System

Abstract

An object of a system according to an embodiment is to clarify a determination criterion of AI and visualize the determination basis.SOLUTION: A system according to an embodiment includes a clarifying unit, a visualizing unit, and an analyzing unit. The clarifying unit clarifies a criterion of the AI. The visualization unit visualizes the determination basis of the AI based on the determination criterion clarified by the clarification unit. The analysis unit identifies and analyzes an erroneous result on the basis of the determination basis visualized by the visualization unit.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] With conventional technology, the AI's decision-making criteria were treated as a black box, making it difficult to understand the basis for the decisions.

[0005] The system according to the embodiment aims to clarify the AI's decision criteria and visualize the basis for those decisions. [Means for solving the problem]

[0006] The system according to the embodiment includes a clarification unit, a visualization unit, and an analysis unit. The clarification unit clarifies the AI's judgment criteria. The visualization unit visualizes the AI's judgment basis based on the judgment criteria clarified by the clarification unit. The analysis unit identifies and analyzes erroneous results based on the judgment basis visualized by the visualization unit. [Effects of the Invention]

[0007] The system according to the embodiment can clarify the AI's decision criteria and visualize the basis for those decisions. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. DETAILED DESCRIPTION OF THE INVENTION

[0009] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0010] First, the terms used in the following description will be explained.

[0011] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, the processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (Tensor Processing Unit).

[0012] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0013] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0014] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), and Bluetooth (registered trademark).

[0015] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0016] [First embodiment] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0017] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0018] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0019] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0020] The reception device 38 includes a touch panel 38A and a microphone 38B, and receives user input. The touch panel 38A detects contact with a pointer (for example, a pen or a finger) to receive user input by the touch of the pointer. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 (see FIG. 2) acquires the data indicating the user input.

[0021] Output device 40 includes a display 40A and a speaker 40B, and presents data to a user by outputting the data in a form of expression that the user can perceive (e.g., audio and / or text). Display 40A displays visible information such as text and images in accordance with instructions from processor 46. Speaker 40B outputs audio in accordance with instructions from processor 46. Camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0022] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0023] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0024] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0025] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0026] In the smart device 14, the specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The specific processing program 60 is used together with the specific processing program 56 by the data processing system 10. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. Note that the smart device 14 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the specific processing unit 290 using these models.

[0027] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device (e.g., a generation server) may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (prediction result, etc.) using the data generation model 58 by communicating with the server device having the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device owned by a user (e.g., a mobile phone, a robot, a home appliance, etc.). Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example 1) A system according to an embodiment of the present invention explains the decision-making process of a machine learning model in terms that are easy for humans to understand. This system clarifies the AI's decision-making criteria, visualizes the basis for black-box AI decisions, and identifies and analyzes erroneous results. For example, the system analyzes the data on which the AI ​​bases its decisions and explains the decision-making criteria in terms that are easy for humans to understand. Next, the system visually displays the process by which the AI ​​made its decision. Furthermore, if the AI ​​makes an incorrect decision, the system identifies the cause and analyzes which part of the data caused the error. This makes the AI's decision-making process transparent, making it easier for users to understand the decision. Furthermore, by helping to identify and analyze erroneous results, the system can contribute to improving the model. This system clarifies and visualizes the AI's decision-making criteria, and helps to identify and analyze erroneous results, thereby improving the reliability and understandability of the AI. For example, by making the AI's decision-making process transparent, users can easily understand the AI's decisions, allowing them to use AI as a more reliable tool. Furthermore, by helping to identify and analyze erroneous results, the system contributes to improving the model and enabling more accurate decisions.

[0029] The system according to the embodiment includes a clarification unit, a visualization unit, and an analysis unit. The clarification unit clarifies the AI's decision criteria. For example, the clarification unit analyzes the data on which the AI ​​bases its decisions and explains the decision criteria in terms that are easy for humans to understand. The clarification unit can also evaluate the reliability of the data used by the AI ​​and clarify the decision criteria based on the reliable data. The clarification unit can also refer to the history of past decisions made by the AI ​​to confirm the consistency of the decision criteria. For example, the clarification unit analyzes the history of past decisions made by the AI ​​and confirms that the decision criteria are consistent. The visualization unit visualizes the basis for the black-box AI's decision based on the decision criteria clarified by the clarification unit. For example, the visualization unit visually displays the process by which the AI ​​made a decision. The visualization unit can also use graphs and charts to show which words and phrases the AI ​​prioritizes when analyzing text. The visualization unit can also display in detail the process flow used by the AI ​​to make a decision and clearly explain the basis for each step. For example, the visualization unit displays the steps in which the AI ​​collects data and clearly explains the rationale for each step. The analysis unit identifies and analyzes erroneous results based on the rationale for the decision visualized by the visualization unit. For example, when the AI ​​makes an incorrect decision, the analysis unit identifies the cause and analyzes which part of the data caused the error. Furthermore, when the AI ​​makes an incorrect decision, the analysis unit can analyze the cause in detail and propose measures to prevent recurrence. Furthermore, the analysis unit can evaluate the quality of the data used by the AI ​​and analyze the impact of a decline in data quality on an incorrect decision. For example, the analysis unit analyzes the impact of a decline in data quality on an incorrect decision and aims to improve data quality. As a result, the system according to the embodiment can clarify and visualize the AI's decision criteria and identify and analyze erroneous results, thereby improving the reliability and understandability of the AI.

[0030] The clarification unit can analyze what data the AI ​​is basing its decisions on and explain the criteria for its decisions in terms that humans can understand. For example, the clarification unit clarifies which parts of an image the AI ​​prioritizes when performing image recognition. For example, the clarification unit indicates that the AI ​​is focusing on a specific area of ​​an image. The clarification unit can also clarify which words or phrases the AI ​​prioritizes when performing text analysis. For example, the clarification unit can indicate that the AI ​​prioritizes a specific keyword. The clarification unit can also clarify which voice features the AI ​​prioritizes when performing voice recognition. For example, the clarification unit can indicate that the AI ​​prioritizes a specific voice pattern. This allows the AI's decision-making criteria to be explained in terms that humans can easily understand, making the AI's decision-making process more transparent. Some or all of the above-described processing in the clarification unit may be performed using, for example, a generation AI (generator AI or LLM), or may be performed without a generation AI. For example, the clarification unit inputs data used by the AI ​​into the generation AI, and the generation AI explains the decision-making criteria based on that data.

[0031] The visualization unit can visually display the process through which the AI ​​makes a decision. For example, the visualization unit displays a flowchart of the process from when the AI ​​collects data, analyzes it, and makes a final decision. For example, the visualization unit displays the steps by which the AI ​​collects data and clearly shows the rationale for each step. The visualization unit can also display the steps by which the AI ​​analyzes data and clearly shows the rationale for each step. The visualization unit can also display the steps by which the AI ​​makes a final decision and clearly shows the rationale for each step. For example, the visualization unit shows the algorithms and methods used by the AI ​​to analyze data. This allows the rationale for the AI's decision to be visually displayed, allowing the user to intuitively understand. Some or all of the above-described processing in the visualization unit may be performed using, for example, a generative AI (generative AI or LLM), or may be performed without using a generative AI. For example, the visualization unit inputs data to be used by the AI ​​into the generative AI, and the generative AI visualizes the process based on that data.

[0032] The visualization unit can use graphs or charts to show which words and phrases the AI ​​emphasizes when performing text analysis. For example, the visualization unit displays the words and phrases that the AI ​​emphasizes when performing text analysis in a bar graph. For example, the visualization unit indicates that the AI ​​makes a decision based on specific keywords. The visualization unit can also display the words and phrases that the AI ​​emphasizes when performing text analysis in a heat map. For example, the visualization unit indicates that the AI ​​makes a decision based on specific phrases. The visualization unit can also display the words and phrases that the AI ​​emphasizes when performing text analysis in a word cloud. For example, the visualization unit indicates that the AI ​​makes a decision based on specific words. This visually shows the basis for the AI's text analysis decisions, making it easier for users to understand. Some or all of the above-mentioned processing in the visualization unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the visualization unit inputs text data to be used by the AI ​​into the generation AI, and the generation AI generates graphs and charts based on the data.

