System

Abstract

An object of a system according to an embodiment is to appropriately evaluate a AI model.SOLUTION: A system includes a model classification unit, a data set generation unit, and an evaluation unit. The model classifying unit classifies a AI model from the source code. The dataset generation unit generates a dataset suitable for the AI model classified by the model classification unit. The evaluation unit evaluates the AI model with a plurality of indexes using the dataset generated by the dataset generation 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] Conventional technology has had the problem of insufficient evaluation of AI models, particularly the difficulty of evaluating them using multiple indicators.

[0005] The system according to the embodiment aims to appropriately evaluate an AI model. [Means for solving the problem]

[0006] A system according to an embodiment includes a model classification unit, a dataset generation unit, and an evaluation unit. The model classification unit classifies an AI model from source code. The dataset generation unit generates a dataset suitable for the AI ​​model classified by the model classification unit. The evaluation unit evaluates the AI ​​model using a plurality of indicators using the dataset generated by the dataset generation unit. [Effects of the Invention]

[0007] The system according to the embodiment can appropriately evaluate an AI model. [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) The AI ​​model evaluation system according to an embodiment of the present invention is a system that classifies AI models from source code, generates datasets suitable for each model, and evaluates the AI ​​models using multiple indicators. As a result, the AI ​​model evaluation system can ensure transparency and accountability for AI models and provide highly reliable evaluations.

[0029] An AI model evaluation system according to an embodiment includes a model classification unit, a dataset generation unit, and an evaluation unit. The model classification unit classifies AI models from source code. For example, the generation AI classifies the purpose and function of each model based on the structure and description of the source code. The generation AI can also distinguish between image recognition models, natural language processing models, and recommendation system models. The generation AI analyzes the entire provided source code to identify different AI models. The dataset generation unit generates datasets appropriate for each of the classified AI models. For example, the generation AI generates an image dataset for an image recognition model and a text dataset for a natural language processing model. The generation AI also generates appropriate datasets based on information regarding the characteristics and purpose of each model. The evaluation unit uses the generated datasets to evaluate AI models using multiple metrics. For example, the generation AI evaluates using metrics such as precision, recall, F1 score, and ROC curve. The generation AI also performs evaluation based on a list of metrics used for evaluation and the AI ​​model to be evaluated. This allows the AI ​​model evaluation system according to an embodiment to ensure transparency and accountability of AI models and provide highly reliable evaluations. For example, developers and users of AI models can use the evaluation results to identify areas for improvement and develop higher quality AI models. Furthermore, by disclosing the evaluation results as a third-party organization, trust in AI can be improved throughout society.

[0030] The model classification unit analyzes source code comments and documentation to classify AI models based on the developer's intent and design philosophy. For example, the generation AI analyzes source code comments to extract the developer's intent and design philosophy. For example, it classifies models based on algorithm explanations and usage examples written in the comments. The model classification unit also analyzes documentation accompanying the source code to understand the developer's design philosophy and objectives. For example, it identifies the role and function of each model based on information written in design documents and specifications. The generation AI also analyzes source code comments and documentation in an integrated manner to classify models taking the developer's intent into consideration. For example, it compares the content of comments and documentation to perform consistent model classification. This enables model classification that takes the developer's intent and design philosophy into consideration.

[0031] The model classification unit analyzes the version history of the source code and can track and classify the evolution and changes of the model. In the model classification unit, for example, the generation AI analyzes the source code version control system and tracks the change history of each version. For example, it analyzes Git or SVN logs to identify the evolution and changes of the model. The model classification unit also extracts changes and additional functions for each model based on the version history. For example, it analyzes new functions and corrections added in a specific version and reflects them in the classification of the model. In the model classification unit, the generation AI also analyzes the version history and visualizes the evolution process of the model. For example, it displays the changes in each version in timeline format to track the evolution of the model. This makes it possible to track and classify the evolution and changes of the model.

[0032] The dataset generation unit can automatically detect and correct bias and imbalance in the data during the dataset generation process. For example, the dataset generation unit automatically detects and corrects bias in the data during the dataset generation process using a generation AI. For example, it distributes data that is biased toward a specific attribute evenly. The dataset generation unit also automatically detects and corrects imbalance in the dataset. For example, it resamples a dataset that has class imbalance to balance it. The dataset generation unit also applies an algorithm that allows the generation AI to detect and correct bias and imbalance in the dataset. For example, it analyzes the distribution of data and performs filtering to remove bias. This makes it possible to automatically correct bias and imbalance in the dataset.

