specified by the user and other elements is output, rather than information that directly indicates the strength of the relationship between multiple elements, thereby helping the user to understand the relationship between the elements specified.

[0109] Furthermore, according to the configuration of this embodiment, after information based on the analysis results using the generative model is output, the reception unit 12′ receives a designation of one or more elements to be targeted for reanalysis using the generative model.

[0110] The instruction unit 13 inputs information indicating the strength of the relationship between the elements narrowed down by the user to the generative model. The output unit 14 outputs information expressing in language the relationship between the elements narrowed down by the user based on the analysis results by the generative model.

[0111] This allows only the information that the user needs to be presented in a format that is easy for the user to understand.

[0112] 13 to 15, a fourth embodiment of the present disclosure will be described. In the fourth embodiment, a configuration will be described in which information indicating the strength of the relationship between a plurality of elements is created from information about the plurality of elements.

[0113] (Example of System) Fig. 13 is a conceptual diagram showing an example of the operation of the system according to embodiment 4. As shown in Fig. 13, information on a plurality of elements (for example, questionnaire results on a plurality of elements) is input to the system.

[0114] The system performs a causal analysis on information about multiple elements, and as a result of the causal analysis, information indicating the strength of the causal relationships between multiple elements (for example, a causal graph (FIG. 8) or a matrix representing the causal relationships) is obtained.

[0115] As shown in FIG. 13, information representing the strength of relationships between multiple elements is input to a generative model (for example, a large-scale language model).

[0116] For example, the input data may be a correlation coefficient matrix composed of correlation coefficients representing correlations between multiple elements. In another example, the input data may be a causal graph (FIG. 8) representing causal relationships between multiple elements.

[0117] The generative model analyzes information that represents the strength of the relationships between multiple elements and outputs the analysis results.

[0118] An information processing device 40 (FIG. 14) described later creates information based on the analysis results output by the generative model.

[0119] As shown in FIG. 13, information based on the analysis results using the generative model is output.

[0120] For example, the output data may be comments about the relationships between the elements. In one example, the comments may include at least one of facts, observations, and suggestions (e.g., policy proposals) derived from the relationships between the elements.

[0121] (Configuration of Information Processing Device 40) The configuration of the information processing device 40 according to the fourth embodiment will be described with reference to Fig. 15. Fig. 15 is a block diagram showing the configuration of the information processing device 40.

[0122] As shown in FIG. 15, the information processing device 40 includes an acquisition unit 41, a reception unit 42, an instruction unit 43, and an output unit 44.

[0123] The acquisition unit 41 performs a causal analysis on information relating to a plurality of elements.

[0124] For example, the acquisition unit 41 acquires information about a plurality of elements from information collected in advance in a database (not shown). The acquisition unit 41 performs a causal analysis on the information about the plurality of elements to acquire a coefficient representing the strength of the causal relationship between the plurality of elements.

[0125] The acquisition unit 41 outputs to the instruction unit 43 information indicating the strength of the causal relationship between the plurality of elements, which information is obtained by performing a causal analysis on the information regarding the plurality of elements.

[0126] The receiving unit 42 receives the designation of one or more elements from among the plurality of elements. The receiving unit 42 is an example of a receiving means.

[0127] For example, the reception unit 42 receives an instruction to specify (select) elements to be analyzed by the generative model from the user terminal 100 ( FIG. 1 ). In one example, the reception unit 42 receives the specification of one or more elements on a reception screen ( FIGS. 7 and 8 ) that displays information indicating the strength of the relationships between multiple elements.

[0128] In addition, similar to the reception unit 12′ described in the third embodiment, the reception unit 42 may receive a designation of one or more specified elements to be subject to reanalysis by the generative model after information based on the analysis results by the generative model is output.

[0129] The receiving unit 42 outputs information indicating the element designated by the user to the instruction unit 43 .

