System

Abstract

An object of the system according to the embodiment is to construct an image model specialized for Japan and to perform a bias check based on discomfort.SOLUTION: A system according to an embodiment includes a collection unit, a construction unit, a check unit, and a presentation unit. The collection unit collects data specific to Japan. The construction unit constructs an image model specialized in Japan based on the data collected by the collection unit. The checking unit performs a bias check for evaluating discomfort with respect to the image model constructed by the constructing unit. The presentation unit presents the bias evidence obtained by the check unit.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] With conventional technology, bias against Asians in image AI has become apparent, and there has been a lack of countermeasures specific to Japan.

[0005] The system of the embodiment aims to build an image model specialized for Japan and perform bias checks based on discomfort. [Means for solving the problem]

[0006] The system according to the embodiment includes a collection unit, a construction unit, a check unit, and a presentation unit. The collection unit collects data specific to Japan. The construction unit constructs an image model specific to Japan based on the data collected by the collection unit. The check unit performs a bias check to evaluate discomfort with the image model constructed by the construction unit. The presentation unit presents bias evidence obtained by the check unit. [Effects of the Invention]

[0007] The system according to the embodiment can build an image model specific to Japan and perform a bias check based on discomfort. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

[0028] (Example 1) A system according to an embodiment of the present invention collects data specific to Japan, constructs a large-scale image model specific to Japan using a generative AI, performs a bias check based on discomfort, and presents bias evidence. This system collects data reflecting Japanese culture and social background and trains a generative AI to construct an image model specific to Japan. Next, it performs a bias check based on discomfort on the constructed image model and corrects the bias based on user feedback. Furthermore, it presents bias evidence and can seriously address issues of AI ethics. For example, the system collects data including traditional Japanese festivals, landscapes, and scenes from everyday life. Next, it constructs an image model specific to Japan using a generative AI. Next, it evaluates whether users feel uncomfortable with the images generated by the generative AI. Next, it corrects the bias based on user feedback. Next, it verifies whether bias exists in the images generated by the generative AI and publishes the results in a report. This allows the system to provide fair services specific to Japan and seriously address issues of AI ethics. This allows the system to address bias issues that arise with the evolution of generative AI and provide fair services specific to Japan. Furthermore, seriously addressing issues of AI ethics can earn the trust of society as a whole.

[0029] A bias check system according to an embodiment includes a collection unit, a construction unit, a check unit, and a presentation unit. The collection unit collects data specific to Japan. The data specific to Japan includes, but is not limited to, Japanese culture, social background, and language. The collection unit collects data including, for example, traditional Japanese festivals, landscapes, and scenes from everyday life. The collection unit can also collect public data on the Internet and data provided by users. For example, the collection unit collects images related to Japan from an image database on the Internet. The collection unit can also collect image data provided by users. The construction unit constructs an image model specific to Japan based on the data collected by the collection unit. The image model is realized using, for example, a deep learning model or a convolutional neural network (CNN), but is not limited to, these examples. For example, the construction unit constructs the image model specific to Japan using deep learning technology. The construction unit can also construct the image model using a convolutional neural network (CNN). The construction unit can also construct the image model using generative AI. For example, the construction unit constructs an image model specialized for Japan by having the generation AI learn data that reflects Japanese culture and social background. The check unit performs a bias check based on discomfort on the image model constructed by the construction unit. The bias check is performed, for example, by evaluating whether a user feels uncomfortable with the generated image, but is not limited to such an example. For example, if a user feels "uncomfortable" with the generated image, the check unit collects that feedback and feeds it back to the generation AI. The check unit can also evaluate discomfort using a questionnaire survey or psychological evaluation. For example, the check unit conducts a questionnaire survey to evaluate the user's discomfort. The check unit can also evaluate discomfort using psychological evaluation. The presentation unit presents bias evidence obtained by the check unit. The bias evidence is presented in the form of, for example, statistical data or user feedback, but is not limited to such an example. For example, the presentation unit publishes the bias evidence in the form of a report.The presentation unit can also visually present bias evidence using graphs or charts. For example, the presentation unit displays bias evidence as graphs or charts. This allows the bias check system according to the embodiment to perform bias checks using data specific to Japan and seriously address issues of AI ethics.

