Analysis support system, analysis support method, and analysis support program

Abstract

To improve the convenience of data analysis support.SOLUTION: An instruction reception unit configured to receive an analysis instruction regarding predetermined data analysis from a user device, a first instruction unit configured to output first instruction data for specifying an analysis technique according to the analysis instruction to an external information processing system based on the analysis instruction, a second instruction unit configured to output second instruction data for generating call data for calling data necessary for analysis according to the specified analysis technique to the external information processing system, and a database management unit configured to manage data necessary for analysis according to the analysis instruction based on the call data; The information processing device includes a provision request unit configured to make a provision request for the provision data, a third instruction unit configured to output, based on the provision data and an analysis instruction, third instruction data for causing an external information processing system to execute analysis according to the analysis instruction, and a result output unit configured to output an analysis result according to the analysis instruction based on a result of the execution.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an analysis support system, an analysis support method, and an analysis support program. [Background technology]

[0002] Conventionally, techniques for supporting data analysis desired by a user have been known.

[0003] For example, the information processing device described in Patent Document 1 includes a setting unit that sets a scenario indicating which AI engines to use among multiple AI engines and the order in which the AI ​​engines are to be used; a transmission unit that sequentially transmits data stored in a storage unit to the AI ​​engines according to the scenario set by the setting unit; and a storage unit that receives processing result data from the AI ​​engines that transmitted the data via the transmission unit and stores the data in the storage unit. Here, the scenario in the device indicates which AI engines to use and the order in which the AI ​​engines are to be used. The device selects a scenario from multiple pre-set scenarios depending on the type of input data to be processed and transmits the data to the AI ​​engines according to the selected scenario. Specifically, for example, if the data to be processed is voice data, the device first converts the voice data to text data and then selects a scenario to process the text data; and if the data to be processed is text data, it first selects a scenario to process the text data. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2020-9337 Summary of the Invention [Problem to be solved by the invention]

[0005] Incidentally, even when the type of data is the same, multiple data analysis methods may be expected. For example, when the data to be processed is numerical data, multiple analysis methods for the numerical data may be expected depending on the purpose of the analysis. However, the information processing device described in Patent Document 1 merely selects a pre-set data processing scenario depending on the type of data to be processed, and does not take into account the case where multiple data analysis methods are expected based on the data to be processed. As a result, user convenience has been limited.

[0006] Therefore, an object of the present invention is to improve the convenience of supporting data analysis. [Means for solving the problem]

[0007] An analysis support system according to one embodiment of the present invention includes an instruction receiving unit that receives analysis instructions regarding a specified data analysis from a user device used by a user; a first instruction unit that outputs first instruction data to an external information processing system based on the analysis instructions, for causing the system to identify an analysis method corresponding to the analysis instructions; a second instruction unit that outputs second instruction data to the external information processing system, for causing the system to generate call data for calling up data necessary for analysis using the identified analysis method; a provision request unit that makes a provision data request to a database that manages data necessary for analysis according to the analysis instructions, based on the generated call data; a third instruction unit that outputs third instruction data to the external information processing system based on the provided data and the analysis instructions, for causing the system to execute an analysis according to the analysis instructions; and a result output unit that outputs analysis results corresponding to the analysis instructions, based on the results of the execution. [Effects of the Invention]

[0008] According to the present invention, it is possible to improve the convenience of supporting data analysis. [Brief explanation of the drawings]

[0009] [Figure 1]FIG. 1 is a diagram showing an overview of processing in an analysis support system 100 according to an embodiment of the present invention. [Figure 2] 1 is a diagram showing the configuration of an analysis support system 100 according to an embodiment of the present invention. [Figure 3] FIG. 10 is a diagram showing an example of first instruction data. [Figure 4] FIG. 10 is a diagram showing an example of second instruction data. [Figure 5] FIG. 10 is a diagram showing an example of third instruction data. [Figure 6] 10A and 10B are diagrams illustrating examples of screens displayed on the user device 200. [Figure 7] 1 is a sequence chart showing an example of processing in analysis support system 100. DETAILED DESCRIPTION OF THE INVENTION

[0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will now be described with reference to the accompanying drawings, in which: Figure 1 is a diagram showing an overview of processing in an analysis support system 100 according to one embodiment of the present invention.

