Information processing device, information processing method, and information processing program

Abstract

To provide an information processing device capable of generating appropriate output using generative AI for a group of numerical values to be input, an information processing method, and an information processing program.SOLUTION: An information processing device comprises an acquisition unit, a generation unit, and an output unit. The acquisition unit acquires numerical value group information indicative of a numerical value group, and purpose information indicative of the intended use of the numerical value group. The generation unit generates output information in response to the intended use using a generative AI on the basis of the numerical value group information and purpose information acquired by the acquisition unit. The output unit outputs the output information generated by the generation unit.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, and an information processing program. [Background technology]

[0002] Conventionally, systems have been provided that perform natural language processing such as question-and-answering using generative AI (Artificial Intelligence) such as large-scale language models (LLMs). The sentences input to the generative AI are also called prompts. Patent Document 1 proposes a technology that generates a prompt by adding additional sentences generated based on candidate sentences related to an input question as reference information to the question. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 7313757 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the technology described in Patent Document 1 is a technology for outputting an answer to a question expressed in text, and has a problem in that it does not take into consideration a group of numerical values.

[0005] The present application has been made in consideration of the above, and aims to provide an information processing device, an information processing method, and an information processing program that can generate appropriate output for an input group of numerical values ​​using a generation AI. [Means for solving the problem]

[0006] The information processing device according to the present application includes an acquisition unit, a generation unit, and an output unit. The acquisition unit acquires numerical group information indicating a numerical group and purpose information indicating a purpose for which the numerical group is to be used. The generation unit generates output information according to the purpose of use using a generation AI based on the numerical group information and purpose information acquired by the acquisition unit. The output unit outputs the output information generated by the generation unit. [Effects of the Invention]

[0007] According to one aspect of the embodiment, a generation AI can be used to generate an appropriate output for an input group of numerical values. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram illustrating an example of information processing according to an embodiment. [Figure 2] FIG. 2 is a diagram illustrating an example of the configuration of an information processing system according to the embodiment. [Figure 3] FIG. 3 is a diagram illustrating an example of the configuration of the information processing device according to the embodiment. [Figure 4] FIG. 4 is a flowchart showing an example of information processing by the processing unit of the information processing device according to the embodiment. [Figure 5] FIG. 5 is a hardware configuration diagram showing an example of a computer that realizes the functions of the information processing device according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, modes for implementing an information processing device, an information processing method, and an information processing program according to the present application (hereinafter referred to as "embodiments") will be described in detail with reference to the drawings. Note that the information processing device, the information processing method, and the information processing program according to the present application are not limited to these embodiments. Furthermore, the respective embodiments can be appropriately combined within the scope of not causing any contradiction in the processing content. Furthermore, the same components in the following embodiments will be assigned the same reference numerals, and redundant explanations will be omitted.

[0010] [1. An example of information processing] First, an example of information processing according to the embodiment will be described with reference to Fig. 1. Fig. 1 is a diagram showing an example of information processing according to the embodiment.

[0011] As shown in Fig. 1, the information processing device 1 acquires numerical value group information from the terminal device 2 (step S1). The numerical value group information includes information indicating a numerical value group. The numerical value group includes, for example, a plurality of combinations of one or more types of numerical values. The numerical value group information is, for example, text information in which one or more types of numerical values ​​are repeatedly arranged on each line, such as information in CSV (Comma Separated Values) format, but is not limited to such an example.

[0012] The information processing device 1 also acquires purpose information indicating the purpose of use of the group of values ​​indicated by the above-mentioned group of values ​​information (step S2). The purpose of use may be, for example, one or more of estimation of the target indicated by the group of values, outlier analysis, status analysis, future prediction, trend analysis, heat map creation, etc., but may also be a more detailed analysis, estimation, prediction, etc.

[0013] The object represented by a group of numerical values ​​is, for example, an attribute item indicated by the numerical values ​​represented by the group of numerical values, such as coordinate position, time, temperature, and various other items, but is not limited to such examples. Anomaly analysis is, for example, an analysis of the number, percentage, and appearance pattern of abnormal values ​​(outliers). Status analysis is an analysis of the status of one object or group of objects represented by a group of numerical values. Future prediction is a future prediction of the numerical values ​​represented by a group of numerical values. Trend analysis is an analysis of the patterns and trends of the numerical values ​​represented by a group of numerical values.

[0014] For example, in step S2, the information processing device 1 acquires purpose information from the terminal device 2. The terminal device 2 acquires candidate purpose information indicating multiple candidate purpose of use, which are candidates for the purpose of use of the group of numerical values, from the information processing device 1, displays the multiple candidate purpose of use indicated in the acquired candidate purpose of use information, and allows the user U of the terminal device 2 to select one or more candidate purpose of use from the multiple candidate purpose of use. The terminal device 2 transmits information on the one or more candidate purpose of use selected by the user U to the information processing device 1 as purpose information. In the example shown in FIG. 1, the purpose of use indicated in the purpose information is estimation of the target indicated by the group of numerical values ​​and trend analysis thereof.

[0015] The information processing device 1 can transmit use purpose candidate information indicating a plurality of predetermined use purpose candidates to the terminal device 2, or can generate use purpose candidate information indicating a plurality of use purpose candidates using a generation AI and transmit the generated use purpose candidate information to the terminal device 2. The generation AI is a text generation AI or a multimodal generation AI. The text generation AI is, for example, a large-scale language model trained to estimate and output the next token from an input token sequence, such as a transformer-based model or an RNN (recurrent neural network)-based model, or a hybrid model of these. The text generation AI may also be a composite system combined with an identifier for preventing fraudulent use.

[0016] Examples of the transformer-based model include, but are not limited to, a generative pre-trained transformer (GPT) and a pathways language model version 2 (PaLM2). Examples of the RNN-based model include, but are not limited to, a receptance weighted key value (RWKV).