[0030] The collection unit can collect data based on Japanese culture and social background. Examples of Japanese culture and social background include, but are not limited to, traditional events and social customs. The collection unit can collect data including, for example, traditional Japanese festivals, landscapes, and scenes from everyday life. The collection unit can also collect public data on the Internet and data provided by users. For example, the collection unit can collect images related to Japan from an image database on the Internet. The collection unit can also collect image data provided by users. This allows for the collection of data reflecting Japanese culture and social background, thereby building a more accurate image model. Some or all of the above-described processing in the collection unit can be performed using, for example, AI, or can be performed without AI. For example, when collecting images related to Japan from an image database on the Internet, the collection unit can automatically select highly relevant images using AI.

[0031] The check unit can evaluate whether a user feels uncomfortable with an image generated by the generation AI. Evaluation of discomfort includes, but is not limited to, methods such as a questionnaire survey and psychological evaluation. For example, if a user feels uncomfortable with a generated image, the check unit collects that feedback and provides it to the generation AI. The check unit can also conduct a questionnaire survey to evaluate the user's discomfort. For example, the check unit can evaluate the user's discomfort with the generated image through a questionnaire survey. The check unit can also evaluate the discomfort using a psychological evaluation. For example, the check unit can evaluate the user's discomfort with the generated image through a psychological evaluation. This makes it easier to detect the presence of bias by evaluating the user's discomfort. Some or all of the above-described processing in the check unit can be performed using AI, for example, or without AI. For example, the check unit can input the results of a questionnaire survey into AI, which can then evaluate the discomfort.

[0032] The check unit can correct biases based on user feedback. Bias corrections include, but are not limited to, types such as gender bias and age bias. For example, the check unit corrects gender biases based on user feedback. The check unit can also correct age biases based on user feedback. For example, the check unit corrects gender biases in images generated by the generation AI based on user feedback. The check unit can also correct age biases in images generated by the generation AI based on user feedback. This allows for correcting biases based on user feedback to provide a fairer image model. Some or all of the above-described processing in the check unit may be performed using AI, for example, or may be performed without using AI. For example, the check unit can input user feedback into AI, which then corrects the biases.

[0033] The presentation unit can verify whether bias exists in the images generated by the generation AI and publish the results in the form of a report. Examples of report formats include, but are not limited to, text reports and graph reports. For example, the presentation unit publishes bias evidence as a text report. The presentation unit can also publish bias evidence as a graph report. For example, the presentation unit publishes a report that describes bias evidence in detail in text format. The presentation unit can also publish a report that visually displays bias evidence using graphs and charts. This makes it possible to verify the existence of bias and publish the results, thereby raising awareness of AI ethics issues. Some or all of the above-described processing in the presentation unit may be performed using AI, or may be performed without using AI. For example, the presentation unit can input bias evidence into AI, which then generates a report.

[0034] The presentation unit can provide information on issues related to AI ethics. Issues related to AI ethics include, but are not limited to, privacy protection and fairness. The presentation unit can provide, for example, information on privacy protection. The presentation unit can also provide information on fairness. For example, the presentation unit can provide guidelines on privacy protection for data collected by AI. The presentation unit can also provide information on fairness in images generated by AI. This can raise awareness of issues related to AI ethics and gain the trust of society as a whole. Some or all of the above-described processing in the presentation unit can be performed using AI, or can be performed without using AI. For example, the presentation unit can input information related to AI ethics into AI, which can then organize and provide the information.

[0035] The collection unit can collect data based on the culture and social background of each region in Japan and construct a region-specific image model. Region-specific data includes, but is not limited to, events and ceremonies, architectural styles, and food cultures unique to each region. For example, the collection unit collects data on events and ceremonies unique to each region, such as the Hokkaido Snow Festival and Okinawa's Eisa. The collection unit can also collect data on architectural styles unique to each region, such as traditional townhouses in Kyoto and modern buildings in Tokyo. For example, the collection unit collects data on regional food cultures and traditional cuisines and reflects them in the image model. In this way, by collecting data unique to each region, a region-specific image model can be constructed. Some or all of the above-described processing by the collection unit may be performed using, for example, AI, or may be performed without AI. For example, the collection unit can input data on regional events and ceremonies into AI, which then automatically collects related data.