[0130] The instruction unit 43 inputs a matrix whose elements are coefficients representing the strength of the causal relationships between one or more elements specified by the user into the generative model. Then, the instruction unit 43 instructs the generative model to analyze the relationships between the specified one or more elements and other elements based on the information representing the strength of the relationships between the multiple elements. The instruction unit 43 is an example of an instruction means.

[0131] For example, the instruction unit 43 receives information indicating the strength of the relationship between a plurality of elements from the acquisition unit 41. The instruction unit 43 also receives information indicating the element designated by the user from the reception unit 42.

[0132] The instruction unit 43 inputs information representing the strength of relationships between multiple elements into the generative model in the form of numerical values ​​such as vectors and matrices. The instruction unit 43 also inputs information indicating elements designated by the user into the generative model. Then, based on the information representing the strength of relationships between the multiple elements, the instruction unit 43 instructs the generative model to analyze the relationships between one or more elements designated by the user and other elements.

[0133] At this time, the instruction unit 43 may receive a prompt (instruction statement) from the user. The generative model analyzes the relationship between one or more elements specified by the user and other elements in accordance with the instructions from the instruction unit 43 and the prompt, and outputs the analysis result.

[0134] Thereafter, the instruction unit 43 notifies the output unit 44 that the generative model has been caused to analyze the information.

[0135] The output unit 44 outputs information based on the analysis results of the correlation coefficient matrix by the generative model (for example, facts, considerations, and conclusions). The output unit 44 is an example of an output means.

[0136] For example, after receiving a notification from the instruction unit 43, the output unit 44 acquires, from the generative model, an analysis result of the relationship between one or more elements specified by the user and other elements. The output unit 44 outputs information based on the acquired analysis result. The information based on the analysis result by the generative model includes information expressing, in language, the relationship between one or more elements specified by the user and other elements.

[0137] In one example, the output unit 44 displays, on the user terminal 100 (FIG. 1), a discussion screen (FIG. 4) on which the user can view the analysis results of the generative model. The discussion screen may include input data (e.g., information indicating the strength of the relationship between multiple elements, or information indicating one or multiple elements specified by the user) along with the analysis results of the generative model. The discussion screen may also include comments entered by the user (e.g., the user's opinion on the analysis results of the generative model).

[0138] (Operation of Information Processing Device 40) The operation of the information processing device 40 according to the fourth embodiment will be described with reference to Fig. 15. Fig. 15 is a flowchart showing the operation of the information processing device 40.

[0139] 15 , first, the acquisition unit 41 acquires information about a plurality of elements. The acquisition unit 41 performs a causal analysis on the information about the plurality of elements (S401). As a result, the acquisition unit 41 acquires information indicating the strength of the causal relationship between the plurality of elements.

[0140] The acquisition unit 41 outputs information indicating the strength of the causal relationship between a plurality of elements to the instruction unit 43 .

[0141] Next, the receiving unit 42 receives the designation of one or more elements from among the plurality of elements (S402).

[0142] The receiving unit 42 outputs information indicating one or more elements designated by the user to the instruction unit 43 .

[0143] The instruction unit 43 inputs a matrix representing the causal relationships between multiple elements into a generative model (Figure 13) and instructs that the generative model be used to analyze the relationships between one or more elements specified by the user and other elements (S403).

[0144] Thereafter, the instruction unit 43 notifies the output unit 44 that the generative model has executed an analysis of the information (a matrix representing the causal relationships between multiple elements).

[0145] Next, the output unit 44 outputs information based on the analysis results of the correlation coefficients using the generative model (S404). For example, the output unit 44 displays, on the user terminal 100 (FIG. 1), a discussion screen (FIG. 8) on which the user can view the analysis results using the generative model.

[0146] This completes the operation of the information processing device 40 according to the fourth embodiment.

[0147] (Effects of this embodiment) According to the configuration of this embodiment, the acquisition unit 41 acquires information indicating the strength of the causal relationships between multiple elements, obtained by performing causal analysis on information about multiple elements. The reception unit 42 receives designation of one or more elements from the multiple elements. The instruction unit 43 inputs a matrix indicating the causal relationships between the designated one or more elements and other elements to the generative model. The output unit 44 outputs information based on the analysis results of the matrix by the generative model.