[0017] The multimodal generative AI is, for example, a generative AI that generates at least one of text and an image from text or a combination of text and an image. Examples of the multimodal generative AI include, but are not limited to, GPT-4 Turbo with vision, Gemini, and CM3Leon (Chameleon Multimodal Model).

[0018] The information processing device 1 inputs, for example, information including the numerical value group information acquired in step S1 and instruction information instructing the output of multiple candidate usage purposes to the generation AI, and causes the generation AI to output information indicating the multiple candidate usage purposes. The generation AI is located in an external information processing device, and the information processing device 1 uses the generation AI via an API (Application Programming Interface), but the generation AI may also be located within the information processing device 1.

[0019] The instruction information may be, for example, a string such as "You are an expert in statistical processing. Each value in the given set of values ​​is obtained from a specific sensor or a specific database. Analyze the characteristics of these sets of values ​​and list five or more possible applications for these sets of values."

[0020] The instruction information may also be, for example, a string such as "You are an expert in statistical processing. Each value in the given set of values ​​is a value obtained from a specific sensor or a specific database. List possible attribute items represented by each value in the given set of values. For the set of values ​​in the listed attribute items, list five or more that you think are appropriate candidates for the intended use for each attribute item." The instruction information may also include information indicating multiple candidate intended uses that are candidates to be enumerated by the generation AI.

[0021] Furthermore, the information processing device 1 can determine the range, number of digits, expression format, etc. of multiple numerical values ​​included in the numerical group for each type. The information processing device 1 has an attribute item dictionary that includes one or more pieces of information from among the numerical range, the number of digits of the numerical value, and the expression format of the numerical value for each attribute item, and can estimate the attribute items of the numerical values ​​included in the numerical group indicated by the numerical group information based on this attribute item dictionary. For example, the information processing device 1 estimates, as the one or more attribute items of the numerical values ​​included in the numerical group, the one or more attribute items whose range, number of digits, and expression format match or have the highest rate of match with the multiple numerical values ​​included in the numerical group.

[0022] Furthermore, the information processing device 1 has a usage purpose information dictionary that includes information indicating one or more attribute items for each usage purpose candidate, and can estimate one or more usage purpose candidates based on this usage purpose information dictionary. For example, the information processing device 1 can estimate, in the usage purpose information dictionary, the usage purpose candidates associated with each attribute item estimated as described above as usage purpose candidates to be presented to the user U.

[0023] Next, the information processing device 1 uses the generation AI to generate output information according to the purpose of use indicated by the purpose information, based on the numerical group information acquired in step S1 and the purpose information acquired in step S2 (step S3). The output information according to the purpose of use indicated by the purpose information is information in which the numerical group has been processed (for example, processed, analyzed, or organized) so that it can be used for the purpose of use indicated by the purpose information, and is, for example, information shown in text or information showing the processed numerical group and its explanation, but may also be information including text and figures.

[0024] For example, the information processing device 1 inputs, as input information, to the generation AI, information including the character string "Analyze the given group of numerical values, process (for example, edit, analyze, or organize) the information into information in accordance with the given purpose of use, and output it," as instruction information, the numerical value group information acquired in step S1, and the purpose information acquired in step S2. This allows the information processing device 1 to cause the generation AI to generate, as output information in accordance with the purpose of use indicated by the purpose information, information processed in accordance with the purpose of use indicated by the purpose information.

[0025] The information processing device 1 also acquires, from an internal storage unit, example information that exemplifies output information according to the intended use of the numerical group indicated by the numerical group information. The example information includes, for example, one or more combinations of information indicating examples of numerical groups and examples of intended uses of the numerical groups, and examples of output information according to the intended uses, for each combination of the numerical group and the intended use.

[0026] In this case, the information processing device 1 inputs information including example information in addition to the instruction information, the numerical group information acquired in step S1, and the objective information acquired in step S2 to the generation AI as input information, and causes the generation AI to generate output information according to the purpose of use.

[0027] The information processing device 1 has an example information dictionary containing example information for each candidate purpose of use, and can obtain example information corresponding to the candidate purpose of use selected by the user U from the example information dictionary, and include the obtained example information in the input information to be input to the generation AI.

[0028] Furthermore, the information processing device 1 can input, as input information, information including instruction information, which is information instructing the generation AI to select, from among the plurality of pieces of example information, example information suitable for the purpose of use indicated by the usage information, the usage information, and the plurality of pieces of example information. This allows the information processing device 1 to cause the generation AI to select, from among the plurality of pieces of example information, example information suitable for the purpose of use indicated by the usage information.

[0029] The generating AI may be generated by learning (for example, fine tuning) using learning information including, for each combination of the numerical group information and the purpose information, numerical group information indicating a numerical group, purpose information indicating a purpose of use of the numerical group, and output information according to the purpose of use. The output information included in the learning information is used as a label.

[0030] In this case, the information processing device 1 inputs information including the numerical group information acquired in step S1 and the objective information acquired in step S2 as input information to the generation AI, and causes the generation AI to generate output information according to the purpose of use indicated by the objective information.

[0031] Furthermore, the information processing device 1 can use each of multiple types of generation AIs to generate output information candidates, which are candidates for output information corresponding to the numerical value group information acquired in step S1. These multiple types of generation AIs are generated to output output information candidates for numerical value groups of different types. For example, each generation AI is trained (e.g., fine-tuned) using training information that includes corresponding specific types of numerical value groups and output information candidates for each numerical value group. Furthermore, the multiple types of generation AIs may be generation AIs generated separately by different types of providers, or may be generation AIs of different types with different architectures.