[0033] If the AI ​​makes an incorrect decision, the analysis unit can identify the cause and analyze which part of the data caused the error. For example, if the AI ​​makes an incorrect decision, the analysis unit analyzes data to identify the cause. For example, the analysis unit evaluates the quality of the data used by the AI ​​and determines whether low-quality data caused the error. The analysis unit can also analyze the algorithms and methods used by the AI ​​and determine whether an algorithm bug or improper settings caused the error. Furthermore, the analysis unit can analyze the data sources used by the AI ​​and determine whether inconsistencies between the data sources caused the error. For example, the analysis unit compares information obtained from different data sources to identify the cause of the incorrect decision. This can identify the cause of the AI's incorrect decision and help improve the model. Some or all of the above-mentioned processing in the analysis unit may be performed using, for example, the generative AI (generative AI or LLM), or may be performed without the generative AI. For example, the analysis unit inputs the data used by the AI ​​into the generative AI, which then analyzes the cause of the error based on that data.

[0034] The clarification unit can evaluate the reliability of data used by the AI ​​and clarify the judgment criteria based on highly reliable data. For example, the clarification unit evaluates the reliability of data sources used by the AI ​​and uses only highly reliable data. For example, the clarification unit evaluates the origin, consistency, and completeness of the data. The clarification unit can also adjust the weighting of the judgment criteria based on the reliability of the data. For example, the clarification unit weights the judgment criteria based on highly reliable data. Furthermore, the clarification unit can eliminate low-reliability data and reconstruct the judgment criteria based on highly reliable data. For example, the clarification unit evaluates the reliability of data and excludes low-reliability data. This allows the clarification unit to clarify the judgment criteria based on highly reliable data, thereby improving the reliability of the AI's judgment. Some or all of the above-mentioned processing in the clarification unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the clarification unit inputs data used by the AI ​​into the generation AI, which then evaluates the reliability based on the data and clarifies the judgment criteria.

[0035] The clarification unit can refer to the history of past decisions made by the AI ​​and confirm the consistency of the decision criteria. For example, the clarification unit analyzes the history of past decisions made by the AI ​​and confirms the consistency of the decision criteria. For example, the clarification unit analyzes past decision results in chronological order and evaluates the consistency of the decision criteria. The clarification unit can also compare the past decision history with the current decision criteria and make adjustments to maintain consistency. For example, the clarification unit compares the past decision criteria with the current decision criteria and checks for inconsistencies. Furthermore, if an inconsistent decision is found, the clarification unit can identify the cause and modify the decision criteria. For example, the clarification unit analyzes the past decision history and identifies the cause of the inconsistent decision. In this way, the consistency of the decision criteria can be maintained by referring to the past decision history. Some or all of the above-mentioned processing in the clarification unit may be performed using, for example, a generation AI (generation AI or LLM) or without a generation AI. For example, the clarification unit inputs the history of past decisions made by the AI ​​into the generation AI, and the generation AI checks the consistency of the decision criteria based on that data.

[0036] The clarification unit can integrate information acquired by the AI ​​from different data sources and clarify comprehensive judgment criteria. The clarification unit, for example, integrates information acquired from different data sources and constructs comprehensive judgment criteria. For example, the clarification unit integrates internal data and external data and clarifies comprehensive judgment criteria. The clarification unit can also adjust the integrated judgment criteria by taking into account the characteristics of each data source. For example, the clarification unit resolves inconsistencies between data sources and clarifies integrated judgment criteria. Furthermore, the clarification unit can resolve inconsistencies between data sources and clarify integrated judgment criteria. For example, the clarification unit compares information acquired from different data sources and resolves inconsistencies. In this way, comprehensive judgment criteria can be provided by integrating information from different data sources. Some or all of the above-mentioned processing in the clarification unit can be performed, for example, using a generation AI (generation AI or LLM) or without using a generation AI. For example, the clarification unit inputs information obtained by the AI ​​from different data sources into the generation AI, which then integrates the information based on that data and clarifies the judgment criteria.

[0037] The clarification unit can customize the method of explaining the judgment criteria according to the type of data used by the AI. The clarification unit customizes the method of explaining the judgment criteria according to, for example, the type of data used by the AI ​​(image, text, audio, etc.). For example, when image data is used, the clarification unit provides a visual explanation. Furthermore, when text data is used, the clarification unit can also provide a written explanation. Furthermore, when audio data is used, the clarification unit can also provide an audio explanation. For example, the clarification unit explains the judgment criteria using audio data. This makes it possible to improve understandability by providing an explanation method according to the type of data. Some or all of the above-mentioned processing in the clarification unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the clarification unit inputs data used by the AI ​​into the generation AI, and the generation AI customizes the explanation method based on the data.

[0038] When the AI ​​uses data specialized for a particular industry or field, the clarification unit can clarify the industry-specific judgment criteria. For example, when using data from the medical field, the clarification unit clarifies the judgment criteria using medical terminology. For example, the clarification unit analyzes data from the medical field and explains the judgment criteria using medical terminology. Furthermore, when using data from the financial field, the clarification unit can also clarify the judgment criteria using financial terminology. For example, the clarification unit analyzes data from the financial field and explains the judgment criteria using financial terminology. Furthermore, when using data from the manufacturing industry, the clarification unit can also clarify judgment criteria specific to the manufacturing industry. For example, the clarification unit analyzes data from the manufacturing industry and explains judgment criteria specific to the manufacturing industry. This enables more specialized judgments by providing industry-specific judgment criteria. Some or all of the above-mentioned processing in the clarification unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the clarification unit inputs industry-specific data used by the AI ​​into the generation AI, and the generation AI clarifies the judgment criteria based on that data.

[0039] The clarification unit can dynamically update the judgment criteria based on data acquired by the AI ​​in real time. The clarification unit dynamically updates the judgment criteria based on, for example, real-time market data. For example, the clarification unit analyzes real-time market data and updates the judgment criteria based on the latest information. The clarification unit can also dynamically update the judgment criteria based on real-time sensor data. For example, the clarification unit analyzes real-time sensor data and updates the judgment criteria based on the latest information. The clarification unit can also dynamically update the judgment criteria based on real-time user data. For example, the clarification unit analyzes real-time user data and updates the judgment criteria based on the latest information. In this way, by updating the judgment criteria based on the real-time data, it is possible to provide a judgment based on the latest information. Some or all of the above-mentioned processing in the clarification unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the clarification unit inputs data acquired by the AI ​​in real time into the generation AI, and the generation AI dynamically updates the judgment criteria based on the data.

[0040] The visualization unit can display in detail the process flow when the AI ​​makes a decision and clearly indicate the basis for each step. For example, the visualization unit can display the steps in which the AI ​​collects data and clearly indicate the basis for each step. For example, the visualization unit can show the sensors and data sources used by the AI ​​to collect data. The visualization unit can also display the steps in which the AI ​​analyzes data and clearly indicate the basis for each step. For example, the visualization unit can show the algorithms and methods used by the AI ​​to analyze data. The visualization unit can also display the steps in which the AI ​​makes a final decision and clearly indicate the basis for each step. For example, the visualization unit can show the factors and criteria considered when the AI ​​makes a final decision. This makes it easier for users to understand the AI's decision-making process by displaying it in detail. Some or all of the above-mentioned processing in the visualization unit can be performed using, for example, a generative AI (generative AI or LLM), or can be performed without using a generative AI. For example, the visualization unit inputs data used by the AI ​​into the generative AI, and the generative AI displays a detailed process flow based on that data.

[0041] The visualization unit can change the visualization format depending on the type of algorithm used by the AI. For example, when a deep learning algorithm is used, the visualization unit visualizes the structure of the neural network. For example, the visualization unit shows each layer and node of the neural network. Furthermore, when a decision tree algorithm is used, the visualization unit can also visualize the branches of the decision tree. For example, the visualization unit shows each node and branching condition of the decision tree. Furthermore, when a clustering algorithm is used, the visualization unit can also visualize the distribution of clusters. For example, the visualization unit shows the centers of clusters and the distribution of data points. This makes it possible to improve understandability by providing a visualization format appropriate for the type of algorithm. Some or all of the above-mentioned processing in the visualization unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the visualization unit inputs data of the algorithm used by the AI ​​into the generation AI, and the generation AI changes the visualization format based on the data.

[0042] The visualization unit can compare past and current judgment results made by the AI ​​and visualize trends in change. For example, the visualization unit compares past and current judgment results in a graph to show trends in change. For example, the visualization unit compares past and current judgment results using time-series data. The visualization unit can also compare past and current judgment results in a heat map to show trends in change. For example, the visualization unit visually shows changes in data using a heat map. Furthermore, the visualization unit can compare past and current judgment results in a time series to show trends in change. For example, the visualization unit shows changes in data using trend lines. This makes it easier to understand trends in change by comparing past and current judgment results. Some or all of the above-described processing in the visualization unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the visualization unit inputs past and current judgment results made by the AI ​​into the generation AI, and the generation AI visualizes trends in change based on the data.