[0033] The dataset generation unit can evaluate the quality and reliability of the data for the dataset it generates and add appropriate metadata. For example, the dataset generation unit adds metadata that allows the generation AI to evaluate the quality of the dataset and ensure its reliability. For example, it adds information about the source of the data and the collection method as metadata. The dataset generation unit also evaluates the reliability of the dataset and adds appropriate metadata. For example, it adds information about the accuracy and completeness of the data as metadata. The dataset generation unit also applies an algorithm that allows the generation AI to evaluate the quality and reliability of the dataset and add metadata. For example, it adds information about the consistency and integrity of the data as metadata. This ensures the quality and reliability of the dataset.

[0034] The dataset generation unit can generate a dataset by integrating different data sources. In the dataset generation unit, for example, the generation AI integrates different data sources to generate a dataset. For example, a dataset is created by combining sensor data and social media data. The dataset generation unit also integrates data collected from different data sources to generate a dataset. For example, data from IoT devices is integrated with user social media posts. In addition, the dataset generation unit analyzes different data sources and generates an integrated dataset. For example, a dataset is created by combining environmental sensor data and user behavior data. This allows a dataset to be generated by integrating different data sources.

[0035] The dataset generation unit can share the generated dataset between different AI models and build a mutually usable dataset. For example, the dataset generation unit shares a dataset generated by the generation AI between different AI models and builds a mutually usable dataset. For example, an image recognition model and a natural language processing model use the same dataset. The dataset generation unit also develops a system that shares datasets between different AI models and builds mutually usable datasets. For example, it integrates datasets corresponding to different tasks. The dataset generation unit also shares the dataset generated by the generation AI between different AI models and maintains data consistency. For example, data consistency is ensured by having multiple models use the same dataset. This makes it possible to share datasets between different AI models and build mutually usable datasets.

[0036] During the evaluation index selection process, the evaluation unit can propose customizable indexes according to the model's characteristics and application. For example, the generation AI proposes evaluation indexes according to the model's characteristics and application. For example, it proposes precision and recall for image recognition models, and BLEU score and ROUGE score for natural language processing models. Furthermore, during the evaluation index selection process, the generation AI analyzes the model's characteristics and proposes the optimal index. For example, it proposes NDCG and MAP for recommendation systems. Furthermore, during the evaluation index selection process, the generation AI proposes customizable indexes according to the model's application. For example, it proposes specificity and sensitivity for models in the medical field. This makes it possible to propose customizable indexes according to the model's characteristics and application.

[0037] The evaluation unit can automatically generate specific action plans for model improvement and optimization based on the evaluation results. For example, the generation AI identifies model improvement points based on the evaluation results and automatically generates specific action plans. For example, it proposes hyperparameter adjustments and data augmentation methods to improve accuracy. The evaluation unit also analyzes the evaluation results, and the generation AI proposes specific action plans for optimization. For example, it proposes regularization methods and dropout rate adjustments to prevent model overfitting. The evaluation unit also builds a system in which the generation AI automatically generates model improvement points and action plans for optimization based on the evaluation results. For example, it proposes the selection of optimal training data and improvements to the model architecture based on the evaluation indicators. This makes it possible to automatically generate specific action plans based on the evaluation results.

[0038] The evaluation unit can combine different evaluation indices to perform a composite evaluation and evaluate the overall model performance. For example, the generation AI combines different evaluation indices to perform a composite evaluation. For example, the generation AI combines precision, recall, and F1 score to evaluate the overall model performance. The evaluation unit also builds a system in which the generation AI combines different evaluation indices to evaluate the overall model performance. For example, the generation AI combines ROC curves and AUCs to evaluate the model performance. The evaluation unit also builds a system in which the generation AI combines different evaluation indices to perform a composite evaluation and evaluate the overall model performance. For example, the generation AI combines BLEU scores and ROUGE scores to evaluate the performance of a natural language processing model. This allows for a composite evaluation to evaluate the overall model performance.

[0039] The evaluation unit can compare the evaluation results with other AI models and perform relative performance evaluation. For example, the evaluation unit allows the generation AI to compare the evaluation results with other AI models and perform relative performance evaluation. For example, it compares the evaluation results of multiple models that solve the same task and identifies the best model. The evaluation unit also builds a system in which the generation AI compares the evaluation results with other AI models and performs relative performance evaluation. For example, it compares the performance of models with different architectures. The evaluation unit also allows the generation AI to compare the evaluation results with other AI models and perform relative performance evaluation. For example, it compares the performance of models trained with different datasets and selects the optimal model. This makes it possible to perform relative performance evaluation by comparing with other AI models.