[0032] The information processing device 1 estimates the importance of each of the multiple output information candidates generated by the multiple types of generation AI with respect to the purpose of use indicated by the purpose information acquired in step S2. For example, the information processing device 1 inputs, as input information to the generation AI for each output information candidate, instruction information including information indicating an instruction to estimate the importance of the output information candidate with respect to the purpose of use indicated in the purpose information acquired in step S2, and information including the purpose information and the output information candidate, causes the generation AI to estimate the importance of the output information candidate, and causes the generation AI to output information indicating the importance.

[0033] In this case, the generation AI that estimates the importance of the output information candidates is generated by, for example, learning (e.g., fine tuning) using training information that includes, for each output information candidate, instruction information, objective information, output information candidate, and information indicating the importance. The information indicating the importance included in the training information is used as a label.

[0034] Furthermore, the generation AI that estimates the importance of the output information candidates may be a generation AI that has been trained (e.g., fine-tuned) using training information that includes, for each output information candidate, target information, output information candidates, and information indicating the importance of the output information candidate. The information indicating the importance included in the training information is used as a label. In this case, the information processing device 1 inputs information including the target information and output information candidates acquired in step S2 as input information to the generation AI for each output information candidate, causes the generation AI to estimate the importance of the output information candidate, and causes the generation AI to output information indicating the importance.

[0035] Based on the importance of each output information candidate estimated as described above, the information processing device 1 determines one or more output information candidates from among the plurality of output information candidates as output information corresponding to the purpose of use indicated by the objective information acquired in step S2. For example, the information processing device 1 can determine the output information candidate with the highest estimated importance as output information, or the output information candidate with an estimated importance equal to or greater than a threshold as output information. Furthermore, the information processing device 1 can also determine m output information candidates as output information in descending order of estimated importance, where m is an integer equal to or greater than 2.

[0036] Furthermore, the information processing device 1 can also cause the generation AI to generate processing information indicating a type of statistical processing according to the purpose of use indicated by the purpose information acquired in step S2. For example, the information processing device 1 causes the generation AI to generate processing information by inputting, as input information, information including the purpose information and instruction information including information instructing the generation AI to output processing information indicating a type of statistical processing according to the purpose of use indicated by the purpose information acquired in step S2.

[0037] In this case, the information processing device 1 generates output information according to the purpose of use by performing statistical processing on the group of values ​​indicated by the group of values ​​using the statistical processing indicated by the processing information. The information processing device 1 can execute a function that performs statistical processing for each type of statistical processing, and performs statistical processing on the group of values ​​indicated by the group of values ​​using a function according to the type of statistical processing indicated by the processing information.

[0038] Next, the information processing device 1 outputs the output information generated in step S3 (step S4). For example, in step S4, the information processing device 1 transmits the output information generated in step S3 to the terminal device 2 and provides the output information generated in step S3 to the user U.

[0039] In this way, the information processing device 1 acquires value group information indicating a value group and purpose information indicating a purpose for which the value group is to be used, generates output information according to the purpose of use using the generation AI based on the acquired value group information and purpose information, and outputs the generated output information. This allows the information processing device 1 to make the generation AI generate an appropriate output for the input value group.

[0040] The configuration of an information processing system including the information processing device 1 that performs such processing and the terminal device 2 will be described in detail below.

[0041] [2. Information Processing System Configuration] 2 is a diagram showing an example of the configuration of an information processing system according to the embodiment. As shown in FIG. 2, the information processing system 100 according to the embodiment includes an information processing device 1 and a plurality of terminal devices 2.

[0042] The multiple terminal devices 2 are used by different users U. The terminal devices 2 are, for example, laptop personal computers (PCs), desktop PCs, smartphones, tablet PCs, and wearable devices. The wearable devices are, for example, smart glasses or smart watches, but are not limited to these examples.

[0043] The information processing device 1 and the terminal device 2 are connected to each other so as to be able to communicate with each other by wire or wirelessly via a network N. Note that the information processing system 100 shown in Fig. 2 may include a plurality of information processing devices 1 and the like.

[0044] The network N includes, for example, a WAN (Wide Area Network) such as the Internet and a mobile communication network such as LTE (Long Term Evolution), 4G (4th Generation), and 5G (5th Generation: 5th generation mobile communication system).

[0045] The terminal device 2 can connect to the network N via short-range wireless communication such as a mobile communication network, Bluetooth (registered trademark), or wireless LAN (Local Area Network), and communicate with the information processing device 1 and the like.

[0046] 3. Configuration of Information Processing Device 1 3 is a diagram showing an example of the configuration of the information processing device 1 according to the embodiment. As shown in FIG. 3, the information processing device 1 includes a communication unit 10, a storage unit 11, and a processing unit 12.

[0047] [3.1. Communication Unit 10] The communication unit 10 is realized by, for example, a communication module or a network interface card (NIC). The communication unit 10 is connected to a network N by wire or wirelessly, and transmits and receives information to and from various other devices. For example, the communication unit 10 transmits and receives information to and from the terminal device 2 via the network N.

[0048] [3.2. Storage section 11] The storage unit 11 is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk.

[0049] The storage unit 11 stores various types of information. For example, the storage unit 11 stores information acquired by the processing unit 12 and information generated by the processing unit 12. The storage unit 11 also stores, for example, an attribute item dictionary, a usage purpose information dictionary, and an example information dictionary.

[0050] [3.3. Processing Unit 12] The processing unit 12 is a controller, and is realized by a processor such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit) executing various programs (corresponding to examples of information processing programs) stored in a storage device inside the information processing device 1 using RAM or the like as a working area.

[0051] The processing unit 12 is a controller, and may be realized by an integrated circuit such as an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or a general purpose graphic processing unit (GPGPU).