[0043] The visualization unit can customize the visualization format depending on the type of data used by the AI. For example, when image data is used, the visualization unit visualizes the data in a heat map format. For example, the visualization unit displays important parts of the image data using a heat map. Furthermore, when text data is used, the visualization unit can also visualize the data in a word cloud format. For example, the visualization unit displays important words in the text data using a word cloud. Furthermore, when audio data is used, the visualization unit can also visualize the data in a waveform format. For example, the visualization unit displays the characteristics of the audio data using waveforms. This makes it possible to improve understandability by providing a visualization format appropriate for the type of data. Some or all of the above-described processing in the visualization unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the visualization unit inputs data to be used by the AI ​​into the generation AI, and the generation AI customizes the visualization format based on the data.

[0044] When the AI ​​uses data specialized for a particular industry or field, the visualization unit can provide a visualization format specific to that industry. For example, when using data in the medical field, the visualization unit visualizes the data in a medical graph or chart format. For example, the visualization unit displays important parts of the medical data in a medical graph. Furthermore, when using data in the financial field, the visualization unit can also visualize the data in a financial chart or graph format. For example, the visualization unit displays important parts of the financial data in a financial chart. Furthermore, when using data in the manufacturing industry, the visualization unit can also visualize the data in a flowchart format of the manufacturing process. For example, the visualization unit displays important parts of the manufacturing data in a flowchart. In this way, by providing an industry-specific visualization format, more specialized information can be displayed. Some or all of the above-mentioned processing in the visualization unit may be performed using, for example, a generative AI (generative AI or LLM), or may be performed without using a generative AI. For example, the visualization unit inputs industry-specific data used by the AI ​​into the generative AI, and the generative AI provides a visualization format based on that data.

[0045] The visualization unit can dynamically update the visualization content based on data acquired by the AI ​​in real time. The visualization unit dynamically updates the visualization content based on, for example, real-time market data. For example, the visualization unit analyzes real-time market data and updates the visualization content based on the latest information. The visualization unit can also dynamically update the visualization content based on real-time sensor data. For example, the visualization unit analyzes real-time sensor data and updates the visualization content based on the latest information. The visualization unit can also dynamically update the visualization content based on real-time user data. For example, the visualization unit analyzes real-time user data and updates the visualization content based on the latest information. In this way, by updating the visualization content based on the real-time data, it is possible to provide the latest information. Some or all of the above-described processing in the visualization unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the visualization unit inputs data acquired by the AI ​​in real time into the generation AI, and the generation AI dynamically updates the visualization content based on the data.

[0046] When the AI ​​makes an incorrect decision, the analysis unit can analyze the cause in detail and propose measures to prevent recurrence. For example, the analysis unit identifies the cause of the incorrect decision and proposes measures to prevent recurrence. For example, the analysis unit makes proposals to improve data quality. The analysis unit can also propose improvements to the algorithm. For example, the analysis unit fixes bugs in the algorithm and proposes measures to prevent recurrence. Furthermore, the analysis unit can evaluate data quality and eliminate low-quality data. For example, the analysis unit evaluates the consistency and completeness of the data and excludes low-quality data. This allows for a detailed analysis of the cause of the incorrect decision and proposes measures to prevent recurrence, thereby improving the accuracy of the AI. Some or all of the above-mentioned processing in the analysis unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the analysis unit inputs data used by the AI ​​into the generation AI, which then analyzes the cause of the error based on that data and proposes measures to prevent recurrence.

[0047] The analysis unit can evaluate the quality of the data used by the AI ​​and analyze the impact of a decline in data quality on erroneous decisions. The analysis unit, for example, analyzes the impact of a decline in data quality on erroneous decisions. For example, the analysis unit evaluates the consistency and completeness of the data and checks whether low-quality data has caused errors. The analysis unit can also evaluate data quality and eliminate low-quality data. For example, the analysis unit evaluates the origin and consistency of the data and eliminates low-quality data. The analysis unit can also evaluate data quality and prioritize the use of high-quality data. For example, the analysis unit evaluates the reliability of the data and uses highly reliable data. This makes it possible to analyze the impact of a decline in data quality on erroneous decisions and improve data quality. Some or all of the above-mentioned processing in the analysis unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the analysis unit inputs data to be used by the AI ​​into the generation AI, and the generation AI evaluates the quality based on the data and analyzes the impact of errors.

[0048] The analysis unit can compare information obtained by the AI ​​from different data sources and identify the cause of an incorrect decision. For example, the analysis unit compares information obtained from different data sources and identifies the cause of an incorrect decision. For example, the analysis unit compares internal data with external data and checks for inconsistencies. The analysis unit can also analyze inconsistencies between data sources and identify the cause of an incorrect decision. For example, the analysis unit integrates information obtained from different data sources and identifies the cause of an incorrect decision. Furthermore, the analysis unit can resolve inconsistencies between data sources and identify the cause of an incorrect decision. For example, the analysis unit compares information obtained from different data sources and resolves inconsistencies. In this way, the cause of an incorrect decision can be identified by comparing information from different data sources. Some or all of the above-mentioned processing in the analysis unit may be performed using, for example, the generation AI (generation AI or LLM), or may be performed without using the generation AI. For example, the analysis unit inputs information from different data sources used by the AI ​​into the generation AI, and the generation AI identifies the cause of an error based on that data.

[0049] The analysis unit can customize the method for identifying erroneous results depending on the type of data used by the AI. For example, when image data is used, the analysis unit identifies erroneous results using image analysis. For example, the analysis unit analyzes the characteristics of the image data to identify errors. Furthermore, when text data is used, the analysis unit can also identify erroneous results using text analysis. For example, the analysis unit analyzes the content of the text data to identify errors. Furthermore, when audio data is used, the analysis unit can also identify erroneous results using audio analysis. For example, the analysis unit analyzes the characteristics of the audio data to identify errors. This allows for accurate identification of erroneous results by providing an identification method according to the type of data. Some or all of the above-mentioned processing in the analysis unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the analysis unit inputs data to be used by the AI ​​into the generation AI, and the generation AI customizes the identification method based on that data.

[0050] When the AI ​​uses data specialized for a particular industry or field, the analysis unit can analyze error patterns specific to that industry. For example, when data from the medical field is used, the analysis unit analyzes error patterns specific to the medical field. For example, the analysis unit analyzes and identifies error patterns in the medical data. Furthermore, when data from the financial field is used, the analysis unit can also analyze error patterns specific to the financial field. For example, the analysis unit analyzes and identifies error patterns in the financial data. Furthermore, when data from the manufacturing industry is used, the analysis unit can also analyze error patterns specific to the manufacturing industry. For example, the analysis unit analyzes and identifies error patterns in the manufacturing data. In this way, by analyzing industry-specific error patterns, erroneous results can be quickly identified. Some or all of the above-mentioned processing in the analysis unit may be performed using, for example, the generation AI (generation AI or LLM), or may be performed without using the generation AI. For example, the analysis unit inputs industry-specific data used by the AI ​​into the generation AI, and the generation AI analyzes error patterns based on that data.

[0051] The analysis unit can dynamically identify and analyze erroneous results based on data acquired by the AI ​​in real time. The analysis unit dynamically identifies and analyzes erroneous results based on, for example, real-time market data. For example, the analysis unit analyzes real-time market data to identify errors. The analysis unit can also dynamically identify and analyze erroneous results based on real-time sensor data. For example, the analysis unit analyzes real-time sensor data to identify errors. The analysis unit can also dynamically identify and analyze erroneous results based on real-time user data. For example, the analysis unit analyzes real-time user data to identify errors. This enables rapid response by identifying and analyzing erroneous results based on real-time data. Some or all of the above-described processing in the analysis unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the analysis unit inputs data acquired by the AI ​​in real time into the generation AI, and the generation AI identifies and analyzes errors based on the data.

[0052] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0053] The clarification unit can evaluate the reliability of data used by the AI ​​and clarify the judgment criteria based on highly reliable data. For example, the clarification unit evaluates the origin, consistency, and completeness of the data and uses only highly reliable data. The clarification unit can also adjust the weighting of the judgment criteria based on the reliability of the data. Furthermore, the clarification unit can eliminate low-reliability data and reconstruct the judgment criteria based on highly reliable data. In this way, by clarifying the judgment criteria based on highly reliable data, the reliability of the AI's judgments can be improved.

[0054] The visualization unit can change the visualization format depending on the type of algorithm used by the AI. For example, if a deep learning algorithm is used, the structure of the neural network can be visualized. Also, if a decision tree algorithm is used, the branching of the decision tree can be visualized. Furthermore, if a clustering algorithm is used, the distribution of clusters can be visualized. This makes it easier to understand by providing a visualization format that suits the type of algorithm.