[0036] The collection unit can filter data according to seasonal events and occasions when collecting data. Examples of seasonal events and occasions include, but are not limited to, cherry blossom viewing and entrance ceremonies in spring, fireworks festivals and beach trips in summer, and Christmas and New Year's in winter. For example, the collection unit collects data on cherry blossom viewing and entrance ceremonies in spring and reflects the data in the image model. The collection unit can also collect data on fireworks festivals and beach trips in summer and reflect the data in the image model. For example, the collection unit collects data on Christmas and New Year's in winter and reflects the data in the image model. This allows for filtering data based on seasonal events and occasions, thereby collecting a wider variety of data. Some or all of the above-described processing in the collection unit may be performed using, or without, AI. For example, the collection unit can input data on seasonal events and occasions into AI, and the AI ​​can automatically filter related data.

[0037] When collecting data, the collection unit can customize the collection method based on the user's past feedback. Customizing the collection method includes, but is not limited to, for example, how to reflect the feedback and adjusting the collection process. For example, the collection unit prioritizes a specific data collection method based on feedback provided by the user in the past. The collection unit can also avoid a data collection method that the user has expressed discomfort with and adopt an alternative method. For example, the collection unit adjusts the frequency and timing of data collection based on the user's feedback. This allows for customizing the collection method to reflect the user's past feedback, thereby collecting more appropriate data. Some or all of the above-described processing by the collection unit may be performed using, or without, AI. For example, the collection unit can input the user's past feedback into AI, which then customizes the collection method.

[0038] When collecting data, the collection unit can prioritize collecting highly relevant data by taking into account the user's geographical location information. Examples of uses of geographical location information include, but are not limited to, GPS data and location information services. For example, if the user is in Tokyo, the collection unit can prioritize collecting data related to Tokyo. Furthermore, if the user is in Kyoto, the collection unit can prioritize collecting data related to Kyoto's traditional culture. For example, if the user is in Okinawa, the collection unit can prioritize collecting data related to Okinawa's natural landscapes. This allows for more relevant data to be collected by taking the user's geographical location information into account. Some or all of the above-described processing in the collection unit may be performed using, or without, AI. For example, the collection unit can input the user's geographical location information into AI, which can then automatically select highly relevant data.

[0039] The collection unit may analyze the user's social media activity during data collection and collect relevant data based on the activity. Examples of analyzing social media activity include, but are not limited to, analyzing posts and followers. The collection unit may collect relevant data based on, for example, images and videos shared by the user on social media. The collection unit may also analyze the user's social media posts to collect data that may be of interest to the user. For example, the collection unit may collect relevant data based on the activities of the user's friends on social media. This allows for the collection of more relevant data by analyzing the user's social media activity. Some or all of the above-described processing by the collection unit may be performed using, for example, AI, or may be performed without using AI. For example, the collection unit may input the user's social media activity data into AI, which then automatically collects the relevant data.

[0040] When constructing an image model, the construction unit can adjust the level of detail of the model according to the importance of the data. Evaluation of the importance of the data includes, but is not limited to, criteria such as the reliability and relevance of the data. For example, the construction unit constructs a detailed image model based on data with high importance. The construction unit can also construct a simplified image model based on data with low importance. For example, the construction unit gradually adjusts the level of detail of the model according to the importance of the data. This allows an image model that reflects more important data to be constructed by adjusting the level of detail of the model based on the importance of the data. Some or all of the above-described processing in the construction unit may be performed using, or without, AI. For example, the construction unit can input the importance of the data into AI, which then adjusts the level of detail of the model.

[0041] When constructing an image model, the construction unit can apply different construction algorithms based on the category of data. Examples of data category classification include, but are not limited to, image category and text category. For example, the construction unit can apply a specific algorithm to landscape images to construct a detailed landscape model. The construction unit can also apply a different algorithm to portrait images to construct a detailed portrait model. For example, the construction unit selects an optimal algorithm for each category and constructs an image model. This allows for the construction of a more accurate image model by applying the optimal algorithm depending on the data category. Some or all of the above-described processing in the construction unit may be performed using, for example, AI, or may be performed without using AI. For example, the construction unit can input the data category into AI, which selects the optimal algorithm to construct an image model.