[0052] 3, the processing unit 12 has an acquisition unit 30, a generation unit 31, and an output unit 32, and realizes or executes the functions and actions of information processing described below. Note that the internal configuration of the processing unit 12 is not limited to the configuration shown in FIG. 3, and may be any other configuration as long as it performs the information processing described below.

[0053] [3.3.1. Acquisition part 30] The acquiring unit 30 acquires various information. For example, the acquiring unit 30 acquires numerical value group information indicating a numerical value group. For example, the acquiring unit 30 acquires the numerical value group information transmitted from the terminal device 2 via the communication unit 10 and the network N.

[0054] The numerical value group information includes information indicating a numerical value group. The numerical value group includes, for example, a plurality of combinations of one or more types of numerical values. The numerical value group information is, for example, text information in which one or more types of numerical values ​​are repeatedly arranged on each line, for example, information in CSV format, but is not limited to such an example.

[0055] The acquiring unit 30 also acquires purpose information indicating the purpose of use of the group of numerical values ​​indicated by the group of numerical values ​​information acquired by the acquiring unit 30. The purpose of use may be, for example, one or more of estimation of the target indicated by the group of numerical values, outlier analysis, status analysis, future prediction, trend analysis, heat map creation, etc., but may also be a more detailed analysis, estimation, prediction, etc.

[0056] The object represented by a group of numerical values ​​is, for example, an attribute item indicated by the numerical values ​​represented by the group of numerical values, such as coordinate position, time, temperature, and various other items, but is not limited to such examples. Anomaly analysis is, for example, an analysis of the number, percentage, and appearance pattern of abnormal values ​​(outliers). Status analysis is an analysis of the status of one object or group of objects represented by a group of numerical values. Future prediction is a future prediction of the numerical values ​​represented by a group of numerical values. Trend analysis is an analysis of the patterns and trends of the numerical values ​​represented by a group of numerical values.

[0057] Furthermore, the acquisition unit 30 acquires the purpose information transmitted from the terminal device 2 via the communication unit 10 and the network N. For example, the acquisition unit 30 acquires, as the purpose information, information on one or more candidate purpose of use selected from a plurality of candidate purpose of use.

[0058] The acquiring unit 30 also acquires example information that exemplifies output information according to the intended use of the numeric group indicated by the acquired numeric group information from the storage unit 11. The example information includes, for example, one or more combinations of information indicating examples of numeric groups and examples of intended uses of the numeric groups, and examples of output information according to the intended uses, for each combination of a numeric group and an intended use.

[0059] The acquisition unit 30 acquires, for example, example information corresponding to the candidate use purpose selected as the use purpose by the user U from an example information dictionary. The example information dictionary includes example information for each candidate use purpose and is stored in the storage unit 11. The acquisition unit 30 acquires, from the example information dictionary stored in the storage unit 11, example information corresponding to the candidate use purpose selected as the use purpose by the user U.

[0060] [3.3.2. Generation unit 31] The generation unit 31 generates various information. Based on the numerical group information and purpose information acquired by the acquisition unit 30, the generation unit 31 uses a generation AI to generate output information according to the purpose of use indicated by the purpose information. The output information according to the purpose of use indicated by the purpose information is information obtained by processing (e.g., processing such as processing, manipulation, analysis, or organization) a numerical group so that it can be used for the purpose of use indicated by the purpose information, and is, for example, information shown in text or information showing a processed numerical group and its explanation, but may also be information including text and figures.

[0061] For example, the generation unit 31 inputs, as input information, a character string as instruction information, "Analyze the given group of numerical values, process (for example, process by manipulating, analyzing, or arranging) the information into information in accordance with the given purpose of use, and output the information," and information including the group of numerical values ​​information and purpose information acquired by the acquisition unit 30 to the generation AI. This allows the generation unit 31 to cause the generation AI to generate, as output information in accordance with the purpose of use indicated by the purpose information, information processed in accordance with the purpose of use indicated by the purpose information.

[0062] The generation AI used by the generation unit 31 is a text generation AI or a multimodal generation AI. The text generation AI is, for example, a large-scale language model trained to estimate and output the next token from an input token sequence, such as a transformer-based model or an RNN-based model, but may also be a hybrid model of these. The text generation AI may also be a composite system combined with an identifier to prevent fraudulent use.

[0063] Examples of the transformer-based model include, but are not limited to, GPT and PaLM2, etc. Examples of the RNN-based model include, but are not limited to, RWKV, etc.

[0064] The multimodal generative AI is, for example, a generative AI that generates at least one of text and an image from text or a combination of text and an image. Examples of the multimodal generative AI include, but are not limited to, GPT-4 Turbo with vision, Gemini, and CM3Leon.

[0065] Furthermore, the generation unit 31 generates output information according to the intended use using the generation AI based on the instruction information and the numerical group information, objective information, and example information acquired by the acquisition unit 30. For example, the acquisition unit 30 inputs information including the instruction information and the numerical group information, objective information, and example information acquired by the acquisition unit 30 as input information to the generation AI, and causes the generation AI to generate output information according to the intended use. In this case, the instruction information includes information such as a string of characters "Analyze the given numerical group with reference to the given example information, and process (e.g., process by processing, analyzing, or arranging) it into information according to the given intended use, and output it."

[0066] Furthermore, the generation unit 31 can input, as input information to the generation AI, information including instruction information, which is information instructing the selection of example information suitable for the purpose of use indicated by the usage information from among the plurality of example information, the usage information, and the plurality of example information. In this way, the generation unit 31 can cause the generation AI to select example information suitable for the purpose of use indicated by the usage information from among the plurality of example information.

[0067] The generating AI may be generated by learning (for example, fine tuning) using learning information including, for each combination of the numerical group information and the purpose information, numerical group information indicating a numerical group, purpose information indicating a purpose of use of the numerical group, and output information according to the purpose of use. The output information included in the learning information is used as a label.