[0011] Analysis support system 100 is an information processing system implemented by an analysis support program. Analysis support system 100 performs analysis using an external information processing system based on an analysis instruction received from user device 200, and outputs the analysis results to the user device.

[0012] First, analysis support system 100 receives an analysis instruction for a predetermined data analysis from user device 200 (S101).

[0013] Next, the analysis support system 100 outputs first instruction data for specifying an analysis method according to the analysis instructions to the information processing system 300 (S102). The analysis support system 100 acquires first result data according to the first instruction data from the information processing system 300 (S103). The analysis support system 100 outputs second instruction data to the information processing system 300 for generating call data for calling up data necessary for analysis using the specified analysis method (S104). The analysis support system 100 acquires second result data according to the second instruction data from the information processing system 300 (S105).

[0014] The analysis support system 100 modifies the call data as necessary (S106). The analysis support system 100 requests the purchase database 400 to provide the provided data based on the call data (including the modified call data) (S107). The analysis support system 100 acquires the provided data from the purchase database 400 (S108).

[0015] The analysis support system 100 outputs third instruction data to the information processing system 300 for executing an analysis in accordance with the analysis instruction, based on the provided data and the analysis instruction (S109). The analysis support system 100 acquires third result data in accordance with the third instruction data from the information processing system 300. The analysis support system 100 then outputs the analysis results in accordance with the analysis instruction to the user device 200 (S111).

[0016] In this way, the analysis support system 100 can output analysis results to the user device 200 by utilizing the information processing system 300, the purchase database 400, and the vector database 500 in a stepwise or sequential manner, so to speak, in response to analysis instructions received from the user device 200. Furthermore, by being able to clearly indicate in detail hypothesis proposals and decision-making materials in data analysis support, particularly purchase data analysis support, the feasibility and convenience of data analysis, particularly purchase data analysis, can be improved. As a result, even when analysis instructions are received from the user device 200 in natural language, the analysis support system 100 can utilize the information processing system 300, the purchase database 400, and the vector database 500 to output the analysis results desired by the user to the user device 200.

[0017] In this embodiment, an example in which the analysis support system 100 executes analysis support for purchase data will be described, but the analysis support executed by the analysis support system 100 is not limited to this.

[0018] 2 is a diagram showing the configuration of an analysis support system 100 according to one embodiment of the present invention. Analysis support system 100 is communicably connected to a user device 200, an information processing system 300, a purchase database 400, and a vector database 500 via a network such as the Internet. Details of analysis support system 100 will be described later.

[0019] The user device 200 is an information processing device used by a user, such as a computer, a smartphone, a tablet terminal, or a personal computer.

[0020] A user accesses the analysis support system 100 using the user device 200 and issues an analysis instruction. The analysis instruction may be, for example, an analysis instruction written in natural language.

[0021] Furthermore, the user accesses analysis support system 100 using user device 200 to obtain the analysis results.

[0022] The information processing system 300 is an information processing system that performs predetermined data processing based on instruction data (first instruction data, second instruction data, third instruction data) acquired from the analysis support system 100.

[0023] The information processing system 300 may be, for example, a predetermined learning model or artificial intelligence (AI), particularly, for example, a generative AI. The generative AI may be, for example, an information processing system trained by a GPT (Generative Pretrained Transformer) model. The information processing system 300 may be, for example, an information processing system that accepts instruction data in a dialogue format using natural language and outputs data processing results, particularly, for example, ChatGPT or GPT4.

[0024] The purchase database 400 and the vector database 500 are databases that manage data necessary for analysis in accordance with analysis instructions. The purchase database 400 may be provided inside the analysis support system 100 or may be provided outside the analysis support system 100.

[0025] Here, the purchase database 400 stores, for example, purchase history data (e.g., big data) in which product information and consumer purchase data are linked. The vector database 500 also stores vector data in which product information such as retail store information, product brand master data, and product category lists is stored in a format that can be processed by the call data correction unit 140 (described later).

[0026] Furthermore, the purchasing database 400 may be a database that outputs data for each of a plurality of organizations in a format that makes it difficult to identify each of the organizations. This allows the analysis support system 100 to execute analysis support processing while maintaining data security and confidentiality.