[0148] In this way, the analysis results are output, rather than the information itself that represents the strength of the causal relationships between multiple elements (here, a matrix that represents the causal relationships between multiple elements), which can help the user understand the relationships between the elements specified.

[0149] Fifth Embodiment A fifth embodiment will be described with reference to Fig. 16. In this fifth embodiment, an example of the configuration of an information processing system including any one of the information processing devices 10, 20, 30, and 40 described in the first to fourth embodiments will be described.

[0150] (Information Processing System 1) Fig. 16 is a diagram schematically illustrating an example of the configuration of an information processing system 1 according to the fifth embodiment. As shown in Fig. 16, the information processing system 1 includes an information processing device 10 (20, 30, 40), a user terminal 100, and an information analysis execution device 200. Note that the number of user terminals 100 included in the information processing system 1 is not limited to three.

[0151] Here, "information processing device 10 (20, 30, 40)" means "any of the information processing devices 10, 20, 30 according to the first to fourth embodiments."

[0152] The user terminal 100 and the information processing device 10 (20, 30, 40) are communicatively connected. The user terminal 100 is used by a user to input information (data) and instructions to the information processing device 10 (20, 30, 40). For example, the user inputs information about multiple elements or information indicating the strength of the relationship between multiple elements to the user terminal 100. The user also inputs instructions (prompts) for the information analysis execution device 200 to the user terminal 100.

[0153] The information processing device 10 (20, 30, 40) acquires information indicating the strength of the relationship between a plurality of elements.

[0154] The information processing device 10 (20, 30, 40) inputs information representing the strength of the relationships between multiple elements to the information analysis execution device 200. At this time, the information processing device 10 (20, 30, 40) may convert the information representing the strength of the relationships between multiple elements into a format compatible with the information analysis execution device 200, and then input the information to the information analysis execution device 200.

[0155] The information analysis execution device 200 is pre-trained to contextualize input data (e.g., text data, image data) and output output data according to the context of the input data. The information analysis execution device 200 is, for example, a text generation model (e.g., a large-scale language model) that analyzes text data, or a generative model that analyzes input diagram data.

[0156] The information analysis execution device 200 analyzes information that indicates the strength of the relationships between multiple elements, which information is input from the information processing device 10 (20, 30, 40), and outputs the analysis results.

[0157] The information processing device 10 (20, 30, 40) acquires analysis results of information indicating the strength of relationships between multiple elements from the information analysis execution device 200. Then, the information processing device 10 (20, 30, 40) outputs information that expresses the relationships between one or multiple elements in language, based on the analysis results from the information analysis execution device 200.

[0158] (Effects of this embodiment) According to the configuration of this embodiment, the information processing device 10 (20, 30, 40) acquires information representing the strength of the relationship between multiple elements. The information processing device 10 (20, 30, 40) accepts the designation of one or more elements from the multiple elements. Based on the information representing the strength of the relationship between the multiple elements, the information processing device 10 (20, 30, 40) instructs the analysis of the relationship between the specified one or more elements using a generative model. Based on the analysis results using the generative model, the information processing device 10 (20, 30, 40) outputs information expressing the relationship between the one or more elements in language.

[0159] In this way, the analysis results are output, rather than the information (here, the correlation coefficient matrix) itself that represents the strength of the relationships between multiple elements, which can help the user understand the relationships between the elements specified.

[0160] (Hardware Configuration) Each of the components of the information processing devices 10, 20, 30, and 40 described in the first to fourth embodiments represents a functional block. Some or all of these components are realized by an information processing device such as that shown in Fig. 17. Fig. 17 is a block diagram showing an example of the hardware configuration of the information processing device.

[0161] 17 , the computer 110 includes a CPU (Central Processing Unit) 111, a main memory 112, a storage device 113, an input interface 114, a display controller 115, a data reader / writer 116, and a communication interface 117. These components are connected to each other via a bus 121 so as to be able to communicate data with each other. Note that the computer 110 may include a GPU (Graphics Processing Unit) or an FPGA (Field-Programmable Gate Array) in addition to or instead of the CPU 111.