[0040] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0041] The model classification unit analyzes source code comments and documentation to classify AI models based on the developer's intent and design philosophy. For example, the generation AI analyzes source code comments and extracts the developer's intent and design philosophy. For example, it classifies models based on algorithm explanations and usage examples written in the comments. The model classification unit also analyzes documentation accompanying the source code to understand the developer's design philosophy and objectives. For example, it identifies the role and function of each model based on information written in design documents and specifications. The generation AI also comprehensively analyzes source code comments and documentation to classify models taking the developer's intent into consideration. For example, it compares the content of comments and documentation to perform consistent model classification. This enables model classification that takes the developer's intent and design philosophy into consideration.

[0042] The model classification unit analyzes the version history of the source code and can track and classify the evolution and changes of the model. For example, the generation AI analyzes the source code version control system and tracks the change history of each version. For example, it analyzes Git or SVN logs to identify the evolution and changes of the model. The model classification unit also extracts changes and additional functions for each model based on the version history. For example, it analyzes the new functions and corrections added in a specific version and reflects them in the classification of the model. The generation AI also analyzes the version history in the model classification unit and visualizes the evolution process of the model. For example, it displays the changes in each version in timeline format to track the evolution of the model. This makes it possible to track and classify the evolution and changes of the model.

[0043] The dataset generation unit can automatically detect and correct bias and imbalance in the data during the dataset generation process. For example, the generation AI automatically detects and corrects bias in the data during the dataset generation process. For example, it distributes data that is biased toward a specific attribute evenly. The dataset generation unit also automatically detects and corrects imbalance in the dataset. For example, it resamples a dataset with class imbalance to balance it. The dataset generation unit also applies an algorithm to detect and correct bias and imbalance in the dataset. For example, it analyzes the distribution of data and performs filtering to remove bias. This makes it possible to automatically correct bias and imbalance in the dataset.

[0044] The dataset generation unit can evaluate the quality and reliability of the data for the dataset it generates and add appropriate metadata. For example, the generation AI evaluates the quality of the dataset and adds metadata to ensure reliability. For example, information about the source of the data and the collection method is added as metadata. The dataset generation unit also evaluates the reliability of the dataset and adds appropriate metadata. For example, information about the accuracy and completeness of the data is added as metadata. The dataset generation unit also applies an algorithm that allows the generation AI to evaluate the quality and reliability of the dataset and add metadata. For example, information about the consistency and integrity of the data is added as metadata. This ensures the quality and reliability of the dataset.

[0045] The dataset generation unit can generate a dataset by integrating different data sources. For example, the generation AI integrates different data sources to generate a dataset. For example, a dataset is created by combining sensor data and social media data. The dataset generation unit also integrates data collected from different data sources to generate a dataset. For example, data from IoT devices is integrated with user social media posts. The dataset generation unit also analyzes different data sources to generate an integrated dataset. For example, a dataset is created by combining environmental sensor data and user behavior data. This makes it possible to generate a dataset by integrating different data sources.

[0046] The dataset generation unit can share the generated dataset between different AI models to build a mutually usable dataset. For example, a dataset generated by the generation AI is shared between different AI models to build a mutually usable dataset. For example, an image recognition model and a natural language processing model use the same dataset. The dataset generation unit also develops a system that shares datasets between different AI models to build a mutually usable dataset. For example, it integrates datasets corresponding to different tasks. The dataset generation unit also shares the dataset generated by the generation AI between different AI models to maintain data consistency. For example, data consistency is ensured by having multiple models use the same dataset. This makes it possible to share datasets between different AI models to build a mutually usable dataset.

[0047] During the evaluation index selection process, the evaluation unit can propose customizable indexes according to the model's characteristics and application. For example, the generative AI proposes evaluation indexes according to the model's characteristics and application. For example, it proposes precision and recall for image recognition models, and BLEU score and ROUGE score for natural language processing models. Furthermore, during the evaluation index selection process, the generative AI analyzes the model's characteristics and proposes the optimal index. For example, it proposes NDCG and MAP for recommendation systems. Furthermore, during the evaluation index selection process, the generative AI proposes customizable indexes according to the model's application. For example, it proposes specificity and sensitivity for models in the medical field. This makes it possible to propose customizable indexes according to the model's characteristics and application.

[0048] The processing flow of the first embodiment will be briefly explained below.