[0027] In Figure 2, one user device 200, one information processing system 300, one purchasing database 400, and one vector database 500 are shown, but there may be multiple user devices 200, multiple information processing systems 300, multiple purchasing databases 400, and multiple vector databases 500.

[0028] Next, details of analysis support system 100 will be described. Analysis support system 100 includes a storage unit 110, an instruction receiving unit 115, a first instruction unit 120, a first result acquisition unit 125, a second instruction unit 130, a second result acquisition unit 135, a call data correction unit 140, a provision request unit 145, a provision data acquisition unit 150, a third instruction unit 155, a third result acquisition unit 160, and a result output unit 165. Each unit shown in Fig. 2 can be realized, for example, by using a storage area or by a processor executing a program stored in the storage area.

[0029] The storage unit 110 stores information processed in the analysis support system 100. The storage unit 110 can store, for example, analysis instruction data, first result data, second result data, corrected call data, and third result data, which will be described later.

[0030] The instruction receiving unit 115 receives an analysis instruction relating to a predetermined data analysis from the user device 200, and stores analysis instruction data relating to the received analysis instruction in the storage unit 110.

[0031] Here, the analysis instruction to be received may be, for example, an analysis instruction in natural language. That is, the user inputs natural language using the user device 200, and the instruction receiving unit 115 can receive the input natural language as an analysis instruction.

[0032] The analysis instruction to be received may be, for example, an analysis instruction to instruct POS analysis based on POS (Point of Sales) data and a competitive analysis comparing multiple companies or products. The competitive analysis may be, for example, an analysis comparing the sales, profit margin, market share, characteristics, or awareness of multiple companies or products. An analysis instruction to instruct POS analysis may be, for example, "Please tell me the sales of Company A." An analysis instruction to instruct competitive analysis may be, for example, "Please compare and analyze the sales of Product A, Product B, and Product C."

[0033] The instruction receiving unit 115 can also receive an analysis instruction including an evaluation instruction for predetermined data.

[0034] Here, the evaluation instruction may be an evaluation instruction instructing an evaluation (for example, a suggestion, explanation, comment, etc.) of the analysis results of a graph or table obtained by analyzing predetermined data. An analysis instruction including an evaluation instruction for predetermined data may be, for example, "Compare the sales of product A, product B, and product C, and evaluate the features of each product."

[0035] The analysis instructions accepted by the instruction accepting unit 115 are not limited to these.

[0036] Based on the analysis instruction, the first instruction unit 120 generates first instruction data for causing the information processing system 300 to identify an analysis method according to the analysis instruction, and outputs the first instruction data.

[0037] The first instruction unit 120 can output the first instruction data to, for example, the information processing system 300. The first instruction unit 120 may output the first instruction data to the information processing system 300 via an information processing system different from the information processing system 300.

[0038] The first instruction data may include details regarding the analysis instruction received by the instruction receiving unit 115 and details regarding an overview of at least one of the plurality of analysis methods.

[0039] Here, the content of the analysis instruction received by the instruction receiving unit 115 and included in the first instruction data may be the analysis instruction itself received by the instruction receiving unit 115 (for example, the text of the analysis instruction itself received by the instruction receiving unit 115), or may be a part of the analysis instruction received by the instruction receiving unit 115 or an edited version (for example, a summary or supplement) of the analysis instruction.

[0040] Furthermore, the content regarding the overview of at least one of the multiple analysis methods included in the first instruction data includes, for example, at least the purpose and use of the at least one analysis method, examples of data required for analysis, examples of call data, and examples of output analysis results. This allows the information processing system 300 to identify an appropriate analysis method from among the multiple analysis methods in response to the received analysis instruction. In particular, for example, when the information processing system 300 is a general-purpose information processing system used for various purposes, it can identify an appropriate analysis method from among the multiple analysis methods in response to the received analysis instruction.

[0041] The first instruction data may be, for example, so-called prompt data for causing the information processing system 300 to execute a predetermined process. The second instruction data and the third instruction data may also be so-called prompt data.

[0042] FIG. 3 is a diagram illustrating an example of the first instruction data.

[0043] The first instruction data includes, for example, instruction data regarding the role and function of the information processing system 300, instruction data regarding processing that the information processing system 300 can execute, instruction data regarding an overview of the analysis method, instruction data regarding an analysis instruction received from the user device 200, and instruction data instructing the identification of an analysis method suitable for the analysis instruction.