[0042] When constructing an image model, the construction unit can improve the accuracy of the model based on the user's past construction results. Use of past construction results includes, but is not limited to, the accuracy of past models and user feedback. For example, the construction unit improves the accuracy of the current model based on image models previously constructed by the user. The construction unit can also adjust the accuracy of the model by referring to the user's past feedback. For example, the construction unit analyzes the user's past construction results and adopts an optimal model construction method. This allows the accuracy of the model to be improved by referring to the user's past construction results. Some or all of the above-described processing in the construction unit may be performed using, for example, AI, or may be performed without using AI. For example, the construction unit can input the user's past construction results into AI, which can then improve the accuracy of the model.

[0043] When constructing an image model, the construction unit can determine the priority of models based on the time of data submission. Use of the time of data submission includes, but is not limited to, the priority of the latest data and the frequency of submission. For example, the construction unit constructs an image model by preferentially using the latest data. The construction unit can also use older data by lowering the priority. For example, the construction unit evaluates the importance of data based on the time of submission and determines the priority of models. By determining the priority of models based on the time of data submission, an image model that reflects the latest data preferentially can be constructed. Some or all of the above-described processing in the construction unit may be performed using, for example, AI, or may be performed without using AI. For example, the construction unit can input the time of data submission into AI, which then determines the priority of models.

[0044] When constructing an image model, the construction unit can adjust the order of models based on the relevance of the data. Evaluation of the relevance of the data includes, but is not limited to, criteria such as correlation analysis and relevance scores. For example, the construction unit constructs an image model by preferentially using highly relevant data. The construction unit can also postpone the use of less relevant data. For example, the construction unit adjusts the construction order of models based on the relevance of the data. In this way, by adjusting the order of models based on the relevance of the data, an image model that preferentially reflects more relevant data can be constructed. Some or all of the above-described processing in the construction unit may be performed using, for example, AI, or may be performed without using AI. For example, the construction unit can input the relevance of the data into AI, which can then adjust the order of the models.

[0045] When constructing an image model, the construction unit can adjust the use of technical terminology in the model based on the user's level of expertise. Evaluation of the level of expertise includes, but is not limited to, criteria such as qualifications and past experience. For example, the construction unit can construct an image model using detailed technical terminology for a user with high expertise. The construction unit can also construct an image model using simpler terminology for a user with low expertise. For example, the construction unit gradually adjusts the use of technical terminology in the model according to the user's level of expertise. This allows for the provision of an image model that is easier to understand by adjusting the use of technical terminology according to the user's level of expertise. Some or all of the above-described processing in the construction unit can be performed using, for example, AI, or without AI. For example, the construction unit can input the user's level of expertise into AI, which can then adjust the use of technical terminology.

[0046] The check unit can improve the accuracy of the bias check based on the interrelationships between images during the bias check. Evaluation of the interrelationships between images can be based on, but not limited to, criteria such as co-occurrence networks and correlation analysis. For example, the check unit can analyze the interrelationships between images to accurately check for the presence or absence of bias. The check unit can also adjust the criteria for the bias check by taking the interrelationships between images into account. For example, the check unit can evaluate the results of the bias check based on the interrelationships between images. This allows the accuracy of the bias check to be improved by taking the interrelationships between images into account. Some or all of the above-described processing in the check unit can be performed using, or without, AI. For example, the check unit can input image interrelationship data into AI, which can then improve the accuracy of the bias check.

[0047] When performing a bias check, the check unit can perform a check based on the attribute information of the image submitter. Use of the submitter's attribute information includes, but is not limited to, age, gender, and occupation. The check unit can, for example, perform a bias check taking into account the age and gender of the image submitter. The check unit can also perform a bias check taking into account the occupation and background of the image submitter. For example, the check unit adjusts the bias check criteria based on the attribute information of the image submitter. This allows for a more accurate bias check by taking into account the attribute information of the image submitter. Some or all of the above-described processing in the check unit can be performed using, for example, AI, or without AI. For example, the check unit can input the submitter's attribute information into AI, which can then perform the bias check.