[0049] Step 1: The model classification unit classifies AI models from the source code. For example, the generative AI classifies the purpose and function of each model based on the structure and content of the source code. The generative AI can also identify image recognition models, natural language processing models, recommendation system models, etc. The generative AI analyzes the entire provided source code and identifies different AI models. Step 2: The dataset generation unit generates a dataset appropriate for each of the separated AI models. For example, the generation AI generates an image dataset for an image recognition model and a text dataset for a natural language processing model. The generation AI also generates an appropriate dataset based on information about the characteristics and purpose of each model. Step 3: The evaluation unit uses the generated dataset to evaluate the AI ​​model using multiple indicators. For example, the generation AI is evaluated using indicators such as precision, recall, F1 score, and ROC curve. The generation AI also performs evaluation based on a list of indicators used for evaluation and the AI ​​model to be evaluated. This allows the AI ​​model evaluation system according to the embodiment to ensure transparency and accountability of AI models and provide highly reliable evaluations.

[0050] (Example 2) The AI ​​model evaluation system according to an embodiment of the present invention is a system that classifies AI models from source code, generates datasets suitable for each model, and evaluates the AI ​​models using multiple indicators. As a result, the AI ​​model evaluation system can ensure transparency and accountability for AI models and provide highly reliable evaluations.

[0051] An AI model evaluation system according to an embodiment includes a model classification unit, a dataset generation unit, and an evaluation unit. The model classification unit classifies AI models from source code. For example, the generation AI classifies the purpose and function of each model based on the structure and description of the source code. The generation AI can also distinguish between image recognition models, natural language processing models, and recommendation system models. The generation AI analyzes the entire provided source code to identify different AI models. The dataset generation unit generates datasets appropriate for each of the classified AI models. For example, the generation AI generates an image dataset for an image recognition model and a text dataset for a natural language processing model. The generation AI also generates appropriate datasets based on information regarding the characteristics and purpose of each model. The evaluation unit uses the generated datasets to evaluate AI models using multiple metrics. For example, the generation AI evaluates using metrics such as precision, recall, F1 score, and ROC curve. The generation AI also performs evaluation based on a list of metrics used for evaluation and the AI ​​model to be evaluated. This allows the AI ​​model evaluation system according to an embodiment to ensure transparency and accountability of AI models and provide highly reliable evaluations. For example, developers and users of AI models can use the evaluation results to identify areas for improvement and develop higher quality AI models. Furthermore, by disclosing the evaluation results as a third-party organization, trust in AI can be improved throughout society.

[0052] The model classification unit analyzes source code comments and documentation to classify AI models based on the developer's intent and design philosophy. For example, the generation AI analyzes source code comments to extract the developer's intent and design philosophy. For example, it classifies models based on algorithm explanations and usage examples written in the comments. The model classification unit also analyzes documentation accompanying the source code to understand the developer's design philosophy and objectives. For example, it identifies the role and function of each model based on information written in design documents and specifications. The generation AI also analyzes source code comments and documentation in an integrated manner to classify models taking the developer's intent into consideration. For example, it compares the content of comments and documentation to perform consistent model classification. This enables model classification that takes the developer's intent and design philosophy into consideration.

[0053] The model classification unit analyzes the version history of the source code and can track and classify the evolution and changes of the model. In the model classification unit, for example, the generation AI analyzes the source code version control system and tracks the change history of each version. For example, it analyzes Git or SVN logs to identify the evolution and changes of the model. The model classification unit also extracts changes and additional functions for each model based on the version history. For example, it analyzes new functions and corrections added in a specific version and reflects them in the classification of the model. In the model classification unit, the generation AI also analyzes the version history and visualizes the evolution process of the model. For example, it displays the changes in each version in timeline format to track the evolution of the model. This makes it possible to track and classify the evolution and changes of the model.

[0054] The model classification unit uses the emotion estimation function to infer the developer's emotions from the source code description and prioritizes analysis of emotionally important parts. For example, the model classification unit uses a generative AI to infer the developer's emotions from the source code description and identify emotionally important parts. For example, it prioritizes analysis of parts with high emotion scores and reflects this in model classification. The model classification unit also uses the emotion estimation function to identify parts where the developer is feeling stressed and focuses analysis on those parts. For example, it analyzes parts with many error messages and debug comments. The model classification unit also infers the developer's emotions from the source code description and prioritizes analysis of parts with strong positive emotions. For example, it analyzes parts that describe successful implementations and areas for improvement. This makes it possible to classify models that take the developer's emotions into account.

[0055] The dataset generation unit can automatically detect and correct bias and imbalance in the data during the dataset generation process. For example, the dataset generation unit automatically detects and corrects bias in the data during the dataset generation process using a generation AI. For example, it distributes data that is biased toward a specific attribute evenly. The dataset generation unit also automatically detects and corrects imbalance in the dataset. For example, it resamples a dataset that has class imbalance to balance it. The dataset generation unit also applies an algorithm that allows the generation AI to detect and correct bias and imbalance in the dataset. For example, it analyzes the distribution of data and performs filtering to remove bias. This makes it possible to automatically correct bias and imbalance in the dataset.