[0044] Note that the first instruction data shown in Figure 3 is an example, and the types of data and analysis methods included in the first instruction data are not limited to those shown in Figure 3. Furthermore, at least a portion of the first instruction data may be written in Japanese or English, or may be written in another language. Furthermore, the first instruction data may include natural language, or may include content other than natural language (such as a programming language, a table, a diagram, or a mathematical formula).

[0045] The first result acquisition unit 125 acquires the first result data output by the information processing system 300 in response to the first instruction data, and stores it in the storage unit 110.

[0046] The first result data includes, for example, content indicating an analysis method according to the analysis instruction, identified in the information processing system 300. The first result data may further include supplementary content (for example, specific grounds or reasons, etc.) of the identified process in the information processing system 300. Specifically, the first result data may be, for example, "POS analysis is appropriate. Please call a tool to obtain data for POS analysis."

[0047] The first result acquisition section 125 may acquire the first result data, for example, directly from the information processing system 300 or indirectly from the information processing system 300 via another information processing system different from the information processing system 300.

[0048] The second instruction unit 130 generates second instruction data for causing the information processing system 300 to generate call data for calling up data required for analysis using the identified analysis method, and outputs the second instruction data.

[0049] The second instruction unit 130 can output the second instruction data to, for example, the information processing system 300. The second instruction unit 130 may output the second instruction data to the information processing system 300 via an information processing system different from the information processing system 300.

[0050] The second instruction data is, for example, instruction data for the information processing system 300 to generate call data for calling data required for analysis by the identified analysis method in the information processing system 300.

[0051] The second instruction data includes content necessary for generating call data in the information processing system 300. The content necessary for generating call data in the information processing system 300 includes, for example, at least one of an example of call data, an example of data managed by the purchase database 400, and an example of data necessary for analysis by the specified analysis method.

[0052] The second instruction data may be different for each identified analysis method, thereby enabling analysis support system 100 to acquire appropriate call data in accordance with the analysis instruction received from the user.

[0053] The call data is data for calling data required for analysis according to the analysis instruction from the purchase database 400. The format of the call data is not particularly limited, but may be, for example, data in SQL (Structured Query Language) format or JSON (JavaScript Object Notation) format.

[0054] The second instruction data may include semantic layer information, which allows the analysis support system 100 to appropriately acquire second acquisition data (described later) and perform high-quality data analysis.

[0055] Here, the semantic layer information is information related to the semantic layer for converting or translating between instructions for the database and instructions from the user in order to facilitate communication between the database and the user and enable the user to easily use the database. The semantic layer information includes, for example, schema information, meta information, and domain information.

[0056] Schema information is, for example, information indicating the structure (table structure) of the purchasing database 400. Meta information is information indicating the nature, meaning, and purpose of each piece of data. Domain information is information indicating the definition of an index calculated from the data (for example, the ratio of sales to the previous period).

[0057] The meta information includes, for example, data type information and description information. The data type information is, for example, information indicating the type of data (e.g., a numerical value, a character string, etc.). The description information is information in which the nature, meaning, and purpose of the data are described using natural language or mathematical formulas. Here, the natural language may be Japanese, English, or another language.

[0058] The semantic layer information may be set for each data or for a predetermined process, or may be set for each analysis method. The semantic layer information set for each analysis method may be, for example, semantic layer information set for each data or for each predetermined process used in the analysis method.

[0059] For example, analysis support system 100 manages semantic layer information set in advance for each analysis method in association with the analysis method. This allows second instruction unit 130 to dynamically include semantic layer information corresponding to the identified analysis method in the second instruction data, enabling analysis support system 100 to perform high-quality data analysis.

[0060] FIG. 4 is a diagram illustrating an example of the second instruction data.

[0061] The second instruction data includes, for example, instruction data regarding the role or function of the information processing system 300, instruction data regarding an analysis instruction received from the user device 200, instruction data regarding the identified analysis method, instruction data instructing the generation of call data, semantic layer information, and information regarding an analysis example. The semantic layer information may be, for example, semantic layer information set for each analysis method that is dynamically inserted according to the identified analysis method.

[0062] 4 is an example. At least a part of the second instruction data may be written in Japanese, English, or another language. The second instruction data may include a natural language, or may include content other than a natural language (such as a programming language, a table, a diagram, or a mathematical formula).