[0048] During bias checks, the check unit can weight checks based on the frequency of image submission. Evaluation of submission frequency includes, but is not limited to, criteria such as the number of submissions and the interval between submissions. For example, the check unit can focus bias checks on images submitted frequently. The check unit can also perform lighter bias checks on images submitted infrequently. For example, the check unit can adjust the weighting of the bias check based on the submission frequency. This allows for more efficient bias checks by weighting the checks based on the frequency of image submission. Some or all of the above-described processing in the check unit can be performed using, for example, AI, or without AI. For example, the check unit can input submission frequency data into AI, which can then weight the bias check.

[0049] When performing a bias check, the check unit can perform a check based on the geographic distribution of the images. Evaluation of the geographic distribution can include, but is not limited to, criteria such as data distribution by region and geographic clusters. For example, the check unit can analyze the geographic distribution of the images and check for bias toward a specific region. The check unit can also adjust the criteria for the bias check based on the geographic distribution. For example, the check unit can evaluate the results of the bias check taking into account the geographic distribution. This makes it possible to check for bias toward a specific region by taking into account the geographic distribution of the images. Some or all of the above-described processing in the check unit can be performed using, or without, AI. For example, the check unit can input geographic distribution data into AI, which can then perform the bias check.

[0050] During the bias check, the check unit can improve the accuracy of the check by referring to related literature of the image. Reference to related literature includes, but is not limited to, criteria such as a literature database and the number of citations. For example, the check unit can refer to literature related to the image to accurately check for the presence or absence of bias. The check unit can also adjust the criteria for the bias check based on the related literature. For example, the check unit can evaluate the results of the bias check by referring to the related literature. Thus, by referring to the related literature of the image, the accuracy of the bias check can be improved. Some or all of the above-described processing in the check unit can be performed using, for example, AI, or can be performed without using AI. For example, the check unit can input related literature data into AI, which can improve the accuracy of the bias check.

[0051] When performing a bias check, the check unit can perform a check based on the market value of the image. Evaluation of market value includes, but is not limited to, criteria such as sales data and evaluation scores. For example, the check unit can focus the bias check on images with high market value. The check unit can also perform a lighter bias check on images with low market value. For example, the check unit can adjust the weighting of the bias check based on market value. This allows the bias check to be focused on more important images by taking the market value of the image into consideration. Some or all of the above-described processing in the check unit can be performed using, or without, AI. For example, the check unit can input market value data into AI, which then performs the bias check.

[0052] When presenting bias evidence, the presentation unit can optimize current evidence based on past evidence data. Use of past evidence data includes, but is not limited to, criteria such as past evaluation results and feedback. The presentation unit, for example, optimizes current bias evidence based on past evidence data. The presentation unit can also adjust the presentation method of bias evidence by referring to past evidence data. For example, the presentation unit analyzes past evidence data and presents optimal bias evidence. This allows current bias evidence to be optimized by referring to the past evidence data. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input past evidence data into AI, which then optimizes current evidence.

[0053] When presenting bias evidence, the presentation unit can improve the method of presenting the evidence based on user feedback. Improvements to the method of presenting the evidence include, but are not limited to, criteria such as a method of reflecting user feedback and adjusting the presentation format. For example, the presentation unit improves the method of presenting the bias evidence based on user feedback. The presentation unit can also adjust the level of detail of the evidence by reflecting user feedback. For example, the presentation unit improves the visual presentation method of the evidence by referring to user feedback. In this way, the presentation method of the evidence can be improved by reflecting user feedback. Some or all of the above-mentioned processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input user feedback into AI, which can improve the method of presenting the evidence.

[0054] When presenting bias evidence, the presentation unit can customize the method of presenting the evidence based on the user's past feedback. Customization of the method of presenting the evidence includes, but is not limited to, criteria such as a method of reflecting the user's past feedback and adjusting the presentation format. For example, the presentation unit customizes the method of presenting the evidence based on the user's past feedback. The presentation unit can also adjust the level of detail of the evidence by reflecting the user's past feedback. For example, the presentation unit customizes the visual presentation method of the evidence by referring to the user's past feedback. In this way, the method of presenting the evidence can be customized by reflecting the user's past feedback. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input the user's past feedback into AI, which can then customize the method of presenting the evidence.