[0055] The analysis unit can compare information obtained by the AI ​​from different data sources and identify the cause of incorrect decisions. For example, it can compare internal data with external data to check for inconsistencies. It can also analyze inconsistencies between data sources and identify the cause of incorrect decisions. It can also resolve inconsistencies between data sources and identify the cause of incorrect decisions. This makes it possible to identify the cause of incorrect decisions by comparing information from different data sources.

[0056] The clarification unit can refer to the history of past decisions made by the AI ​​and check the consistency of the decision criteria. For example, it can analyze past decision results in chronological order and evaluate the consistency of the decision criteria. It can also compare the past decision history with the current decision criteria and make adjustments to maintain consistency. Furthermore, if an inconsistent decision is found, it can identify the cause and revise the decision criteria. In this way, by referring to the past decision history, it is possible to maintain the consistency of the decision criteria.

[0057] The visualization unit can compare the AI's past and current judgment results and visualize trends in change. For example, past and current judgment results can be compared in a graph to show trends in change. Past and current judgment results can also be compared in a heat map to show trends in change. Furthermore, past and current judgment results can be compared in chronological order to show trends in change. This makes it easier to understand trends in change by comparing past and current judgment results.

[0058] The processing flow of the first embodiment will be briefly explained below.

[0059] Step 1: The clarification unit clarifies the AI's decision-making criteria. For example, the clarification unit analyzes what data the AI ​​uses to make its decisions and explains the decision-making criteria in terms that are easy for humans to understand. The clarification unit can also evaluate the reliability of the data used by the AI ​​and clarify the decision-making criteria based on highly reliable data. Furthermore, the clarification unit can refer to the history of past decisions made by the AI ​​to confirm the consistency of the decision-making criteria. Step 2: The visualization unit visualizes the basis for the black-box AI's decision based on the criteria clarified by the clarification unit. For example, the visualization unit visually displays the process the AI ​​went through to make its decision. The visualization unit can also use graphs and charts to show which words and phrases the AI ​​prioritizes when analyzing text. Furthermore, the visualization unit can display in detail the process flow when the AI ​​makes a decision and clearly state the basis for the decision at each step. Step 3: The analysis unit identifies and analyzes erroneous results based on the decision-making basis visualized by the visualization unit. For example, if the AI ​​makes an incorrect decision, the analysis unit identifies the cause and analyzes which part of the data caused the error. The analysis unit can also analyze the cause in detail when the AI ​​makes an incorrect decision and propose measures to prevent recurrence. Furthermore, the analysis unit can evaluate the quality of the data used by the AI ​​and analyze the impact that a decline in data quality has on erroneous decisions.

[0060] (Example 2) A system according to an embodiment of the present invention explains the decision-making process of a machine learning model in terms that are easy for humans to understand. This system clarifies the AI's decision-making criteria, visualizes the basis for black-box AI decisions, and identifies and analyzes erroneous results. For example, the system analyzes the data on which the AI ​​bases its decisions and explains the decision-making criteria in terms that are easy for humans to understand. Next, the system visually displays the process by which the AI ​​made its decision. Furthermore, if the AI ​​makes an incorrect decision, the system identifies the cause and analyzes which part of the data caused the error. This makes the AI's decision-making process transparent, making it easier for users to understand the decision. Furthermore, by helping to identify and analyze erroneous results, the system can contribute to improving the model. This system clarifies and visualizes the AI's decision-making criteria, and helps to identify and analyze erroneous results, thereby improving the reliability and understandability of the AI. For example, by making the AI's decision-making process transparent, users can easily understand the AI's decisions, allowing them to use AI as a more reliable tool. Furthermore, by helping to identify and analyze erroneous results, the system contributes to improving the model and enabling more accurate decisions.

[0061] The system according to the embodiment includes a clarification unit, a visualization unit, and an analysis unit. The clarification unit clarifies the AI's decision criteria. For example, the clarification unit analyzes the data on which the AI ​​bases its decisions and explains the decision criteria in terms that are easy for humans to understand. The clarification unit can also evaluate the reliability of the data used by the AI ​​and clarify the decision criteria based on the reliable data. The clarification unit can also refer to the history of past decisions made by the AI ​​to confirm the consistency of the decision criteria. For example, the clarification unit analyzes the history of past decisions made by the AI ​​and confirms that the decision criteria are consistent. The visualization unit visualizes the basis for the black-box AI's decision based on the decision criteria clarified by the clarification unit. For example, the visualization unit visually displays the process by which the AI ​​made a decision. The visualization unit can also use graphs and charts to show which words and phrases the AI ​​prioritizes when analyzing text. The visualization unit can also display in detail the process flow used by the AI ​​to make a decision and clearly explain the basis for each step. For example, the visualization unit displays the steps in which the AI ​​collects data and clearly explains the rationale for each step. The analysis unit identifies and analyzes erroneous results based on the rationale for the decision visualized by the visualization unit. For example, when the AI ​​makes an incorrect decision, the analysis unit identifies the cause and analyzes which part of the data caused the error. Furthermore, when the AI ​​makes an incorrect decision, the analysis unit can analyze the cause in detail and propose measures to prevent recurrence. Furthermore, the analysis unit can evaluate the quality of the data used by the AI ​​and analyze the impact of a decline in data quality on an incorrect decision. For example, the analysis unit analyzes the impact of a decline in data quality on an incorrect decision and aims to improve data quality. As a result, the system according to the embodiment can clarify and visualize the AI's decision criteria and identify and analyze erroneous results, thereby improving the reliability and understandability of the AI.

[0062] The clarification unit can analyze what data the AI ​​is basing its decisions on and explain the criteria for its decisions in terms that humans can understand. For example, the clarification unit clarifies which parts of an image the AI ​​prioritizes when performing image recognition. For example, the clarification unit indicates that the AI ​​is focusing on a specific area of ​​an image. The clarification unit can also clarify which words or phrases the AI ​​prioritizes when performing text analysis. For example, the clarification unit can indicate that the AI ​​prioritizes a specific keyword. The clarification unit can also clarify which voice features the AI ​​prioritizes when performing voice recognition. For example, the clarification unit can indicate that the AI ​​prioritizes a specific voice pattern. This allows the AI's decision-making criteria to be explained in terms that humans can easily understand, making the AI's decision-making process more transparent. Some or all of the above-described processing in the clarification unit may be performed using, for example, a generation AI (generator AI or LLM), or may be performed without a generation AI. For example, the clarification unit inputs data used by the AI ​​into the generation AI, and the generation AI explains the decision-making criteria based on that data.

[0063] The visualization unit can visually display the process through which the AI ​​makes a decision. For example, the visualization unit displays a flowchart of the process from when the AI ​​collects data, analyzes it, and makes a final decision. For example, the visualization unit displays the steps by which the AI ​​collects data and clearly shows the rationale for each step. The visualization unit can also display the steps by which the AI ​​analyzes data and clearly shows the rationale for each step. The visualization unit can also display the steps by which the AI ​​makes a final decision and clearly shows the rationale for each step. For example, the visualization unit shows the algorithms and methods used by the AI ​​to analyze data. This allows the rationale for the AI's decision to be visually displayed, allowing the user to intuitively understand. Some or all of the above-described processing in the visualization unit may be performed using, for example, a generative AI (generative AI or LLM), or may be performed without using a generative AI. For example, the visualization unit inputs data to be used by the AI ​​into the generative AI, and the generative AI visualizes the process based on that data.

[0064] The visualization unit can use graphs or charts to show which words and phrases the AI ​​emphasizes when performing text analysis. For example, the visualization unit displays the words and phrases that the AI ​​emphasizes when performing text analysis in a bar graph. For example, the visualization unit indicates that the AI ​​makes a decision based on specific keywords. The visualization unit can also display the words and phrases that the AI ​​emphasizes when performing text analysis in a heat map. For example, the visualization unit indicates that the AI ​​makes a decision based on specific phrases. The visualization unit can also display the words and phrases that the AI ​​emphasizes when performing text analysis in a word cloud. For example, the visualization unit indicates that the AI ​​makes a decision based on specific words. This visually shows the basis for the AI's text analysis decisions, making it easier for users to understand. Some or all of the above-mentioned processing in the visualization unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the visualization unit inputs text data to be used by the AI ​​into the generation AI, and the generation AI generates graphs and charts based on the data.