[0063] The second result acquisition unit 135 acquires the second result data output by the information processing system 300 in response to the second instruction data, and stores it in the storage unit 110.

[0064] The second result data includes, for example, call data for calling up data required for analysis by the identified analysis method.

[0065] The second result acquisition section 135 may acquire the second result data, for example, directly from the information processing system 300 or indirectly from the information processing system 300 via another information processing system different from the information processing system 300.

[0066] The call data correction unit 140 corrects at least one piece of data included in the call data generated in accordance with the second instruction data based on the similarity between data (e.g., vector data) registered in the vector database 500 and the data included in the call data, and stores the corrected call data in the memory unit 110.

[0067] The processing by the call data correction unit 140 will be explained using an example in which a user provides an analysis instruction for a company named "Shirayuki Pharmacy Co., Ltd.", and data for that company is registered in the vector database 500 under the company name "Shirahana Pharmacy Co., Ltd."

[0068] This case may occur, for example, when a user mistakenly provides an analysis instruction for the company name "Shirayuki Pharmacy Co., Ltd.", even though the correct company name is "Shirayuki Pharmacy Co., Ltd." In this case, through processing in analysis support system 100, second result acquisition unit 135 acquires second result data including, for example, call data for calling data for the company name "Shirayuki Pharmacy Co., Ltd." from purchase database 400. However, because data is registered in purchase database 400 under the company name "Shirayuki Pharmacy Co., Ltd.", provision request unit 145, described below, may not be able to appropriately request provision of data for the company name "Shirayuki Pharmacy Co., Ltd." based on the call data.

[0069] Therefore, the call data correction unit 140 corrects the call data generated in response to the second instruction data to, for example, call data for calling data related to the company name "Shirahana Pharmacy Co., Ltd." based on the similarity between the data registered in the vector database 500 and the data included in the call data generated in response to the second instruction data. This allows the provision request unit 145, which will be described later, to appropriately request the provision of data related to the company name "Shirahana Pharmacy Co., Ltd." based on the corrected call data.

[0070] Here, for example, based on the similarity between data registered in the vector database 500 and data included in the call data generated in response to the second instruction data, if the data included in the call data generated in response to the second instruction data is not registered in the vector database 500, the call data correction unit 140 may replace the data included in the call data generated in response to the second instruction data with data having the highest similarity among the data registered in the vector database 500. Here, the similarity may be, for example, cosine similarity.

[0071] Specifically, the call data correction unit 140 first identifies the data with the company "Shirahana Pharmacy Co., Ltd." as the data with the highest similarity among the data registered in the vector database 500. Next, the call data correction unit 140 replaces the data with the company "Shirayuki Pharmacy Co., Ltd." included in the call data generated in accordance with the second instruction data with the data with the company "Shirahana Pharmacy Co., Ltd."

[0072] In this embodiment, unless otherwise specified and as far as technical substitution is possible, the call data may include modified call data.

[0073] For example, the call data correction unit 140 may evaluate the similarity between the data registered in the vector database 500 and the data included in the call data generated in accordance with the second instruction data for all of the call data generated in the information processing system 300. This allows the analysis support system 100 to correct the call data generated in accordance with the second instruction data into appropriate call data and perform the analysis support process.

[0074] Furthermore, for example, when the provision request process by the provision request unit 145 described below fails, the call data correction unit 140 may evaluate the similarity between the data registered in the vector database 500 and the data included in the call data generated in accordance with the second instruction data, and correct the call data. This allows the analysis support system 100 to execute the call data correction process when the call data correction process is necessary, thereby efficiently executing the analysis support process. Note that here, failure of the provision request process includes, for example, at least one of obtaining data from the vector database 500 indicating that data corresponding to the call data is not registered in the vector database 500, receiving an instruction from the user to, for example, re-execute the analysis, and detecting inaccuracy in the analysis based on predetermined conditions.

[0075] The provision request unit 145 makes a provision request for the provision data to the purchase database 400, which manages data necessary for analysis according to the analysis instruction, based on the call data. The provision request unit 145 may make the provision request based on the call data included in the second result data, for example, or may make the provision request based on the corrected call data.