[0068] In this case, the generation unit 31 inputs information including the numerical group information and purpose information acquired by the acquisition unit 30 to the generation AI as input information, and causes the generation AI to generate output information according to the purpose of use indicated by the purpose information.

[0069] The generation unit 31 can also use each of multiple types of generation AIs to generate output information candidates, which are candidates for output information corresponding to the numerical value group information acquired by the acquisition unit 30. These multiple types of generation AIs are generated to output output information candidates for numerical value groups of different types. For example, each generation AI is trained (e.g., fine-tuned) using learning information that includes corresponding specific types of numerical value groups and output information candidates for each numerical value group. Furthermore, the multiple types of generation AIs may be generation AIs generated separately by different types of providers, or may be generation AIs of different types with different architectures.

[0070] The generation unit 31 estimates the importance of each of the multiple output information candidates generated by the multiple types of generation AI with respect to the purpose of use indicated in the purpose information acquired by the acquisition unit 30. For example, the generation unit 31 inputs, as input information, instruction information including information indicating an instruction to estimate the importance of the output information candidate with respect to the purpose of use indicated in the purpose information acquired by the acquisition unit 30, information including the purpose information and the output information candidate to the generation AI for each output information candidate, causes the generation AI to estimate the importance of the output information candidate, and causes the generation AI to output information indicating the importance.

[0071] In this case, the generation AI that estimates the importance of the output information candidates is generated by, for example, learning (e.g., fine tuning) using training information that includes, for each output information candidate, instruction information, objective information, output information candidate, and information indicating the importance. The information indicating the importance included in the training information is used as a label.

[0072] Furthermore, the generation AI that estimates the importance of the output information candidates may be a generation AI that has been trained (e.g., fine-tuned) using training information that includes, for each output information candidate, target information, output information candidates, and information indicating the importance of the output information candidate. The information indicating the importance included in the training information is used as a label. In this case, the generation unit 31 inputs information including the target information and output information candidates acquired by the acquisition unit 30 as input information to the generation AI for each output information candidate, causes the generation AI to estimate the importance of the output information candidate, and causes the generation AI to output information indicating the importance.

[0073] Based on the importance of each output information candidate estimated as described above, the generation unit 31 determines one or more output information candidates from among the multiple output information candidates as output information corresponding to the purpose of use indicated by the purpose information acquired by the acquisition unit 30. For example, the generation unit 31 can determine the output information candidate with the highest estimated importance as the output information, or the output information candidate with an estimated importance equal to or greater than a threshold as the output information. Furthermore, the generation unit 31 can also determine m output information candidates as the output information in descending order of estimated importance, where m is an integer equal to or greater than 2.

[0074] Furthermore, the generation unit 31 can also cause the generation AI to generate processing information indicating a type of statistical processing according to the purpose of use indicated in the purpose information acquired by the acquisition unit 30. For example, the generation unit 31 causes the generation AI to generate processing information by inputting, to the generation AI, information including the purpose information and instruction information including information instructing the output of processing information indicating a type of statistical processing according to the purpose of use indicated in the purpose information acquired by the acquisition unit 30 as input information.

[0075] In this case, the generation unit 31 performs statistical processing on the group of values ​​indicated by the group of values ​​using the statistical processing indicated by the processing information, thereby generating output information according to the purpose of use. The generation unit 31 can execute a function that performs statistical processing for each type of statistical processing, and performs statistical processing on the group of values ​​indicated by the group of values ​​using a function according to the type of statistical processing indicated by the processing information.

[0076] Furthermore, the generation unit 31 can use the generation AI to generate candidate purpose information indicating a plurality of candidate purpose of use, which are candidates for the purpose of use of the numerical group indicated by the numerical group information acquired by the acquisition unit 30. For example, the generation unit 31 inputs information including the numerical group information acquired by the acquisition unit 30 and instruction information indicating a plurality of candidate purpose of use to the generation AI, and causes the generation AI to output information indicating the plurality of candidate purpose of use.

[0077] The instruction information may be, for example, a string such as "You are an expert in statistical processing. Each value in the given set of values ​​is obtained from a specific sensor or a specific database. Analyze the characteristics of these sets of values ​​and list five or more possible applications for these sets of values."

[0078] The instruction information may also be, for example, a string such as "You are an expert in statistical processing. Each value in the given set of values ​​is a value obtained from a specific sensor or a specific database. List possible attribute items represented by each value in the given set of values. For the set of values ​​in the listed attribute items, list five or more that you think are appropriate candidates for the intended use for each attribute item." The instruction information may also include information indicating multiple candidate intended uses that are candidates to be enumerated by the generation AI.

[0079] The generation unit 31 also determines the range, number of digits, expression format, etc. of the multiple numerical values ​​included in the numerical group for each type. The generation unit 31 can estimate attribute items of the numerical values ​​included in the numerical group indicated by the numerical group information, for example, based on an attribute item dictionary stored in the storage unit 11 and acquired by the acquisition unit 30. The generation unit 31 estimates, for example, one or more attribute items whose range, number of digits, and expression format match or have the highest rate of match with the multiple numerical values ​​included in the numerical group as one or more attribute items of the numerical values ​​included in the numerical group. The attribute item dictionary includes, for each attribute item, information on one or more of the numerical range, the number of digits of the numerical value, and the expression format of the numerical value.

[0080] Furthermore, the generation unit 31 can estimate one or more use purpose candidates based on the use purpose information dictionary stored in the storage unit 11 and acquired by the acquisition unit 30. The use purpose information dictionary includes information indicating one or more attribute items for each use purpose candidate. For example, the generation unit 31 can estimate, in the use purpose information dictionary, the use purpose candidates associated with each attribute item estimated as described above as the use purpose candidates to be presented to the user U.