[0065] If the AI ​​makes an incorrect decision, the analysis unit can identify the cause and analyze which part of the data caused the error. For example, if the AI ​​makes an incorrect decision, the analysis unit analyzes data to identify the cause. For example, the analysis unit evaluates the quality of the data used by the AI ​​and determines whether low-quality data caused the error. The analysis unit can also analyze the algorithms and methods used by the AI ​​and determine whether an algorithm bug or improper settings caused the error. Furthermore, the analysis unit can analyze the data sources used by the AI ​​and determine whether inconsistencies between the data sources caused the error. For example, the analysis unit compares information obtained from different data sources to identify the cause of the incorrect decision. This can identify the cause of the AI's incorrect decision and help improve the model. Some or all of the above-mentioned processing in the analysis unit may be performed using, for example, the generative AI (generative AI or LLM), or may be performed without the generative AI. For example, the analysis unit inputs the data used by the AI ​​into the generative AI, which then analyzes the cause of the error based on that data.

[0066] The clarification unit can estimate the user's emotions and adjust the explanation method for the judgment criteria based on the estimated user emotions. For example, if the user is feeling stressed, the clarification unit can provide a simple and intuitive explanation to facilitate understanding. For example, the clarification unit can avoid technical terms and use concise expressions. Furthermore, if the user is relaxed, the clarification unit can provide a detailed explanation to promote deeper understanding. For example, the clarification unit can provide an explanation including detailed data and analysis results. Furthermore, if the user is in a hurry, the clarification unit can provide a concise explanation that focuses on the main points to facilitate quick understanding. For example, the clarification unit can emphasize only the important points to enable understanding in a short time. This can improve ease of understanding by providing an explanation method that corresponds to the user's emotions. Emotion estimation is realized using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the clarification unit can be performed using, for example, a generation AI (generation AI or LLM) or without a generation AI. For example, the clarification unit inputs the user's emotional data into the generation AI, which then adjusts its explanation method based on that data.

[0067] The clarification unit can evaluate the reliability of data used by the AI ​​and clarify the judgment criteria based on highly reliable data. For example, the clarification unit evaluates the reliability of data sources used by the AI ​​and uses only highly reliable data. For example, the clarification unit evaluates the origin, consistency, and completeness of the data. The clarification unit can also adjust the weighting of the judgment criteria based on the reliability of the data. For example, the clarification unit weights the judgment criteria based on highly reliable data. Furthermore, the clarification unit can eliminate low-reliability data and reconstruct the judgment criteria based on highly reliable data. For example, the clarification unit evaluates the reliability of data and excludes low-reliability data. This allows the clarification unit to clarify the judgment criteria based on highly reliable data, thereby improving the reliability of the AI's judgment. Some or all of the above-mentioned processing in the clarification unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the clarification unit inputs data used by the AI ​​into the generation AI, which then evaluates the reliability based on the data and clarifies the judgment criteria.

[0068] The clarification unit can refer to the history of past decisions made by the AI ​​and confirm the consistency of the decision criteria. For example, the clarification unit analyzes the history of past decisions made by the AI ​​and confirms the consistency of the decision criteria. For example, the clarification unit analyzes past decision results in chronological order and evaluates the consistency of the decision criteria. The clarification unit can also compare the past decision history with the current decision criteria and make adjustments to maintain consistency. For example, the clarification unit compares the past decision criteria with the current decision criteria and checks for inconsistencies. Furthermore, if an inconsistent decision is found, the clarification unit can identify the cause and modify the decision criteria. For example, the clarification unit analyzes the past decision history and identifies the cause of the inconsistent decision. In this way, the consistency of the decision criteria can be maintained by referring to the past decision history. Some or all of the above-mentioned processing in the clarification unit may be performed using, for example, a generation AI (generation AI or LLM) or without a generation AI. For example, the clarification unit inputs the history of past decisions made by the AI ​​into the generation AI, and the generation AI checks the consistency of the decision criteria based on that data.

[0069] The clarification unit can integrate information acquired by the AI ​​from different data sources and clarify comprehensive judgment criteria. The clarification unit, for example, integrates information acquired from different data sources and constructs comprehensive judgment criteria. For example, the clarification unit integrates internal data and external data and clarifies comprehensive judgment criteria. The clarification unit can also adjust the integrated judgment criteria by taking into account the characteristics of each data source. For example, the clarification unit resolves inconsistencies between data sources and clarifies integrated judgment criteria. Furthermore, the clarification unit can resolve inconsistencies between data sources and clarify integrated judgment criteria. For example, the clarification unit compares information acquired from different data sources and resolves inconsistencies. In this way, comprehensive judgment criteria can be provided by integrating information from different data sources. Some or all of the above-mentioned processing in the clarification unit can be performed, for example, using a generation AI (generation AI or LLM) or without using a generation AI. For example, the clarification unit inputs information obtained by the AI ​​from different data sources into the generation AI, which then integrates the information based on that data and clarifies the judgment criteria.

[0070] The clarification unit can estimate the user's emotions and determine the priority of the criteria based on the estimated user's emotions. For example, if the user is feeling anxious, the clarification unit prioritizes criteria that provide a sense of security. For example, the clarification unit adjusts the priority of the criteria according to the user's emotions. Furthermore, if the user is excited, the clarification unit can also prioritize quick decisions. For example, the clarification unit determines the priority of the criteria based on the user's emotions. Furthermore, if the user is relaxed, the clarification unit can also prioritize detailed criteria. For example, the clarification unit determines the priority of the criteria based on the user's emotions. This allows for providing more appropriate criteria by providing priorities according to the user's emotions. Emotion estimation is realized using an emotion estimation function, for example, using an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the clarification unit can be performed using, for example, a generation AI (generation AI or LLM) or without a generation AI. For example, the clarification unit inputs the user's emotional data into the generation AI, which then determines the priority of the judgment criteria based on that data.

[0071] The clarification unit can customize the method of explaining the judgment criteria according to the type of data used by the AI. The clarification unit customizes the method of explaining the judgment criteria according to, for example, the type of data used by the AI ​​(image, text, audio, etc.). For example, when image data is used, the clarification unit provides a visual explanation. Furthermore, when text data is used, the clarification unit can also provide a written explanation. Furthermore, when audio data is used, the clarification unit can also provide an audio explanation. For example, the clarification unit explains the judgment criteria using audio data. This makes it possible to improve understandability by providing an explanation method according to the type of data. Some or all of the above-mentioned processing in the clarification unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the clarification unit inputs data used by the AI ​​into the generation AI, and the generation AI customizes the explanation method based on the data.

[0072] When the AI ​​uses data specialized for a particular industry or field, the clarification unit can clarify the industry-specific judgment criteria. For example, when using data from the medical field, the clarification unit clarifies the judgment criteria using medical terminology. For example, the clarification unit analyzes data from the medical field and explains the judgment criteria using medical terminology. Furthermore, when using data from the financial field, the clarification unit can also clarify the judgment criteria using financial terminology. For example, the clarification unit analyzes data from the financial field and explains the judgment criteria using financial terminology. Furthermore, when using data from the manufacturing industry, the clarification unit can also clarify judgment criteria specific to the manufacturing industry. For example, the clarification unit analyzes data from the manufacturing industry and explains judgment criteria specific to the manufacturing industry. This enables more specialized judgments by providing industry-specific judgment criteria. Some or all of the above-mentioned processing in the clarification unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the clarification unit inputs industry-specific data used by the AI ​​into the generation AI, and the generation AI clarifies the judgment criteria based on that data.

[0073] The clarification unit can dynamically update the judgment criteria based on data acquired by the AI ​​in real time. The clarification unit dynamically updates the judgment criteria based on, for example, real-time market data. For example, the clarification unit analyzes real-time market data and updates the judgment criteria based on the latest information. The clarification unit can also dynamically update the judgment criteria based on real-time sensor data. For example, the clarification unit analyzes real-time sensor data and updates the judgment criteria based on the latest information. The clarification unit can also dynamically update the judgment criteria based on real-time user data. For example, the clarification unit analyzes real-time user data and updates the judgment criteria based on the latest information. In this way, by updating the judgment criteria based on the real-time data, it is possible to provide a judgment based on the latest information. Some or all of the above-mentioned processing in the clarification unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the clarification unit inputs data acquired by the AI ​​in real time into the generation AI, and the generation AI dynamically updates the judgment criteria based on the data.