[0162] The CPU 111 loads the program (code) of this embodiment stored in the storage device 113 into the main memory 112 and executes it in a predetermined order to perform various calculations. The main memory 112 is typically a volatile storage device such as a DRAM (Dynamic Random Access Memory). The program of this embodiment is provided in a state stored in a computer-readable recording medium 120. The program of this embodiment may be distributed over the Internet connected via the communication interface 117.

[0163] Specific examples of the storage device 113 include a hard disk drive and a semiconductor storage device such as a flash memory. The input interface 114 mediates data transmission between the CPU 111 and input devices 118 such as a keyboard and a mouse. The display controller 115 is connected to a display device 119 and controls the display on the display device 119.

[0164] The data reader / writer 116 mediates data transmission between the CPU 111 and the recording medium 120, reads programs from the recording medium 120, and writes processing results from the computer 110 to the recording medium 120. The communication interface 117 mediates data transmission between the CPU 111 and other computers.

[0165] Specific examples of the recording medium 120 include general-purpose semiconductor storage devices such as CF (Compact Flash (registered trademark)) and SD (Secure Digital), magnetic recording media such as flexible disks, or optical recording media such as CD-ROMs (Compact Disk Read Only Memory).

[0166] (Supplementary Notes) A ​​part or all of the above-described embodiments can be described as, but are not limited to, the following supplementary notes.

[0167] (Supplementary Note 1) An information processing device comprising: an acquisition means for acquiring information representing the strength of the relationship between a plurality of elements; a reception means for receiving the designation of one or more elements from the plurality of elements; an instruction means for instructing that the relationship between the one or more designated elements and other elements be analyzed using a generative model based on the information representing the strength of the relationship between the plurality of elements; and an output means for outputting information expressing the relationship between the one or more elements in language based on the analysis results using the generative model.

[0168] (Supplementary Note 2) The information processing device according to Supplementary Note 1, characterized in that the acquisition means acquires the information representing the strength of the causal relationship between the plurality of elements, obtained by performing a causal analysis on information relating to the plurality of elements.

[0169] (Supplementary Note 3) The information processing device according to Supplementary Note 1, wherein the accepting means accepts designation of the one or more elements on a screen displaying information indicating a causal relationship between the plurality of elements.

[0170] (Supplementary Note 4) The information processing device described in Supplementary Note 1, characterized in that the receiving means receives a designation of an element from among the one or more elements to be subject to reanalysis by the generative model after the information based on the analysis results by the generative model is output.

[0171] (Supplementary Note 5) The information processing device according to Supplementary Note 1, wherein the instruction means causes the generative model to analyze a relationship between the one or more specified elements and other elements.

[0172] (Supplementary Note 6) The information processing device according to Supplementary Note 1, wherein the instruction means causes the generative model to analyze a relationship between the one or more specified elements and a specific other element.

[0173] (Supplementary Note 7) The information processing device according to Supplementary Note 1, wherein the output means displays information representing a causal relationship between the one or more elements based on an analysis result by the generative model.

[0174] (Supplementary Note 8) The information processing device according to Supplementary Note 1, wherein the information representing the strength of the relationship between the plurality of elements is numerical data or chart data representing the strength of the relationship between the plurality of elements.

[0175] (Supplementary Note 9) The information processing device according to Supplementary Note 1, wherein the plurality of elements include a target variable that is a result and a plurality of explanatory variables that are causes.

[0176] (Supplementary Note 10) The information processing device according to Supplementary Note 1, wherein the instruction means causes the generative model to analyze the relationship between the specified one or more elements and other elements that have a causal relationship with the specified one or more elements.