[0081] As shown in FIG. 3, the generation unit 31 includes a first generation processing unit 40, an estimation processing unit 41, a determination processing unit 42, and a second generation processing unit 43. The first generation processing unit 40 uses each of the multiple types of generation AIs described above to generate output information candidates that are candidates for output information corresponding to the numerical value group information acquired by the acquisition unit 30. The multiple types of generation AIs are generated to output output information candidates for numerical value groups of different types. For example, each generation AI is trained (e.g., fine-tuned) using training information that includes a corresponding specific type of numerical value group and output information candidates for each numerical value group.

[0082] The estimation processing unit 41 estimates the importance of each of the multiple output information candidates generated by the first generation processing unit 40 for the purpose of use indicated by the purpose information. For example, the estimation processing unit 41 inputs, as input information, instruction information including information indicating an instruction to estimate the importance of the output information candidate for the purpose of use indicated by the purpose information acquired by the acquisition unit 30, information including the purpose information and the output information candidate, to the generation AI for each output information candidate, causes the generation AI to estimate the importance of the output information candidate, and causes the generation AI to output information indicating the importance.

[0083] In this case, the generation AI that estimates the importance of the output information candidates is generated by, for example, learning (e.g., fine tuning) using training information that includes, for each output information candidate, instruction information, objective information, output information candidate, and information indicating the importance. The information indicating the importance included in the training information is used as a label.

[0084] Furthermore, the generation AI that estimates the importance of the output information candidates may be a generation AI that has been trained (e.g., fine-tuned) using training information that includes, for each output information candidate, target information, output information candidates, and information indicating the importance of the output information candidate. The information indicating the importance included in the training information is used as a label. In this case, the estimation processing unit 41 inputs information including the target information and output information candidates acquired by the acquisition unit 30 as input information to the generation AI for each output information candidate, causes the generation AI to estimate the importance of the output information candidate, and causes the generation AI to output information indicating the importance.

[0085] The determination processing unit 42 determines, as output information, one or more output information candidates from among the plurality of output information candidates based on the importance estimated by the estimation processing unit 41. For example, the determination processing unit 42 can determine, as output information, the output information candidate with the highest estimated importance, or the output information candidate with an estimated importance equal to or greater than a threshold. Furthermore, the determination processing unit 42 can also determine, as output information, m output information candidates in descending order of estimated importance, for example.

[0086] Furthermore, the first generation processing unit 40 causes the generation AI to generate processing information indicating a type of statistical processing according to the purpose of use indicated in the purpose information. For example, the first generation processing unit 40 can also cause the generation AI to generate processing information indicating a type of statistical processing according to the purpose of use indicated in the purpose information acquired by the acquisition unit 30. For example, the first generation processing unit 40 inputs, as input information to the generation AI, information including the purpose information and instruction information including information instructing the generation AI to output processing information indicating a type of statistical processing according to the purpose of use indicated in the purpose information acquired by the acquisition unit 30, thereby causing the generation AI to generate processing information.

[0087] The second generation processing unit 43 generates output information according to the purpose of use by performing statistical processing on the group of numerical values ​​indicated in the numerical value group information using the statistical processing indicated in the processing information generated by the first generation processing unit 40. The second generation processing unit 43 can execute a function that performs statistical processing for each type of statistical processing, and performs statistical processing on the group of numerical values ​​indicated in the numerical value group information using a function according to the type of statistical processing indicated in the processing information generated by the first generation processing unit 40.

[0088] 3.3.3. Output section 32 The output unit 32 outputs various information. For example, the output unit 32 outputs various information by transmitting the various information to the terminal device 2 via the communication unit 10 and the network N, and provides the various information to the user U.

[0089] For example, the output unit 32 transmits the use purpose candidate information generated by the generation unit 31 to the terminal device 2 via the communication unit 10 and the network N so that one or more use purpose candidates from among the multiple use purpose candidates can be selected.

[0090] The terminal device 2 acquires the candidate purpose of use information transmitted from the information processing device 1, and based on the acquired candidate purpose of use information, displays one or more selectable candidate purpose of use from the multiple candidate purpose of use, and allows the user U of the terminal device 2 to select one or more candidate purpose of use from the multiple candidate purpose of use. The terminal device 2 transmits information indicating the one or more candidate purpose of use selected by the user U to the information processing device 1 as purpose information.

[0091] Furthermore, the output unit 32 outputs the output information generated by the generation unit 31. For example, the output unit 32 transmits the output information generated by the generation unit 31 to the terminal device 2 via the communication unit 10 and the network N. The terminal device 2 acquires the output information transmitted from the information processing device 1 by the output unit 32 and displays the acquired output information.

[0092] [4. Processing Procedure] Next, a procedure of information processing by the processing unit 12 of the information processing device 1 according to the embodiment will be described. Fig. 4 is a flowchart showing an example of information processing by the processing unit 12 of the information processing device 1 according to the embodiment.

[0093] 4, the processing unit 12 of the information processing device 1 determines whether or not numeric group information has been acquired (step S10). If it determines that numeric group information has been acquired (step S10: Yes), the processing unit 12 uses a generation AI to generate candidate purpose information indicating a plurality of candidate purpose information that are candidates for the purpose of use of the numeric group indicated by the numeric group information acquired in step S11 (step S11), and outputs the generated candidate purpose information (step S12).

[0094] The processing unit 12 acquires the purpose information transmitted in response to the output of the use purpose candidate information in step S12 (step S13). Then, the processing unit 12 generates output information candidates, which are candidates for output information corresponding to the numerical value group information acquired in step S11, using multiple types of generation AI (step S14).