[0076] Provision request unit 145 can make a provision request, for example, by outputting the call data to purchase database 400. Purchase database 400 provides analysis support system 100 with the data registered in purchase database 400 in response to the call data.

[0077] The provided data acquisition unit 150 acquires the provided data provided from the purchase database 400 in response to the call data, and stores the provided data in the storage unit 110.

[0078] The third instruction unit 155 generates third instruction data for causing the information processing system 300 to execute an analysis in accordance with the analysis instruction, based on the provided data and the analysis instruction, and outputs the third instruction data.

[0079] The third instruction unit 155 can output the third instruction data to, for example, the information processing system 300. The third instruction unit 155 may output the third instruction data to the information processing system 300 via an information processing system different from the information processing system 300.

[0080] The third instruction data is instruction data for executing an analysis according to the analysis instruction in the information processing system 300 based on the provided data and the analysis instruction. The third instruction data includes, for example, content indicating an analysis method to be executed in the information processing system 300, and provided data necessary for the analysis using the analysis method.

[0081] Furthermore, when the analysis instruction includes an instruction to evaluate predetermined data, the third instruction data may be instruction data for executing an analysis relating to the evaluation of the provided data based on the evaluation instruction.

[0082] FIG. 5 is a diagram illustrating an example of the third instruction data.

[0083] The third instruction data includes, for example, instruction data related to the role and function of the information processing system 300, instruction data related to an analysis instruction received from the user device 200, instruction data related to the identified analysis method, instruction data related to the analysis result to be output, and instruction data related to specific data required for the analysis. The third instruction data may also include data related to specifying the format of the third result data (specifying output using text or output using graphs, or specifying the file format to be output), which will be described later.

[0084] 5 is an example. At least a part of the third instruction data may be written in Japanese, English, or another language. The third instruction data may include a natural language, or may include content other than a natural language (such as a programming language, a table, a diagram, or a mathematical formula).

[0085] Furthermore, the third instruction data may further include data that is not necessary for the analysis. This allows the analysis support system 100 to utilize the information processing system 300 to select and discard data that is necessary for the analysis.

[0086] The third result acquisition unit 160 acquires the third result data output by the information processing system 300 in response to the third instruction data, and stores it in the storage unit 110.

[0087] The third result data includes, for example, the content of the analysis result executed by the information processing system 300 in response to the third instruction data.

[0088] The third result data may include, for example, numerical values ​​indicating the analysis results, diagrams visualizing the numerical values ​​(graphs, tables, etc.), and sentences in natural language, which allow the user to easily understand the analysis results according to the analysis instructions.

[0089] The third result acquiring section 160 may acquire the third result data, for example, directly from the information processing system 300 or indirectly from the information processing system 300 via another information processing system different from the information processing system 300.

[0090] The result output unit 165 outputs the analysis result according to the analysis instruction based on the result of the execution of the analysis according to the analysis instruction. Furthermore, the result output unit 165 can output the analysis result including the evaluation result according to the evaluation instruction based on the result of the execution of the analysis according to the analysis instruction.

[0091] Here, the result of executing the analysis in response to the analysis instruction may be, for example, at least a part of the content indicated by the third result data.

[0092] The analysis results output by the result output unit 165 may include, for example, analyzed numerical values, diagrams visualizing the numerical values ​​(graphs, tables, etc.), and sentences in natural language, which allows the user to easily understand the analysis results according to the analysis instructions.

[0093] At this time, the result output unit 165 may output the analysis result to the user device 200, for example.

[0094] In addition, if the purchasing database 400 is a database that outputs data for each of multiple organizations in a format that makes it difficult to identify each of the multiple organizations, the result output unit 165 can output the analysis results based on the data for each of the multiple organizations in a format that makes it difficult to identify each of the multiple organizations.

[0095] FIG. 6 is a diagram showing an example of a screen displayed on the user device 200. As shown in FIG.

[0096] The screen shown in Fig. 6 is an example of a usage screen displayed on the user device 200 of a user who uses the analysis support system 100. Area 601 displays details related to the analysis instructions received from the user. Areas 602 and 603 display the analysis results according to the analysis instructions. In the screen shown in Fig. 6, as an example, area 602 displays a graph, and area 603 displays a table. Area 604 displays the evaluation results of the analysis results as the analysis results according to the analysis instructions.