[0177] (Supplementary Note 11) An information processing system comprising: an information processing device; one or more user terminals used by a user to input information and instructions to the information processing device; and an information analysis execution device trained to contextualize the input data and output output data according to the context of the input data, wherein the information processing device acquires information representing the strength of the relationship between a plurality of elements, accepts designation of one or more elements from the plurality of elements, instructs the use of a generative model to analyze the relationship between the designated one or more elements and other elements based on the information representing the strength of the relationship between the plurality of elements, and outputs information expressing the relationship between the one or more elements in language based on the analysis results by the generative model.

[0178] (Appendix 12) The information processing system described in Appendix 11 is characterized in that the information processing device acquires the information representing the strength of the causal relationship between the multiple elements, obtained by performing causal analysis on information regarding the multiple elements.

[0179] (Supplementary Note 13) The information processing system according to Supplementary Note 11, wherein the information processing device accepts designation of the one or more elements on a screen that displays information representing a causal relationship between the plurality of elements.

[0180] (Appendix 14) The information processing system described in Appendix 11 is characterized in that, after the information based on the analysis results by the generative model is output, the information processing device accepts designation of elements among the one or more elements to be targeted for reanalysis by the generative model.

[0181] (Supplementary Note 15) The information processing system according to Supplementary Note 11, wherein the information processing device causes the generative model to analyze a relationship between the one or more specified elements and other elements.

[0182] (Supplementary Note 16) The information processing system according to Supplementary Note 11, wherein the information processing device causes the generative model to analyze the relationship between the one or more specified elements and specific other elements.

[0183] (Supplementary Note 17) The information processing system according to Supplementary Note 11, wherein the information processing device displays information representing a causal relationship between the one or more elements based on an analysis result by the generative model.

[0184] (Supplementary Note 18) The information processing system according to Supplementary Note 11, wherein the information representing the strength of the relationship between the plurality of elements is numerical data or chart data representing the strength of the relationship between the plurality of elements.

[0185] (Supplementary Note 19) The information processing system according to Supplementary Note 11, wherein the plurality of elements include a response variable that is a result and a plurality of explanatory variables that are causes.

[0186] (Supplementary Note 20) The information processing system described in Supplementary Note 11 is characterized in that the information processing device causes the generative model to analyze the relationship between the specified one or more elements and other elements that have a causal relationship with the specified one or more elements.

[0187] (Supplementary Note 21) An information processing method in which a computer obtains information representing the strength of the relationship between a plurality of elements, accepts the designation of one or more elements from the plurality of elements, instructs the computer to use a generative model to analyze the relationship between the designated one or more elements and other elements based on the information representing the strength of the relationship between the plurality of elements, and outputs information expressing the relationship between the one or more elements in language based on the analysis results using the generative model.

[0188] (Supplementary Note 22) The information processing method described in Supplementary Note 21, characterized in that the computer acquires the information representing the strength of the causal relationship between the plurality of elements, obtained by performing causal analysis on information regarding the plurality of elements.

[0189] (Supplementary Note 23) The information processing method according to Supplementary Note 21, wherein the computer accepts designation of the one or more elements on a screen that displays information indicating causal relationships between the plurality of elements.

[0190] (Appendix 24) The information processing method described in Appendix 21, characterized in that the computer, after outputting the information based on the analysis results by the generative model, accepts designation of elements from the one or more elements to be subject to reanalysis by the generative model.

[0191] (Supplementary Note 25) The information processing method according to Supplementary Note 21, wherein the computer causes the generative model to analyze the relationship between the one or more specified elements and other elements.

[0192] (Supplementary Note 26) The information processing method according to Supplementary Note 21, wherein the computer causes the generative model to analyze the relationship between the one or more specified elements and specific other elements.

[0193] (Supplementary Note 27) The information processing method according to Supplementary Note 21, characterized in that the computer displays information representing a causal relationship between the one or more elements based on the analysis results of the generative model.

[0194] (Supplementary Note 28) The information processing method according to Supplementary Note 21, wherein the information representing the strength of the relationship between the plurality of elements is numerical data or chart data representing the strength of the relationship between the plurality of elements.