[0095] Next, the processing unit 12 estimates the importance of each output information candidate for the purpose of use indicated by the purpose information acquired in step S13 (step S15). Then, the processing unit 12 selects one or more output information candidates as output information based on the importance estimated in step S16 (step S16), and outputs the selected output information (step S17).

[0096] When the processing of step S17 is completed or when it is determined that the numerical value group information has not been acquired (step S10: No), the processing unit 12 determines whether or not the operation end timing has arrived (step S18). The processing unit 12 determines that the operation end timing has arrived when, for example, the power supply of the information processing device 1 is turned off.

[0097] If the processing unit 12 determines that the operation end time has not yet arrived (step S18: No), it proceeds to step S10, and if it determines that the operation end time has arrived (step S18: Yes), it terminates the processing shown in Figure 4.

[0098] [5. Modifications] The numeric value group information acquired by the acquisition unit 30 may include, as a header, information on attribute items that indicate the type of numeric value, which enables the generation unit 31 to more accurately output output information that corresponds to the purpose of use indicated by the purpose information.

[0099] In addition, the generation unit 31 can, for example, obtain various numerical values ​​stored in a database or the like with headers (attribute items) attached from the database, and based on the various numerical values ​​with headers, have the generation AI determine or learn the range of the numerical values, the number of digits of the numerical values, the numerical expression format, etc. for each attribute item.

[0100] The processing unit 12 can also have the function of a determination unit that, for example, determines the range of numeric values, the number of digits of numeric values, and the expression format of numeric values ​​indicated in the numeric value group information for each type of numeric value (for example, types of numeric values ​​separated by commas) using a generation AI or by using pattern matching processing. In this case, the generation unit 31 inputs information to the generation AI including information indicating the range of numeric values, the number of digits of numeric values, and the expression format of numeric values ​​for each type of usage purpose determined by the determination unit, as well as instruction information instructing the output of multiple usage purpose candidates, and causes the generation AI to output the usage purpose candidate information.

[0101] [6. Hardware Configuration] The information processing device 1 according to the embodiment described above is realized by, for example, a computer 80 configured as shown in Fig. 5. Fig. 5 is a hardware configuration diagram showing an example of the computer 80 that realizes the functions of the information processing device 1 according to the embodiment. The computer 80 has a CPU 81, a RAM 82, a ROM (Read Only Memory) 83, an HDD (Hard Disk Drive) 84, a communication interface (I / F) 85, an input / output interface (I / F) 86, and a media interface (I / F) 87.

[0102] The CPU 81 operates and controls each part based on programs stored in the ROM 83 or the HDD 84. The ROM 83 stores a boot program executed by the CPU 81 when the computer 80 starts up, programs that depend on the hardware of the computer 80, and the like.

[0103] The HDD 84 stores programs executed by the CPU 81, data used by such programs, etc. The communication interface 85 receives data from other devices via the network N (see FIG. 2) and sends it to the CPU 81, and transmits data generated by the CPU 81 to other devices via the network N.

[0104] The CPU 81 controls output devices such as a display and a printer, and input devices such as a keyboard or a mouse, via the input / output interface 86. The CPU 81 acquires data from the input devices via the input / output interface 86. The CPU 81 also outputs generated data to the output devices via the input / output interface 86.

[0105] The media interface 87 reads a program or data stored in a recording medium 88 and provides it to the CPU 81 via the RAM 82. The CPU 81 loads the program or data from the recording medium 88 onto the RAM 82 via the media interface 87 and executes the loaded program. The recording medium 88 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or a PD (Phase Change Rewritable Disc), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory.

[0106] For example, when the computer 80 functions as the information processing device 1 according to the embodiment, the CPU 81 of the computer 80 executes programs loaded onto the RAM 82 to realize the functions of the processing unit 12. In addition, the HDD 84 stores data in the storage unit 11. The CPU 81 of the computer 80 reads and executes these programs from a recording medium 88, but as another example, the CPU 81 may obtain these programs from another device via the network N.

[0107] [7. Other] Furthermore, among the processes described in the above embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically using known methods. In addition, the information including the processing procedures, specific names, various data, and parameters shown in the above documents and drawings can be changed as desired unless otherwise specified. For example, the various information shown in each drawing is not limited to the information shown in the drawings.

[0108] Furthermore, the components of each device shown in the figure are conceptual functional components and do not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figure, and all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc.

[0109] For example, the information processing device 1 described above may be realized by a terminal device and a server computer, or may be realized by multiple server computers. Furthermore, depending on the function, the configuration can be flexibly changed, such as by calling an external platform using an API or network computing.

[0110] Furthermore, the above-described embodiments and modifications can be combined as appropriate within the scope of not causing any contradiction in the processing content.

[0111] [8. Effects] As described above, the information processing device 1 according to the embodiment includes an acquisition unit 30, a generation unit 31, and an output unit 32. The acquisition unit 30 acquires numerical group information indicating a numerical group and purpose information indicating a purpose for which the numerical group is to be used. The generation unit 31 generates output information according to the purpose of use using a generation AI based on the numerical group information and purpose information acquired by the acquisition unit 30. The output unit 32 outputs the output information generated by the generation unit 31. This allows the information processing device 1 to generate an appropriate output for an input numerical group using the generation AI.

[0112] The acquisition unit 30 also includes example information that exemplifies output information according to the intended use of the group of numerical values, and the generation unit 31 uses the generation AI to generate output information according to the intended use, based on the group of numerical values ​​information, purpose information, and example information acquired by the acquisition unit 30. This allows the information processing device 1 to generate a more appropriate output for the group of numerical values ​​that is input using the generation AI.