[0055] When presenting bias evidence, the presentation unit can select an optimal evidence presentation method based on the user's geographical location information. Use of geographical location information includes, but is not limited to, GPS data, location information services, etc. For example, if the user is in Tokyo, the presentation unit can prioritize presenting bias evidence related to Tokyo. Furthermore, if the user is in Kyoto, the presentation unit can also prioritize presenting bias evidence related to Kyoto. For example, if the user is in Okinawa, the presentation unit can prioritize presenting bias evidence related to Okinawa. This allows for providing more relevant evidence by taking the user's geographical location information into consideration. Some or all of the above-described processing in the presentation unit may be performed using, or without, AI. For example, the presentation unit can input the user's geographical location information into AI, which can then select an optimal evidence presentation method.

[0056] When presenting bias evidence, the presentation unit can analyze the user's social media activity and suggest a method for presenting the evidence based on the activity. Analysis of social media activity includes, but is not limited to, criteria such as analysis of post content and analysis of followers. The presentation unit can present relevant bias evidence, for example, based on images and videos shared by the user on social media. The presentation unit can also analyze the user's social media posts to present bias evidence that may be of interest. For example, the presentation unit can present relevant bias evidence based on the activity of the user's friends on social media. This allows for analysis of the user's social media activity to provide more relevant evidence. Some or all of the above-described processing by the presentation unit can be performed using, for example, AI, or without AI. For example, the presentation unit can input the user's social media activity data into AI, which can then suggest a method for presenting evidence.

[0057] When presenting bias evidence, the presentation unit can customize the method of presenting the evidence based on the user's past feedback. Customization of the method of presenting the evidence includes, but is not limited to, criteria such as a method of reflecting the user's past feedback and adjusting the presentation format. For example, the presentation unit customizes the method of presenting the evidence based on the user's past feedback. The presentation unit can also adjust the level of detail of the evidence by reflecting the user's past feedback. For example, the presentation unit customizes the visual presentation method of the evidence by referring to the user's past feedback. In this way, the method of presenting the evidence can be customized by reflecting the user's past feedback. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input the user's past feedback into AI, which can then customize the method of presenting the evidence.

[0058] When performing a bias check, the check unit can perform a check based on the market value of the image. Evaluation of market value includes, but is not limited to, criteria such as sales data and evaluation scores. For example, the check unit can focus the bias check on images with high market value. The check unit can also perform a lighter bias check on images with low market value. For example, the check unit can adjust the weighting of the bias check based on market value. This allows the bias check to be focused on more important images by taking the market value of the image into consideration. Some or all of the above-described processing in the check unit can be performed using, or without, AI. For example, the check unit can input market value data into AI, which then performs the bias check.

[0059] When presenting bias evidence, the presentation unit can optimize current evidence based on past evidence data. Use of past evidence data includes, but is not limited to, criteria such as past evaluation results and feedback. The presentation unit, for example, optimizes current bias evidence based on past evidence data. The presentation unit can also adjust the presentation method of bias evidence by referring to past evidence data. For example, the presentation unit analyzes past evidence data and presents optimal bias evidence. This allows current bias evidence to be optimized by referring to the past evidence data. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input past evidence data into AI, which then optimizes current evidence.

[0060] When presenting bias evidence, the presentation unit can improve the method of presenting the evidence based on user feedback. Improvements to the method of presenting the evidence include, but are not limited to, criteria such as a method of reflecting user feedback and adjusting the presentation format. For example, the presentation unit improves the method of presenting the bias evidence based on user feedback. The presentation unit can also adjust the level of detail of the evidence by reflecting user feedback. For example, the presentation unit improves the visual presentation method of the evidence by referring to user feedback. In this way, the presentation method of the evidence can be improved by reflecting user feedback. Some or all of the above-mentioned processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input user feedback into AI, which can improve the method of presenting the evidence.

[0061] When presenting bias evidence, the presentation unit can customize the method of presenting the evidence based on the user's past feedback. Customization of the method of presenting the evidence includes, but is not limited to, criteria such as a method of reflecting the user's past feedback and adjusting the presentation format. For example, the presentation unit customizes the method of presenting the evidence based on the user's past feedback. The presentation unit can also adjust the level of detail of the evidence by reflecting the user's past feedback. For example, the presentation unit customizes the visual presentation method of the evidence by referring to the user's past feedback. In this way, the method of presenting the evidence can be customized by reflecting the user's past feedback. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input the user's past feedback into AI, which can then customize the method of presenting the evidence.