[0195] (Supplementary Note 29) The information processing method according to Supplementary Note 21, wherein the plurality of elements includes a response variable that is a result and a plurality of explanatory variables that are causes.

[0196] (Supplementary Note 30) The information processing method according to Supplementary Note 21, characterized in that the computer causes the generative model to analyze the relationship between the specified one or more elements and other elements that have a causal relationship with the specified one or more elements.

[0197] (Supplementary Note 31) A program for causing a computer to execute the following processes: a process for acquiring information representing the strength of the relationship between a plurality of elements; a process for accepting the designation of one or more elements from the plurality of elements; a process for instructing the use of a generative model to analyze the relationship between the one or more designated elements and other elements based on the information representing the strength of the relationship between the plurality of elements; and a process for outputting information expressing the relationship between the one or more elements in language based on the analysis results using the generative model.

[0198] (Supplementary Note 32) The recording medium according to Supplementary Note 31, wherein the program causes the computer to execute a process of acquiring the information representing the strength of the causal relationship between the plurality of elements, the information being obtained by performing a causal analysis on information relating to the plurality of elements.

[0199] (Appendix 33) The recording medium described in Appendix 31, characterized in that the program causes the computer to execute a process of accepting designation of the one or more elements on a screen displaying information representing the causal relationships between the multiple elements.

[0200] (Appendix 34) The recording medium described in Appendix 31, characterized in that the program causes the computer to execute a process of accepting designation of an element from among the one or more elements to be subject to reanalysis by the generative model after the information based on the analysis results by the generative model is output.

[0201] (Supplementary Note 35) The recording medium according to Supplementary Note 31, wherein the program causes the computer to execute a process of causing the generative model to analyze the relationship between the one or more specified elements and other elements.

[0202] (Supplementary Note 36) The recording medium described in Supplementary Note 31, wherein the program causes the computer to execute a process of having the generative model analyze the relationship between the specified one or more elements and specific other elements.

[0203] (Supplementary Note 37) The recording medium described in Supplementary Note 31, characterized in that the program causes the computer to execute a process of displaying information representing the causal relationship between the one or more elements based on the analysis results of the generative model.

[0204] (Supplementary Note 38) The recording medium according to Supplementary Note 31, wherein the information representing the strength of the relationship between the plurality of elements is numerical data or chart data representing the strength of the relationship between the plurality of elements.

[0205] (Supplementary Note 39) The recording medium according to Supplementary Note 31, wherein the plurality of elements include a response variable that is a result and a plurality of explanatory variables that are causes.

[0206] (Supplementary Note 40) The recording medium described in Supplementary Note 31, wherein the program causes the computer to execute a process of having the generative model analyze the relationship between the specified one or more elements and other elements that have a causal relationship with the specified one or more elements.

[0207] The present disclosure has been described above with reference to several embodiments. However, the present disclosure is not limited to the above embodiments. Each embodiment can be combined with other embodiments as appropriate. Furthermore, various modifications that can be understood by those skilled in the art can be made to the configurations and details of the above embodiments within the scope of the present disclosure.

[0208] This application claims priority based on Japanese Patent Application No. 2024-006392, filed January 18, 2024, the disclosure of which is incorporated herein by reference in its entirety.

[0209] The present disclosure can be used, for example, in information processing using generative models.

[0210] REFERENCE SIGNS LIST 1 Information processing system 10 Information processing device 11 Acquisition unit 12 Reception unit 12' Reception unit 13 Instruction unit 14 Output unit 20 Information processing device 30 Information processing device 40 Information processing device 41 Acquisition unit 42 Reception unit 43 Instruction unit 44 Output unit 100 User terminal 200 Information analysis execution device

Claims

1. An information processing apparatus comprising: an acquisition means for acquiring information representing the strength of relationships among a plurality of elements; a reception means for receiving a specification of one or more elements among the plurality of elements; an instruction means for instructing, based on the information representing the strength of relationships among the plurality of elements, to analyze the relationships between the specified one or more elements and other elements using a generation model; and an output means for outputting information representing in language the relationships among the one or more elements based on the analysis result by the generation model.