[0097] In this way, analysis support system 100 can accept analysis instructions including natural language and output analysis results including natural language. Furthermore, analysis support system 100 can accept analysis instructions from user device 200, for example, through an interactive screen (chat screen) displayed on user device 200, and output analysis results to user device 200. This allows analysis support system 100 to provide analysis support with high convenience.

[0098] FIG. 7 is a sequence chart showing an example of processing in analysis support system 100.

[0099] First, the instruction receiving unit 115 receives an analysis instruction from the user device 200 (S701). The first instruction unit 120 generates first instruction data and outputs it to the information processing system 300 (S702). The first result acquisition unit 125 acquires first result data from the information processing system 300 (S703).

[0100] The second instruction unit 130 generates second instruction data and outputs it to the information processing system 300 (S704). The second result acquisition unit 135 acquires second result data from the information processing system 300 (S705). The call data correction unit 140 corrects the call data (S706). At this time, the call data correction unit 140 exchanges data with the vector database 500 (not shown) to correct the call data. The provision request unit 145 outputs the call data to the purchase database 400 (S707). The provision data acquisition unit 150 acquires the provision data from the purchase database 400 (S708).

[0101] The third instruction unit 155 generates third instruction data and outputs it to the information processing system 300 (S709). The third result acquisition unit 160 acquires the third result data from the information processing system 300 (S710). The result output unit 165 outputs the analysis result to the user device 200 (S711).

[0102] In this way, analysis support system 100 can output instruction data to information processing system 300 in multiple installments in response to analysis instructions received from user device 200. This allows analysis support system 100 to output appropriate analysis results in response to analysis instructions containing natural language. Furthermore, even if there is an error in the analysis instructions containing natural language, analysis support system 100 can correct the call data as necessary, allowing the user to receive highly convenient analysis support.

[0103] The above describes one embodiment of the present invention. Analysis support system 100 accepts an analysis instruction from user device 200, outputs first instruction data to information processing system 300 based on the analysis instruction, outputs second instruction data to information processing system 300, requests purchase database 400 to provide data based on the generated call data, outputs third instruction data to information processing system 300 based on the provided data and analysis instruction, for executing an analysis in accordance with the analysis instruction, and outputs analysis results in accordance with the analysis instruction based on the execution results. This allows analysis support system 100 to improve convenience in data analysis support.

[0104] Furthermore, the analysis support system 100 can receive an analysis instruction including an instruction to evaluate predetermined data, output third instruction data to the information processing system 300 for executing an analysis related to the evaluation of the provided data based on the evaluation instruction, and output an evaluation result according to the evaluation instruction. This allows the analysis support system 100 to output an evaluation of the analysis result to the user, thereby improving convenience for the user in data analysis support.

[0105] Furthermore, analysis support system 100 can accept analysis instructions including input in natural language and output analysis results including output in natural language, allowing users to conveniently input analysis instructions and easily understand the analysis results.

[0106] Furthermore, the analysis support system 100 can output first instruction data including an analysis instruction and an overview of at least one of the plurality of analysis techniques, and the first instruction data further including at least one of examples of data required for analysis, examples of call data, and examples of analysis results for at least one of the plurality of analysis techniques. This allows the analysis support system 100 to utilize the information processing system 300 to identify an analysis technique suitable for the analysis instruction received from the user.

[0107] Furthermore, analysis support system 100 can correct at least one piece of data included in the generated call data in accordance with the data registered in vector database 500, based on the similarity between the data registered in vector database 500 and the data included in the generated call data, and can issue a provision request based on the corrected call data. This allows analysis support system 100 to output appropriate analysis results to the user even if the analysis instructions received from the user contain, for example, an error in the description, thereby improving convenience in data analysis support.

[0108] Furthermore, analysis support system 100 can output analysis results based on the data of multiple organizations in a format that makes it difficult to identify each of the multiple organizations. This allows analysis support system 100 to improve the convenience of data analysis support while maintaining security and confidentiality.

[0109] It should be noted that the present embodiment is provided to facilitate understanding of the present invention and is not intended to limit the present invention. The present invention may be modified or improved without departing from the spirit thereof, and equivalents thereof are also included in the present invention.