[0056] The dataset generation unit can evaluate the quality and reliability of the data for the dataset it generates and add appropriate metadata. For example, the dataset generation unit adds metadata that allows the generation AI to evaluate the quality of the dataset and ensure its reliability. For example, it adds information about the source of the data and the collection method as metadata. The dataset generation unit also evaluates the reliability of the dataset and adds appropriate metadata. For example, it adds information about the accuracy and completeness of the data as metadata. The dataset generation unit also applies an algorithm that allows the generation AI to evaluate the quality and reliability of the dataset and add metadata. For example, it adds information about the consistency and integrity of the data as metadata. This ensures the quality and reliability of the dataset.

[0057] The dataset generation unit can use the emotion estimation function to consider user emotions during the dataset generation process and prioritize generating data that elicits positive emotions. For example, the dataset generation unit uses a generation AI to consider user emotions during the dataset generation process and prioritize generating data that elicits positive emotions. For example, it prioritizes collecting data that is likely to interest users. The dataset generation unit also uses the emotion estimation function to monitor user emotions in real time during the dataset generation process and generate data that elicits positive emotions. For example, it adjusts the data based on user feedback. The dataset generation unit also considers user emotions during the dataset generation process and provides an interface for eliciting positive emotions. For example, it designs an interface that allows users to enjoy collecting data. This makes it possible to generate datasets that take user emotions into consideration.

[0058] The dataset generation unit can generate a dataset by integrating different data sources. In the dataset generation unit, for example, the generation AI integrates different data sources to generate a dataset. For example, a dataset is created by combining sensor data and social media data. The dataset generation unit also integrates data collected from different data sources to generate a dataset. For example, data from IoT devices is integrated with user social media posts. In addition, the dataset generation unit analyzes different data sources and generates an integrated dataset. For example, a dataset is created by combining environmental sensor data and user behavior data. This allows a dataset to be generated by integrating different data sources.

[0059] The dataset generation unit can share the generated dataset between different AI models and build a mutually usable dataset. For example, the dataset generation unit shares a dataset generated by the generation AI between different AI models and builds a mutually usable dataset. For example, an image recognition model and a natural language processing model use the same dataset. The dataset generation unit also develops a system that shares datasets between different AI models and builds mutually usable datasets. For example, it integrates datasets corresponding to different tasks. The dataset generation unit also shares the dataset generated by the generation AI between different AI models and maintains data consistency. For example, data consistency is ensured by having multiple models use the same dataset. This makes it possible to share datasets between different AI models and build mutually usable datasets.

[0060] The dataset generation unit can use the emotion estimation function to monitor user emotions in real time during the dataset generation process and generate data that is easily emotionally relatable. For example, the dataset generation unit uses a generation AI to monitor user emotions in real time during the dataset generation process and generate data that is easily emotionally relatable. For example, the dataset generation unit adjusts the data based on the user's emotion score. The dataset generation unit also uses the emotion estimation function to take user emotions into consideration during the dataset generation process and prioritize generating data that is easily relatable. For example, it collects data that elicits positive emotions. The dataset generation unit also builds a system that monitors user emotions in real time during the dataset generation process and generates data that is easily emotionally relatable. For example, it adjusts the data based on user feedback. This makes it possible to monitor user emotions in real time and generate data that is easily emotionally relatable.

[0061] During the evaluation index selection process, the evaluation unit can propose customizable indexes according to the model's characteristics and application. For example, the generation AI proposes evaluation indexes according to the model's characteristics and application. For example, it proposes precision and recall for image recognition models, and BLEU score and ROUGE score for natural language processing models. Furthermore, during the evaluation index selection process, the generation AI analyzes the model's characteristics and proposes the optimal index. For example, it proposes NDCG and MAP for recommendation systems. Furthermore, during the evaluation index selection process, the generation AI proposes customizable indexes according to the model's application. For example, it proposes specificity and sensitivity for models in the medical field. This makes it possible to propose customizable indexes according to the model's characteristics and application.