[0074] The visualization unit can estimate the user's emotions and adjust the visualization display method based on the estimated user emotions. For example, when the user is nervous, the visualization unit provides a simple, highly visible display method. For example, the visualization unit changes the color or layout. Furthermore, when the user is relaxed, the visualization unit can provide a display method that includes detailed information. For example, the visualization unit displays detailed data and analysis results. Furthermore, when the user is in a hurry, the visualization unit can provide a display method that focuses on the main points. For example, the visualization unit highlights only important points to enable quick understanding. This improves understandability by providing a display method that corresponds to the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the visualization unit can be performed using, for example, a generation AI (generation AI or LLM) or without a generation AI. For example, the visualization unit inputs the user's emotional data into the generation AI, which then adjusts the display method based on that data.

[0075] The visualization unit can display in detail the process flow when the AI ​​makes a decision and clearly indicate the basis for each step. For example, the visualization unit can display the steps in which the AI ​​collects data and clearly indicate the basis for each step. For example, the visualization unit can show the sensors and data sources used by the AI ​​to collect data. The visualization unit can also display the steps in which the AI ​​analyzes data and clearly indicate the basis for each step. For example, the visualization unit can show the algorithms and methods used by the AI ​​to analyze data. The visualization unit can also display the steps in which the AI ​​makes a final decision and clearly indicate the basis for each step. For example, the visualization unit can show the factors and criteria considered when the AI ​​makes a final decision. This makes it easier for users to understand the AI's decision-making process by displaying it in detail. Some or all of the above-mentioned processing in the visualization unit can be performed using, for example, a generative AI (generative AI or LLM), or can be performed without using a generative AI. For example, the visualization unit inputs data used by the AI ​​into the generative AI, and the generative AI displays a detailed process flow based on that data.

[0076] The visualization unit can change the visualization format depending on the type of algorithm used by the AI. For example, when a deep learning algorithm is used, the visualization unit visualizes the structure of the neural network. For example, the visualization unit shows each layer and node of the neural network. Furthermore, when a decision tree algorithm is used, the visualization unit can also visualize the branches of the decision tree. For example, the visualization unit shows each node and branching condition of the decision tree. Furthermore, when a clustering algorithm is used, the visualization unit can also visualize the distribution of clusters. For example, the visualization unit shows the centers of clusters and the distribution of data points. This makes it possible to improve understandability by providing a visualization format appropriate for the type of algorithm. Some or all of the above-mentioned processing in the visualization unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the visualization unit inputs data of the algorithm used by the AI ​​into the generation AI, and the generation AI changes the visualization format based on the data.

[0077] The visualization unit can compare past and current judgment results made by the AI ​​and visualize trends in change. For example, the visualization unit compares past and current judgment results in a graph to show trends in change. For example, the visualization unit compares past and current judgment results using time-series data. The visualization unit can also compare past and current judgment results in a heat map to show trends in change. For example, the visualization unit visually shows changes in data using a heat map. Furthermore, the visualization unit can compare past and current judgment results in a time series to show trends in change. For example, the visualization unit shows changes in data using trend lines. This makes it easier to understand trends in change by comparing past and current judgment results. Some or all of the above-described processing in the visualization unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the visualization unit inputs past and current judgment results made by the AI ​​into the generation AI, and the generation AI visualizes trends in change based on the data.

[0078] The visualization unit can estimate the user's emotions and determine a visualization priority based on the estimated user's emotions. For example, if the user is feeling anxious, the visualization unit prioritizes displaying information that provides a sense of security. For example, the visualization unit adjusts the display content according to the user's emotions. Furthermore, if the user is excited, the visualization unit can prioritize displaying quick information. For example, the visualization unit adjusts the display content based on the user's emotions. Furthermore, if the user is relaxed, the visualization unit can prioritize displaying detailed information. For example, the visualization unit adjusts the display content based on the user's emotions. This allows more appropriate information to be displayed by providing priorities according to the user's emotions. Emotion estimation is realized using an emotion estimation function, for example, using an emotion engine or a generation AI. The generation AI can be, for example, a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above-described processing in the visualization unit may be performed using, for example, a generation AI (generation AI or LLM) or without a generation AI. For example, the visualization unit inputs the user's emotional data into the generation AI, which then determines the priority of the display content based on that data.

[0079] The visualization unit can customize the visualization format depending on the type of data used by the AI. For example, when image data is used, the visualization unit visualizes the data in a heat map format. For example, the visualization unit displays important parts of the image data using a heat map. Furthermore, when text data is used, the visualization unit can also visualize the data in a word cloud format. For example, the visualization unit displays important words in the text data using a word cloud. Furthermore, when audio data is used, the visualization unit can also visualize the data in a waveform format. For example, the visualization unit displays the characteristics of the audio data using waveforms. This makes it possible to improve understandability by providing a visualization format appropriate for the type of data. Some or all of the above-described processing in the visualization unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the visualization unit inputs data to be used by the AI ​​into the generation AI, and the generation AI customizes the visualization format based on the data.

[0080] When the AI ​​uses data specialized for a particular industry or field, the visualization unit can provide a visualization format specific to that industry. For example, when using data in the medical field, the visualization unit visualizes the data in a medical graph or chart format. For example, the visualization unit displays important parts of the medical data in a medical graph. Furthermore, when using data in the financial field, the visualization unit can also visualize the data in a financial chart or graph format. For example, the visualization unit displays important parts of the financial data in a financial chart. Furthermore, when using data in the manufacturing industry, the visualization unit can also visualize the data in a flowchart format of the manufacturing process. For example, the visualization unit displays important parts of the manufacturing data in a flowchart. In this way, by providing an industry-specific visualization format, more specialized information can be displayed. Some or all of the above-mentioned processing in the visualization unit may be performed using, for example, a generative AI (generative AI or LLM), or may be performed without using a generative AI. For example, the visualization unit inputs industry-specific data used by the AI ​​into the generative AI, and the generative AI provides a visualization format based on that data.

[0081] The visualization unit can dynamically update the visualization content based on data acquired by the AI ​​in real time. The visualization unit dynamically updates the visualization content based on, for example, real-time market data. For example, the visualization unit analyzes real-time market data and updates the visualization content based on the latest information. The visualization unit can also dynamically update the visualization content based on real-time sensor data. For example, the visualization unit analyzes real-time sensor data and updates the visualization content based on the latest information. The visualization unit can also dynamically update the visualization content based on real-time user data. For example, the visualization unit analyzes real-time user data and updates the visualization content based on the latest information. In this way, by updating the visualization content based on the real-time data, it is possible to provide the latest information. Some or all of the above-described processing in the visualization unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the visualization unit inputs data acquired by the AI ​​in real time into the generation AI, and the generation AI dynamically updates the visualization content based on the data.

[0082] The analysis unit can estimate the user's emotions and adjust the method for identifying erroneous results based on the estimated user emotions. For example, if the user is feeling anxious, the analysis unit can quickly identify erroneous results and provide a sense of security. For example, the analysis unit adjusts the identification method based on the user's emotions. Furthermore, if the user is excited, the analysis unit can perform a detailed analysis to clarify the cause of the erroneous results. For example, the analysis unit adjusts the identification method based on the user's emotions. Furthermore, if the user is relaxed, the analysis unit can perform a detailed analysis to clarify the cause of the erroneous results. For example, the analysis unit adjusts the identification method based on the user's emotions. In this way, by providing an identification method according to the user's emotions, erroneous results can be identified quickly and accurately. Emotion estimation is realized using an emotion estimation function, for example, using an emotion engine or a generative AI. The generative AI can be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above-mentioned processing in the analysis unit can be performed using, for example, a generative AI (e.g., LLM) or without a generative AI. For example, the analysis unit inputs the user's emotional data into the generation AI, which then adjusts its identification method based on that data.

[0083] When the AI ​​makes an incorrect decision, the analysis unit can analyze the cause in detail and propose measures to prevent recurrence. For example, the analysis unit identifies the cause of the incorrect decision and proposes measures to prevent recurrence. For example, the analysis unit makes proposals to improve data quality. The analysis unit can also propose improvements to the algorithm. For example, the analysis unit fixes bugs in the algorithm and proposes measures to prevent recurrence. Furthermore, the analysis unit can evaluate data quality and eliminate low-quality data. For example, the analysis unit evaluates the consistency and completeness of the data and excludes low-quality data. This allows for a detailed analysis of the cause of the incorrect decision and proposes measures to prevent recurrence, thereby improving the accuracy of the AI. Some or all of the above-mentioned processing in the analysis unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the analysis unit inputs data used by the AI ​​into the generation AI, which then analyzes the cause of the error based on that data and proposes measures to prevent recurrence.