[0110] In addition, in the present invention, a "unit" does not simply mean a physical means, but also includes cases where the functions of the "unit" are realized by software. Furthermore, the functions of one "unit" or device may be realized by two or more physical means, devices, or software, and the functions of two or more "units" or devices may be realized by one physical means, device, or software. [Explanation of symbols]

[0111] 100 Analysis support system, 110 Memory unit, 115 Instruction receiving unit, 120 First instruction unit, 125 First result acquisition unit, 130 Second instruction unit, 135 Second result acquisition unit, 140 Call data correction unit, 145 Provision request unit, 150 Provision data acquisition unit, 155 Third instruction unit, 160 Third result acquisition unit, 165 Result output unit, 200 User device, 300 Information processing system, 400 Purchase database, 500 Vector database

Claims

1. an instruction receiving unit that receives an analysis instruction regarding a predetermined data analysis from a user device used by a user; a first instruction unit that outputs, based on the analysis instruction, first instruction data for specifying an analysis method corresponding to the analysis instruction to an external information processing system; a second instruction unit that outputs second instruction data to the external information processing system to generate call data for calling data necessary for analysis by the specified analysis method; a provision request unit that issues a provision data request to a database that manages data necessary for analysis according to the analysis instruction based on the generated call data; a third instruction unit that outputs third instruction data to the external information processing system based on the provided data and the analysis instruction, the third instruction data causing the external information processing system to execute an analysis in accordance with the analysis instruction; a result output unit that outputs an analysis result according to the analysis instruction based on the execution result; An analysis support system equipped with:

2. the instruction receiving unit receives the analysis instruction including an evaluation instruction for predetermined data; the third instruction unit outputs the third instruction data to the external information processing system to cause the external information processing system to execute an analysis related to an evaluation of the provided data based on the evaluation instruction; the result output unit outputs the evaluation result in response to the evaluation instruction. The analysis support system according to claim 1 .

3. the analysis instructions include input in natural language; the analysis results include natural language output.

3. The analysis support system according to claim 1 or 2.

4. The analysis support system according to claim 3 , wherein the first instruction data includes the analysis instruction and content relating to an overview of at least one of a plurality of analysis methods.

5. The analysis support system according to claim 4 , wherein the first instruction data further includes at least one of an example of data required for analysis, an example of call data, and an example of an analysis result for at least one of a plurality of analysis methods.

6. a call data correcting unit that corrects at least one piece of data included in the generated call data in accordance with data registered in the database, based on a similarity between data registered in the database and data included in the generated call data; the provision request unit makes the provision request based on the corrected call data.

3. The analysis support system according to claim 1 or 2.

7. the database includes a database that outputs data of each of a plurality of organizations in a format that makes it difficult to identify each of the plurality of organizations; the result output unit outputs the analysis result based on the data of each of the plurality of organizations in a format that makes it difficult to identify each of the plurality of organizations.

3. The analysis support system according to claim 1 or 2.

8. The computer receiving an analysis instruction regarding a predetermined data analysis from a user device used by a user; outputting, to an external information processing system based on the analysis instruction, first instruction data for specifying an analysis method according to the analysis instruction; outputting second instruction data to the external information processing system to generate call data for calling data necessary for analysis by the specified analysis method; Based on the generated call data, a request for provision of data is made to a database that manages data necessary for analysis in accordance with the analysis instruction; outputting third instruction data to the external information processing system based on the provided data and the analysis instruction, the third instruction data being for causing the external information processing system to execute an analysis in accordance with the analysis instruction; outputting an analysis result according to the analysis instruction based on the execution result; Analysis support method.

9. On the computer, an instruction receiving unit that receives an analysis instruction regarding a predetermined data analysis from a user device used by a user; a first instruction unit that outputs, based on the analysis instruction, first instruction data for specifying an analysis method corresponding to the analysis instruction to an external information processing system; a second instruction unit that outputs second instruction data to the external information processing system to generate call data for calling data necessary for analysis by the specified analysis method; a provision request unit that issues a provision data request to a database that manages data necessary for analysis according to the analysis instruction based on the generated call data; a third instruction unit that outputs third instruction data to the external information processing system based on the provided data and the analysis instruction, the third instruction data causing the external information processing system to execute an analysis in accordance with the analysis instruction; a result output unit that outputs an analysis result according to the analysis instruction based on the execution result; An analysis support program to achieve this.

Images