[0062] The evaluation unit can automatically generate specific action plans for model improvement and optimization based on the evaluation results. For example, the generation AI identifies model improvement points based on the evaluation results and automatically generates specific action plans. For example, it proposes hyperparameter adjustments and data augmentation methods to improve accuracy. The evaluation unit also analyzes the evaluation results, and the generation AI proposes specific action plans for optimization. For example, it proposes regularization methods and dropout rate adjustments to prevent model overfitting. The evaluation unit also builds a system in which the generation AI automatically generates model improvement points and action plans for optimization based on the evaluation results. For example, it proposes the selection of optimal training data and improvements to the model architecture based on the evaluation indicators. This makes it possible to automatically generate specific action plans based on the evaluation results.

[0063] The evaluation unit uses the emotion estimation function to analyze the user's emotional response to the evaluation results and can emphasize emotionally positive evaluation results. For example, the evaluation unit uses a generation AI to analyze the user's emotional response to the evaluation results and emphasize positive evaluation results. For example, the evaluation unit prominently displays positive evaluation results based on the user's emotional score. The evaluation unit also uses the emotion estimation function to monitor the user's emotional response to the evaluation results in real time and emphasizes positive evaluation results. For example, the evaluation results are adjusted based on user feedback. The evaluation unit also builds a system in which the generation AI analyzes the user's emotional response to the evaluation results and emphasizes emotionally positive evaluation results. For example, evaluation results with a high positive emotional score are preferentially displayed. This makes it possible to analyze the user's emotional response and emphasize positive evaluation results.

[0064] The evaluation unit can combine different evaluation indices to perform a composite evaluation and evaluate the overall model performance. For example, the generation AI combines different evaluation indices to perform a composite evaluation. For example, the generation AI combines precision, recall, and F1 score to evaluate the overall model performance. The evaluation unit also builds a system in which the generation AI combines different evaluation indices to evaluate the overall model performance. For example, the generation AI combines ROC curves and AUCs to evaluate the model performance. The evaluation unit also builds a system in which the generation AI combines different evaluation indices to perform a composite evaluation and evaluate the overall model performance. For example, the generation AI combines BLEU scores and ROUGE scores to evaluate the performance of a natural language processing model. This allows for a composite evaluation to evaluate the overall model performance.

[0065] The evaluation unit can compare the evaluation results with other AI models and perform relative performance evaluation. For example, the evaluation unit allows the generation AI to compare the evaluation results with other AI models and perform relative performance evaluation. For example, it compares the evaluation results of multiple models that solve the same task and identifies the best model. The evaluation unit also builds a system in which the generation AI compares the evaluation results with other AI models and performs relative performance evaluation. For example, it compares the performance of models with different architectures. The evaluation unit also allows the generation AI to compare the evaluation results with other AI models and perform relative performance evaluation. For example, it compares the performance of models trained with different datasets and selects the optimal model. This makes it possible to perform relative performance evaluation by comparing with other AI models.

[0066] The evaluation unit uses the emotion estimation function to monitor the user's emotional response to the evaluation results in real time, and can continuously improve the evaluation process. For example, the evaluation unit uses the generation AI to monitor the user's emotional response to the evaluation results in real time, and can continuously improve the evaluation process. For example, the evaluation unit adjusts the evaluation index based on the user's emotional score. The evaluation unit also uses the emotion estimation function to build a system that monitors the user's emotional response to the evaluation results in real time, and improves the evaluation process. For example, the evaluation unit optimizes the evaluation process based on user feedback. The evaluation unit also uses the generation AI to monitor the user's emotional response to the evaluation results in real time, and can continuously improve the evaluation process. For example, the evaluation index with a high number of positive emotional responses is prioritized. This allows the user's emotional response to be monitored in real time, and can continuously improve the evaluation process.

[0067] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0068] The model classification unit analyzes source code comments and documentation to classify AI models based on the developer's intent and design philosophy. For example, the generation AI analyzes source code comments and extracts the developer's intent and design philosophy. For example, it classifies models based on algorithm explanations and usage examples written in the comments. The model classification unit also analyzes documentation accompanying the source code to understand the developer's design philosophy and objectives. For example, it identifies the role and function of each model based on information written in design documents and specifications. The generation AI also comprehensively analyzes source code comments and documentation to classify models taking the developer's intent into consideration. For example, it compares the content of comments and documentation to perform consistent model classification. This enables model classification that takes the developer's intent and design philosophy into consideration.

[0069] The model classification unit analyzes the version history of the source code and can track and classify the evolution and changes of the model. For example, the generation AI analyzes the source code version control system and tracks the change history of each version. For example, it analyzes Git or SVN logs to identify the evolution and changes of the model. The model classification unit also extracts changes and additional functions for each model based on the version history. For example, it analyzes the new functions and corrections added in a specific version and reflects them in the classification of the model. The generation AI also analyzes the version history in the model classification unit and visualizes the evolution process of the model. For example, it displays the changes in each version in timeline format to track the evolution of the model. This makes it possible to track and classify the evolution and changes of the model.