[0084] The analysis unit can evaluate the quality of the data used by the AI ​​and analyze the impact of a decline in data quality on erroneous decisions. The analysis unit, for example, analyzes the impact of a decline in data quality on erroneous decisions. For example, the analysis unit evaluates the consistency and completeness of the data and checks whether low-quality data has caused errors. The analysis unit can also evaluate data quality and eliminate low-quality data. For example, the analysis unit evaluates the origin and consistency of the data and eliminates low-quality data. The analysis unit can also evaluate data quality and prioritize the use of high-quality data. For example, the analysis unit evaluates the reliability of the data and uses highly reliable data. This makes it possible to analyze the impact of a decline in data quality on erroneous decisions and improve data quality. Some or all of the above-mentioned processing in the analysis unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the analysis unit inputs data to be used by the AI ​​into the generation AI, and the generation AI evaluates the quality based on the data and analyzes the impact of errors.

[0085] The analysis unit can compare information obtained by the AI ​​from different data sources and identify the cause of an incorrect decision. For example, the analysis unit compares information obtained from different data sources and identifies the cause of an incorrect decision. For example, the analysis unit compares internal data with external data and checks for inconsistencies. The analysis unit can also analyze inconsistencies between data sources and identify the cause of an incorrect decision. For example, the analysis unit integrates information obtained from different data sources and identifies the cause of an incorrect decision. Furthermore, the analysis unit can resolve inconsistencies between data sources and identify the cause of an incorrect decision. For example, the analysis unit compares information obtained from different data sources and resolves inconsistencies. In this way, the cause of an incorrect decision can be identified by comparing information from different data sources. Some or all of the above-mentioned processing in the analysis unit may be performed using, for example, the generation AI (generation AI or LLM), or may be performed without using the generation AI. For example, the analysis unit inputs information from different data sources used by the AI ​​into the generation AI, and the generation AI identifies the cause of an error based on that data.

[0086] The analysis unit can estimate the user's emotions and prioritize erroneous results based on the estimated user emotions. For example, if the user is feeling anxious, the analysis unit prioritizes identifying erroneous results. For example, the analysis unit adjusts the priority according to the user's emotions. Furthermore, if the user is excited, the analysis unit can prioritize detailed analysis. For example, the analysis unit adjusts the priority based on the user's emotions. Furthermore, if the user is relaxed, the analysis unit can prioritize detailed analysis. For example, the analysis unit adjusts the priority based on the user's emotions. This allows for quick identification of erroneous results by providing priorities according to the user's emotions. The emotion estimation is realized using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, for example, a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above-described processing in the analysis unit can be performed using, for example, a generation AI (generation AI or LLM) or without a generation AI. For example, the analysis unit inputs the user's emotional data into the generation AI, which then determines priorities based on that data.

[0087] The analysis unit can customize the method for identifying erroneous results depending on the type of data used by the AI. For example, when image data is used, the analysis unit identifies erroneous results using image analysis. For example, the analysis unit analyzes the characteristics of the image data to identify errors. Furthermore, when text data is used, the analysis unit can also identify erroneous results using text analysis. For example, the analysis unit analyzes the content of the text data to identify errors. Furthermore, when audio data is used, the analysis unit can also identify erroneous results using audio analysis. For example, the analysis unit analyzes the characteristics of the audio data to identify errors. This allows for accurate identification of erroneous results by providing an identification method according to the type of data. Some or all of the above-mentioned processing in the analysis unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the analysis unit inputs data to be used by the AI ​​into the generation AI, and the generation AI customizes the identification method based on that data.

[0088] When the AI ​​uses data specialized for a particular industry or field, the analysis unit can analyze error patterns specific to that industry. For example, when data from the medical field is used, the analysis unit analyzes error patterns specific to the medical field. For example, the analysis unit analyzes and identifies error patterns in the medical data. Furthermore, when data from the financial field is used, the analysis unit can also analyze error patterns specific to the financial field. For example, the analysis unit analyzes and identifies error patterns in the financial data. Furthermore, when data from the manufacturing industry is used, the analysis unit can also analyze error patterns specific to the manufacturing industry. For example, the analysis unit analyzes and identifies error patterns in the manufacturing data. In this way, by analyzing industry-specific error patterns, erroneous results can be quickly identified. Some or all of the above-mentioned processing in the analysis unit may be performed using, for example, the generation AI (generation AI or LLM), or may be performed without using the generation AI. For example, the analysis unit inputs industry-specific data used by the AI ​​into the generation AI, and the generation AI analyzes error patterns based on that data.

[0089] The analysis unit can dynamically identify and analyze erroneous results based on data acquired by the AI ​​in real time. The analysis unit dynamically identifies and analyzes erroneous results based on, for example, real-time market data. For example, the analysis unit analyzes real-time market data to identify errors. The analysis unit can also dynamically identify and analyze erroneous results based on real-time sensor data. For example, the analysis unit analyzes real-time sensor data to identify errors. The analysis unit can also dynamically identify and analyze erroneous results based on real-time user data. For example, the analysis unit analyzes real-time user data to identify errors. This enables rapid response by identifying and analyzing erroneous results based on real-time data. Some or all of the above-described processing in the analysis unit may be performed using, for example, a generation AI (generation AI or LLM), or may be performed without using a generation AI. For example, the analysis unit inputs data acquired by the AI ​​in real time into the generation AI, and the generation AI identifies and analyzes errors based on the data. === Hard Collateral 1-1 === Each of the multiple elements including the above-mentioned clarification unit, visualization unit, and analysis unit is realized, for example, by at least one of the smart device 14 and the data processing device 12. For example, the clarification unit is realized by the control unit 46A of the smart device 14 or the specific processing unit 290 of the data processing device 12. For example, the visualization unit is realized by the display 40A of the smart device 14 or the specific processing unit 290 of the data processing device 12. For example, the analysis unit is realized by the control unit 46A of the smart device 14 or the specific processing unit 290 of the data processing device 12. === Hard Collateral 1-2 === Each of the multiple elements including the above-mentioned clarification unit, visualization unit, and analysis unit is realized, for example, by at least one of the smart glasses 214 and the data processing device 12. For example, the clarification unit is realized by the control unit 46A of the smart glasses 214 or the specific processing unit 290 of the data processing device 12. For example, the visualization unit is realized by the display of the smart glasses 214 or the specific processing unit 290 of the data processing device 12. For example, the analysis unit is realized by the control unit 46A of the smart glasses 214 or the specific processing unit 290 of the data processing device 12. === Hard Collateral 1-3 === Each of the multiple elements including the above-mentioned clarification unit, visualization unit, and analysis unit is realized, for example, by at least one of the headset type terminal 314 and the data processing device 12. For example, the clarification unit is realized by the control unit 46A of the headset type terminal 314 or the specific processing unit 290 of the data processing device 12. For example, the visualization unit is realized by the display 343 of the headset type terminal 314 or the specific processing unit 290 of the data processing device 12. For example, the analysis unit is realized by the control unit 46A of the headset type terminal 314 or the specific processing unit 290 of the data processing device 12. === Hard Collateral 1-4 === Each of the multiple elements including the above-mentioned clarification unit, visualization unit, and analysis unit is realized, for example, by at least one of the robot 414 and the data processing device 12. For example, the clarification unit is realized by the control unit 46A of the robot 414 or the specific processing unit 290 of the data processing device 12. For example, the visualization unit is realized by the display of the robot 414 or the specific processing unit 290 of the data processing device 12. For example, the analysis unit is realized by the control unit 46A of the robot 414 or the specific processing unit 290 of the data processing device 12.

[0090] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0091] The clarification unit can evaluate the reliability of data used by the AI ​​and clarify the judgment criteria based on highly reliable data. For example, the clarification unit evaluates the origin, consistency, and completeness of the data and uses only highly reliable data. The clarification unit can also adjust the weighting of the judgment criteria based on the reliability of the data. Furthermore, the clarification unit can eliminate low-reliability data and reconstruct the judgment criteria based on highly reliable data. In this way, by clarifying the judgment criteria based on highly reliable data, the reliability of the AI's judgments can be improved.

[0092] The visualization unit can change the visualization format depending on the type of algorithm used by the AI. For example, if a deep learning algorithm is used, the structure of the neural network can be visualized. Also, if a decision tree algorithm is used, the branching of the decision tree can be visualized. Furthermore, if a clustering algorithm is used, the distribution of clusters can be visualized. This makes it easier to understand by providing a visualization format that suits the type of algorithm.

[0093] The analysis unit can compare information obtained by the AI ​​from different data sources and identify the cause of incorrect decisions. For example, it can compare internal data with external data to check for inconsistencies. It can also analyze inconsistencies between data sources and identify the cause of incorrect decisions. It can also resolve inconsistencies between data sources and identify the cause of incorrect decisions. This makes it possible to identify the cause of incorrect decisions by comparing information from different data sources.