[0113] Furthermore, the generation unit 31 uses the generation AI to generate candidate purpose information indicating multiple candidates for the purpose of use of the group of numeric values ​​indicated by the numeric value group information acquired by the acquisition unit 30, the output unit 32 outputs the candidate purpose information in a selectable manner, with one or more of the multiple candidate purposes of use being selected, and the acquisition unit 30 acquires information on the one or more candidates selected from the multiple candidates as purpose information. This allows the information processing device 1 to generate a more appropriate output for the group of numeric values ​​input using the generation AI.

[0114] The generating AI is generated using learning information that includes, for each combination of the numerical group information and the purpose information, numerical group information indicating a numerical group, purpose information indicating a purpose of use of the numerical group, and output information corresponding to the purpose of use. This allows the information processing device 1 to generate a more appropriate output for an input numerical group using the generating AI.

[0115] The generation unit 31 also includes a first generation processing unit 40 that uses each of a plurality of types of generation AI to generate output information candidates that are candidates for output information according to the numerical value group information acquired by the acquisition unit 30, an estimation processing unit 41 that estimates the importance of each of the plurality of output information candidates generated by the first generation processing unit 40 with respect to the purpose of use indicated by the purpose information, and a determination processing unit 42 that determines one or more output information candidates from the plurality of output information candidates as output information based on the importance estimated by the estimation processing unit 41. This allows the information processing device 1 to generate a more appropriate output for the input numerical value group using the generation AI.

[0116] Furthermore, the multiple types of generation AIs are generated to output output information candidates for different types of numerical value groups, which allows the information processing device 1 to generate more appropriate output for the input numerical value group using the generation AI.

[0117] The generation unit 31 also includes a first generation processing unit 40 that causes the generation AI to generate processing information indicating a type of statistical processing according to the purpose of use indicated by the objective information, and a second generation processing unit 43 that generates output information according to the purpose of use by performing statistical processing on the group of numbers indicated by the number group information using the statistical processing indicated by the processing information. This allows the information processing device 1 to generate a more appropriate output for the group of numbers that is input using the generation AI.

[0118] The above describes the embodiments of the present application in detail based on the drawings, but this is merely an example, and the present invention can be implemented in other forms that include the embodiments described in the Disclosure of the Invention section and that have been modified and improved in various ways based on the knowledge of those skilled in the art.

[0119] Furthermore, the above-mentioned "section, module, unit" can be read as "means" or "circuit," etc. For example, an acquisition unit can be read as an acquisition means or an acquisition circuit. [Explanation of symbols]

[0120] 1. Information processing equipment 2. Terminal Device 10. Communications Department 11 Storage section 12 Processing section 30 Acquisition Department 31 Generation part 32 Output section 40 First generation processing unit 41 Estimation processing unit 42 Decision Processing Unit 43 Second generation processing unit 100 Information Processing Systems N Network

Claims

1. an acquisition unit that acquires numerical group information indicating a numerical group and purpose information indicating a purpose for using the numerical group; a generation unit that generates output information according to the purpose of use using a generation AI based on the numerical group information and the purpose information acquired by the acquisition unit; an output unit that outputs the output information generated by the generation unit; 1. An information processing device comprising:

2. The acquisition unit including example information showing examples of output information according to the purpose of use of the group of numerical values; The generation unit Generate output information according to the purpose of use using the generation AI based on the numerical group information, the objective information, and the example information acquired by the acquisition unit.

2. The information processing apparatus according to claim 1, wherein:

3. The generation unit generating, using a generation AI, candidate use purpose information indicating a plurality of candidate use purposes of the numerical group indicated by the numerical group information acquired by the acquisition unit; The output unit outputting the candidate purpose of use information so that one or more candidates from among the plurality of candidate purposes of use can be selected; The acquisition unit Information on the one or more candidates selected from the plurality of candidates is acquired as the target information.

3. The information processing apparatus according to claim 1, wherein the information processing apparatus is a computer.

4. The generated AI is The AI ​​is generated using learning information including, for each combination of the numerical group information and the purpose information, numerical group information indicating a numerical group, purpose information indicating a purpose of use of the numerical group, and output information according to the purpose of use.

3. The information processing apparatus according to claim 1, wherein the information processing apparatus is a computer.

5. The generation unit a first generation processing unit that generates output information candidates that are output information candidates according to the numerical value group information acquired by the acquisition unit, using each of a plurality of types of generation AIs; an estimation processing unit that estimates the importance of each of the plurality of output information candidates generated by the first generation processing unit with respect to the purpose of use indicated by the purpose information; a determination processing unit that determines one or more output information candidates from among the plurality of output information candidates as the output information based on the importance estimated by the estimation processing unit.

3. The information processing apparatus according to claim 1, wherein the information processing apparatus is a computer.

6. The plurality of types of generated AIs are: The output information candidates are generated so as to be output for groups of numerical values ​​of different types.

6. The information processing apparatus according to claim 5,

7. The generation unit a first generation processing unit that causes the generation AI to generate processing information indicating a type of statistical processing according to the purpose of use indicated by the purpose information; a second generation processing unit that generates output information according to the purpose of use by performing statistical processing on the group of numerical values ​​indicated by the numerical value group information by the statistical processing indicated by the processing information; 3. The information processing apparatus according to claim 1, wherein the information processing apparatus is a computer.

8. 1. A computer-implemented information processing method, comprising: an acquisition step of acquiring numerical group information indicating a numerical group and purpose information indicating a purpose for using the numerical group; a generation step of generating output information according to the purpose of use using a generation AI based on the numerical group information and the purpose information acquired by the acquisition step; an output step of outputting the output information generated by the generation step. An information processing method comprising:

9. an acquisition procedure for acquiring numerical group information indicating a numerical group and purpose information indicating a purpose for using the numerical group; a generation step of generating output information according to the purpose of use using a generation AI based on the numerical group information and the purpose information acquired by the acquisition step; an output step of outputting the output information generated by the generation step, An information processing program characterized by:

Images