[0070] The model classification unit uses the emotion estimation function to infer the developer's emotions from the source code description and prioritizes analysis of emotionally important parts. For example, the generative AI infers the developer's emotions from the source code description and identifies the emotionally important parts. For example, it prioritizes analysis of parts with high emotion scores and reflects this in the model classification. The model classification unit also uses the emotion estimation function to identify parts that cause stress to the developer and focuses analysis on those parts. For example, it analyzes parts with many error messages and debug comments. The model classification unit also infers the developer's emotions from the source code description and prioritizes analysis of parts with strong positive emotions. For example, it analyzes parts that describe successful implementations and areas for improvement. This makes it possible to classify models that take the developer's emotions into account.

[0071] The dataset generation unit can automatically detect and correct bias and imbalance in the data during the dataset generation process. For example, the generation AI automatically detects and corrects bias in the data during the dataset generation process. For example, it distributes data that is biased toward a specific attribute evenly. The dataset generation unit also automatically detects and corrects imbalance in the dataset. For example, it resamples a dataset with class imbalance to balance it. The dataset generation unit also applies an algorithm to detect and correct bias and imbalance in the dataset. For example, it analyzes the distribution of data and performs filtering to remove bias. This makes it possible to automatically correct bias and imbalance in the dataset.

[0072] The dataset generation unit can evaluate the quality and reliability of the data for the dataset it generates and add appropriate metadata. For example, the generation AI evaluates the quality of the dataset and adds metadata to ensure reliability. For example, information about the source of the data and the collection method is added as metadata. The dataset generation unit also evaluates the reliability of the dataset and adds appropriate metadata. For example, information about the accuracy and completeness of the data is added as metadata. The dataset generation unit also applies an algorithm that allows the generation AI to evaluate the quality and reliability of the dataset and add metadata. For example, information about the consistency and integrity of the data is added as metadata. This ensures the quality and reliability of the dataset.

[0073] The dataset generation unit can use the emotion estimation function to consider user emotions during the dataset generation process and prioritize generating data that elicits positive emotions. For example, the generation AI considers user emotions during the dataset generation process and prioritizes generating data that elicits positive emotions. For example, it prioritizes collecting data that is likely to interest users. The dataset generation unit also uses the emotion estimation function to monitor user emotions in real time during the dataset generation process and generate data that elicits positive emotions. For example, it adjusts the data based on user feedback. The dataset generation unit also considers user emotions during the dataset generation process and provides an interface that elicits positive emotions. For example, it designs an interface that allows users to enjoy collecting data. This makes it possible to generate datasets that take user emotions into consideration.

[0074] The dataset generation unit can generate a dataset by integrating different data sources. For example, the generation AI integrates different data sources to generate a dataset. For example, a dataset is created by combining sensor data and social media data. The dataset generation unit also integrates data collected from different data sources to generate a dataset. For example, data from IoT devices is integrated with user social media posts. The dataset generation unit also analyzes different data sources to generate an integrated dataset. For example, a dataset is created by combining environmental sensor data and user behavior data. This makes it possible to generate a dataset by integrating different data sources.

[0075] The dataset generation unit can share the generated dataset between different AI models to build a mutually usable dataset. For example, a dataset generated by the generation AI is shared between different AI models to build a mutually usable dataset. For example, an image recognition model and a natural language processing model use the same dataset. The dataset generation unit also develops a system that shares datasets between different AI models to build a mutually usable dataset. For example, it integrates datasets corresponding to different tasks. The dataset generation unit also shares the dataset generated by the generation AI between different AI models to maintain data consistency. For example, data consistency is ensured by having multiple models use the same dataset. This makes it possible to share datasets between different AI models to build a mutually usable dataset.

[0076] During the evaluation index selection process, the evaluation unit can propose customizable indexes according to the model's characteristics and application. For example, the generative AI proposes evaluation indexes according to the model's characteristics and application. For example, it proposes precision and recall for image recognition models, and BLEU score and ROUGE score for natural language processing models. Furthermore, during the evaluation index selection process, the generative AI analyzes the model's characteristics and proposes the optimal index. For example, it proposes NDCG and MAP for recommendation systems. Furthermore, during the evaluation index selection process, the generative AI proposes customizable indexes according to the model's application. For example, it proposes specificity and sensitivity for models in the medical field. This makes it possible to propose customizable indexes according to the model's characteristics and application.