[0094] The clarification unit can refer to the history of past decisions made by the AI ​​and check the consistency of the decision criteria. For example, it can analyze past decision results in chronological order and evaluate the consistency of the decision criteria. It can also compare the past decision history with the current decision criteria and make adjustments to maintain consistency. Furthermore, if an inconsistent decision is found, it can identify the cause and revise the decision criteria. In this way, by referring to the past decision history, it is possible to maintain the consistency of the decision criteria.

[0095] The visualization unit can compare the AI's past and current judgment results and visualize trends in change. For example, past and current judgment results can be compared in a graph to show trends in change. Past and current judgment results can also be compared in a heat map to show trends in change. Furthermore, past and current judgment results can be compared in chronological order to show trends in change. This makes it easier to understand trends in change by comparing past and current judgment results.

[0096] The clarification unit can estimate the user's emotions and adjust the explanation method for the judgment criteria based on the estimated user emotions. For example, if the user is feeling stressed, a simple and intuitive explanation can be provided to help the user understand. If the user is relaxed, a detailed explanation can be provided to encourage deeper understanding. Furthermore, if the user is in a hurry, a concise explanation that focuses on the main points can be provided to help the user understand quickly. This makes it possible to improve ease of understanding by providing an explanation method that suits the user's emotions.

[0097] The visualization unit can estimate the user's emotions and adjust the visualization display method based on the estimated user emotions. For example, if the user is nervous, a simple, highly visible display method can be provided. If the user is relaxed, a display method including detailed information can be provided. Furthermore, if the user is in a hurry, a display method that focuses on the main points can be provided. In this way, by providing a display method that corresponds to the user's emotions, it is possible to improve ease of understanding.

[0098] The analysis unit can estimate the user's emotions and adjust the method for identifying erroneous results based on the estimated user emotions. For example, if the user is feeling anxious, the analysis unit can quickly identify erroneous results and provide a sense of security. If the user is excited, the analysis unit can perform a detailed analysis to clarify the cause of the erroneous results. If the user is relaxed, the analysis unit can perform a detailed analysis to clarify the cause of the erroneous results. This allows for quick and accurate identification of erroneous results by providing an identification method that corresponds to the user's emotions.

[0099] The clarification unit can estimate the user's emotions and determine the priority of the judgment criteria based on the estimated user's emotions. For example, if the user feels anxious, the clarification unit can prioritize judgment criteria that give a sense of security. Also, if the user is excited, the clarification unit can prioritize quick judgment. Furthermore, if the user is relaxed, the clarification unit can prioritize detailed judgment criteria. In this way, by providing priorities according to the user's emotions, more appropriate judgment criteria can be provided.

[0100] The visualization unit can estimate the user's emotions and determine the priority of visualization based on the estimated user's emotions. For example, if the user is feeling anxious, information that gives a sense of security can be displayed with priority. Also, if the user is excited, quick information can be displayed with priority. Furthermore, if the user is relaxed, detailed information can be displayed with priority. In this way, by providing priorities according to the user's emotions, more appropriate information can be displayed.

[0101] The processing flow of the second embodiment will be briefly explained below.

[0102] Step 1: The clarification unit clarifies the AI's decision-making criteria. For example, the clarification unit analyzes what data the AI ​​uses to make its decisions and explains the decision-making criteria in terms that are easy for humans to understand. The clarification unit can also evaluate the reliability of the data used by the AI ​​and clarify the decision-making criteria based on highly reliable data. Furthermore, the clarification unit can refer to the history of past decisions made by the AI ​​to confirm the consistency of the decision-making criteria. Step 2: The visualization unit visualizes the basis for the black-box AI's decision based on the criteria clarified by the clarification unit. For example, the visualization unit visually displays the process the AI ​​went through to make its decision. The visualization unit can also use graphs and charts to show which words and phrases the AI ​​prioritizes when analyzing text. Furthermore, the visualization unit can display in detail the process flow when the AI ​​makes a decision and clearly state the basis for the decision at each step. Step 3: The analysis unit identifies and analyzes erroneous results based on the decision-making basis visualized by the visualization unit. For example, if the AI ​​makes an incorrect decision, the analysis unit identifies the cause and analyzes which part of the data caused the error. The analysis unit can also analyze the cause in detail when the AI ​​makes an incorrect decision and propose measures to prevent recurrence. Furthermore, the analysis unit can evaluate the quality of the data used by the AI ​​and analyze the impact that a decline in data quality has on erroneous decisions.

[0103] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0104] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AIs include the data generation model 58, such as a neural network model (e.g., a neural network model), and a neural network model (e.g., a neural network model). The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating speech, text data indicating text, and image data indicating an image is also input to the data generation model 58. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0105] Furthermore, the processing by the data processing system 10 described above is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0106] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0107] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0108] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0109] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0110] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0111] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0112] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0113] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0114] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0115] The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0116] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0117] In the smart glasses 214, the specific processing is performed by the processor 46. A specific processing program 60 is stored in the storage 50. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0118] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0119] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0120] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AI other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0121] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart glasses 214 or an external device, etc., and the smart glasses 214 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0122] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0123] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0124] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0125] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0126] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0127] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0128] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0129] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0130] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0131] The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0132] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0133] In the headset type terminal 314, the identification process is performed by the processor 46. A identification program 60 is stored in the storage 50. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as a control unit 46A in accordance with the identification program 60 executed on the RAM 48. Note that the headset type terminal 314 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the identification processing unit 290 using these models.

[0134] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0135] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0136] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AI other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0137] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset type terminal 314, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset type terminal 314. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the headset type terminal 314 or an external device, etc., and the headset type terminal 314 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0138] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0139] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0140] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0141] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0142] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[0143] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0144] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS image sensor or a CCD image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0145] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0146] The control object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[0147] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0148] The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0149] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0150] In the robot 414, the processor 46 performs the identification process. The storage 50 stores the identification program 60. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as the control unit 46A in accordance with the identification program 60 executed on the RAM 48. The robot 414 also has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform the same process as the identification processing unit 290 using these models.

[0151] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0152] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0153] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AI other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0154] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the robot 414 or an external device, etc., and the robot 414 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0155] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0156] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0157] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion encompasses both emotions and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[0158] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[0159] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[0160] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is expressed, and when they approach the ideal, a state of pleasure is expressed. Emotions can also be created for robots, cars, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is expressed, and when they approach the ideal, a state of pleasure is expressed. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on speech emotion recognition and brain physiological signal analysis systems for emotions, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[0161] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[0162] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[0163] In the above embodiment, an example was given in which a specific process is performed by one computer 22, but the technology disclosed herein is not limited to this, and distributed processing of the specific process may be performed by multiple computers including computer 22.

[0164] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[0165] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[0166] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[0167] The hardware resource for executing a specific process can be any of the following types of processors: A processor, for example, is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. A processor also includes a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[0168] The hardware resource that executes the specific process may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific process may be a single processor.

[0169] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[0170] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[0171] In the above example, the first to fourth embodiments have been described separately, but some or all of these embodiments may be combined. The smart device 14, smart glasses 214, headset terminal 314, and robot 414 are merely examples, and they may be combined, or other devices may be used. In the above example, the first and second embodiments have been described separately, but they may be combined.

[0172] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[0173] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[0174] [Explanation of symbols]

[0175] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot

Claims

1. A clarification unit that clarifies the AI's judgment criteria; A visualization unit that visualizes the basis for AI judgment based on the judgment criteria clarified by the clarification unit; and an analysis unit that identifies and analyzes erroneous results based on the judgment grounds visualized by the visualization unit. A system characterized by:

2. The clarification unit Analyze what data AI uses to make decisions and explain the criteria in terms humans can understand.

2. The system of claim 1.

3. The visualization unit Visually display the process through which the AI ​​made its decision 2. The system of claim 1.

4. The visualization unit Graphs and charts show which words and phrases are prioritized when AI analyzes text.

2. The system of claim 1.

5. The analysis unit If the AI ​​makes an incorrect decision, identify the cause and analyze which part of the data caused the error.

2. The system of claim 1.

6. The clarification unit Inferring user emotions and adjusting the explanation method for the criteria based on the estimated user emotions 2. The system of claim 1.

7. The clarification unit Evaluate the reliability of data used by AI and clarify decision-making criteria based on reliable data.

2. The system of claim 1.

8. The clarification unit Check the consistency of the judgment criteria by looking at the history of past decisions made by the AI.

2. The system of claim 1.

Images