2. The information processing apparatus according to claim 1, wherein the acquisition means acquires the information representing the strength of causal relationships among the plurality of elements, which is obtained by performing causal analysis on the information regarding the plurality of elements.

3. The information processing apparatus according to claim 1, wherein the reception means receives the specification of the one or more elements on a screen displaying information representing the causal relationships among the plurality of elements.

4. The information processing apparatus according to claim 1, wherein the reception means receives the specification of an element to be re-analyzed by the generation model among the one or more elements after the information based on the analysis result by the generation model is output.

5. The information processing apparatus according to claim 1, wherein the instruction means causes the generation model to analyze the relationships between the specified one or more elements and other elements.

6. The information processing apparatus according to claim 1, wherein the instruction means causes the generation model to analyze the relationships between the specified one or more elements and specific other elements.

7. The information processing apparatus according to claim 1, wherein the output means displays information representing the causal relationships among the one or more elements based on the analysis result by the generation model.

8. The information processing apparatus according to claim 1, wherein the information representing the strength of relationships among the plurality of elements is numerical data or chart data representing the strength of relationships among the plurality of elements.

9. The information processing apparatus according to claim 1, wherein the plurality of elements include a target variable as a result and a plurality of explanatory variables as causes.

10. The information processing apparatus according to claim 1, wherein the instruction means causes the generation model to analyze the relationship between the specified one or more elements and other elements that are causally related to the one or more elements.

11. An information processing system comprising: an information processing apparatus; one or more user terminals used by a user to input information and instructions to the information processing apparatus; and an information analysis execution apparatus trained to contextualize input data and output output data according to the context of the input data, wherein the information processing apparatus obtains information representing the strength of the relationship between a plurality of elements, receives a specification of one or more elements among the plurality of elements, and based on the information representing the strength of the relationship between the plurality of elements, instructs to analyze the relationship between the specified one or more elements and other elements using a generation model, and outputs information representing the relationship between the one or more elements in language based on the analysis result by the generation model.

12. The information processing system according to claim 11, wherein the information processing apparatus obtains the information representing the strength of the causal relationship between the plurality of elements, which is obtained by performing causal analysis on the information regarding the plurality of elements.

13. The information processing system according to claim 11, wherein the information processing apparatus receives a specification of the one or more elements on a screen displaying information representing the causal relationship between the plurality of elements.

14. The information processing system according to claim 11, wherein after the information based on the analysis result by the generation model is output, the information processing apparatus receives a specification of an element to be re-analyzed by the generation model among the one or more elements.

15. The information processing system according to claim 11, wherein the information processing apparatus causes the generation model to analyze the relationship between the specified one or more elements and other elements.

16. The information processing system according to claim 11, wherein the information processing apparatus causes the generation model to analyze the relationship between the specified one or more elements and specific other elements.

17. The information processing apparatus according to claim 11, wherein the information processing apparatus displays information representing a causal relationship between the one or more elements based on the analysis result by the generation model.

18. The information processing system according to claim 11, wherein the information representing the strength of the relationship between the plurality of elements is numerical data or chart data representing the strength of the relationship between the plurality of elements.

19. An information processing method, comprising: a computer obtaining information representing the strength of a relationship between a plurality of elements; receiving a designation of one or more of the plurality of elements; instructing, based on the information representing the strength of the relationship between the plurality of elements, to analyze, using a generation model, a relationship between the designated one or more elements and other elements; and outputting information expressing, in language, the relationship between the one or more elements based on the analysis result by the generation model.

20. A non-transitory recording medium storing a program for causing a computer to execute: a process of obtaining information representing the strength of a relationship between a plurality of elements; a process of receiving a designation of one or more of the plurality of elements; a process of instructing, based on the information representing the strength of the relationship between the plurality of elements, to analyze, using a generation model, a relationship between the designated one or more elements and other elements; and a process of outputting information expressing, in language, the relationship between the one or more elements based on the analysis result by the generation model.

Images