[0077] The evaluation unit uses the emotion estimation function to analyze the user's emotional response to the evaluation results and can emphasize emotionally positive evaluation results. For example, the generation AI analyzes the user's emotional response to the evaluation results and emphasizes positive evaluation results. For example, positive evaluation results are displayed prominently based on the user's emotional score. The evaluation unit also uses the emotion estimation function to monitor the user's emotional response to the evaluation results in real time and emphasize positive evaluation results. For example, the evaluation results are adjusted based on user feedback. The evaluation unit also builds a system in which the generation AI analyzes the user's emotional response to the evaluation results and emphasizes emotionally positive evaluation results. For example, evaluation results with a high positive emotional score are displayed preferentially. This makes it possible to analyze the user's emotional response and emphasize positive evaluation results.

[0078] The processing flow of the second embodiment will be briefly explained below.

[0079] Step 1: The model classification unit classifies AI models from the source code. For example, the generative AI classifies the purpose and function of each model based on the structure and content of the source code. The generative AI can also identify image recognition models, natural language processing models, recommendation system models, etc. The generative AI analyzes the entire provided source code and identifies different AI models. Step 2: The dataset generation unit generates a dataset appropriate for each of the separated AI models. For example, the generation AI generates an image dataset for an image recognition model and a text dataset for a natural language processing model. The generation AI also generates an appropriate dataset based on information about the characteristics and purpose of each model. Step 3: The evaluation unit uses the generated dataset to evaluate the AI ​​model using multiple indicators. For example, the generation AI is evaluated using indicators such as precision, recall, F1 score, and ROC curve. The generation AI also performs evaluation based on a list of indicators used for evaluation and the AI ​​model to be evaluated. This allows the AI ​​model evaluation system according to the embodiment to ensure transparency and accountability of AI models and provide highly reliable evaluations.

[0080] 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.

[0081] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<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.

[0082] 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.

[0083] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0084] 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.

[0085] 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.

[0086] 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.

[0087] 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.

[0088] 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).

[0089] 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.

[0090] 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.

[0091] 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.

[0092] 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.

[0093] 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.

[0094] 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.

[0095] 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.

[0096] 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 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.

[0097] 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.

[0098] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0099] 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.

[0100] 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.

[0101] 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.

[0102] 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.

[0103] 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).

[0104] 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.

[0105] 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.

[0106] 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.

[0107] 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.

[0108] In the headset type terminal 314, 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. 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 specific processing unit 290 using these models.

[0109] 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.

[0110] 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.

[0111] 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 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.

[0112] 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.

[0113] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0114] 7, the 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.

[0115] 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.

[0116] 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.

[0117] 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.

[0118] 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).

[0119] 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.

[0120] 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.

[0121] 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.

[0122] 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.

[0123] 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.

[0124] In the robot 414, 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 robot 414 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.

[0125] 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.

[0126] 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.

[0127] 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 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.

[0128] 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.

[0129] 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.

[0130] 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.

[0131] 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.

[0132] 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).

[0133] 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 area called "reaction," where sensation is dominant. The right half of the emotion map lists emotions belonging to the area called "situation," where situational awareness is dominant.

[0134] 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."

[0135] 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.

[0136] 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.

[0137] 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.

[0138] 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.

[0139] 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.

[0140] 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.

[0141] 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.

[0142] 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.

[0143] 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.

[0144] 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.

[0145] 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.

[0146] 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. [Explanation of symbols]

[0147] 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 model classification unit that classifies AI models from source code; a dataset generation unit that generates a dataset suitable for the AI ​​model classified by the model classification unit; an evaluation unit that evaluates the AI ​​model using a plurality of indexes using the dataset generated by the dataset generation unit. A system characterized by:

2. The model classification unit Analyze the comments and documents in the source code and classify the AI ​​models based on the developer's intentions and design philosophy.

2. The system of claim 1.

3. The model classification unit Analyzing the version history of the source code to track and separate the evolution and changes of the model 2. The system of claim 1.

4. The model classification unit The developer's emotions are estimated from the source code description, and the emotionally important parts are prioritized for analysis.

2. The system of claim 1.

5. The dataset generation unit Automatically detect and correct data bias and imbalance during the creation of said dataset.

2. The system of claim 1.

6. The dataset generation unit Evaluating the quality and reliability of the data set to be generated and adding appropriate metadata 2. The system of claim 1.

7. The dataset generation unit In the process of generating the data set, data that elicits positive emotions is preferentially generated based on the user's emotions.

2. The system of claim 1.

8. The dataset generation unit Integrating different data sources to generate said dataset 2. The system of claim 1.

Images