[0062] When presenting bias evidence, the presentation unit can select an optimal evidence presentation method based on the user's geographical location information. Use of geographical location information includes, but is not limited to, GPS data, location information services, etc. For example, if the user is in Tokyo, the presentation unit can prioritize presenting bias evidence related to Tokyo. Furthermore, if the user is in Kyoto, the presentation unit can also prioritize presenting bias evidence related to Kyoto. For example, if the user is in Okinawa, the presentation unit can prioritize presenting bias evidence related to Okinawa. This allows for providing more relevant evidence by taking the user's geographical location information into consideration. Some or all of the above-described processing in the presentation unit may be performed using, or without, AI. For example, the presentation unit can input the user's geographical location information into AI, which can then select an optimal evidence presentation method.

[0063] When presenting bias evidence, the presentation unit can analyze the user's social media activity and suggest a method for presenting the evidence based on the activity. Analysis of social media activity includes, but is not limited to, criteria such as analysis of post content and analysis of followers. The presentation unit can present relevant bias evidence, for example, based on images and videos shared by the user on social media. The presentation unit can also analyze the user's social media posts to present bias evidence that may be of interest. For example, the presentation unit can present relevant bias evidence based on the activity of the user's friends on social media. This allows for analysis of the user's social media activity to provide more relevant evidence. Some or all of the above-described processing by the presentation unit can be performed using, for example, AI, or without AI. For example, the presentation unit can input the user's social media activity data into AI, which can then suggest a method for presenting evidence.

[0064] When presenting bias evidence, the presentation unit can customize the method of presenting the evidence by reflecting the user's past feedback. Customization of the method of presenting the evidence includes, but is not limited to, criteria such as a method of reflecting the user's past feedback and adjusting the presentation format. For example, the presentation unit customizes the method of presenting the evidence based on the user's past feedback. The presentation unit can also adjust the level of detail of the evidence by reflecting the user's past feedback. For example, the presentation unit customizes the visual presentation method of the evidence by referring to the user's past feedback. In this way, the method of presenting the evidence can be customized by reflecting the user's past feedback. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input the user's past feedback into AI, which can then customize the method of presenting the evidence.

[0065] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0066] The collection unit can collect data based on the culture and social background of each region in Japan and construct a region-specific image model. Region-specific data includes, but is not limited to, events and ceremonies, architectural styles, and food cultures unique to each region. For example, the collection unit collects data on events and ceremonies unique to each region, such as the Hokkaido Snow Festival and Okinawa's Eisa. The collection unit can also collect data on architectural styles unique to each region, such as traditional townhouses in Kyoto and modern buildings in Tokyo. For example, the collection unit collects data on regional food cultures and traditional cuisines and reflects them in the image model. In this way, by collecting data unique to each region, a region-specific image model can be constructed. Some or all of the above-described processing by the collection unit may be performed using, for example, AI, or may be performed without AI. For example, the collection unit can input data on regional events and ceremonies into AI, which then automatically collects related data.

[0067] The collection unit can filter data according to seasonal events and occasions when collecting data. Examples of seasonal events and occasions include, but are not limited to, cherry blossom viewing and entrance ceremonies in spring, fireworks festivals and beach trips in summer, and Christmas and New Year's in winter. For example, the collection unit collects data on cherry blossom viewing and entrance ceremonies in spring and reflects the data in the image model. The collection unit can also collect data on fireworks festivals and beach trips in summer and reflect the data in the image model. For example, the collection unit collects data on Christmas and New Year's in winter and reflects the data in the image model. This allows for filtering data based on seasonal events and occasions, thereby collecting a wider variety of data. Some or all of the above-described processing in the collection unit may be performed using, or without, AI. For example, the collection unit can input data on seasonal events and occasions into AI, and the AI ​​can automatically filter related data.

[0068] When collecting data, the collection unit can customize the collection method based on the user's past feedback. Customizing the collection method includes, but is not limited to, for example, how to reflect the feedback and adjusting the collection process. For example, the collection unit prioritizes a specific data collection method based on feedback provided by the user in the past. The collection unit can also avoid a data collection method that the user has expressed discomfort with and adopt an alternative method. For example, the collection unit adjusts the frequency and timing of data collection based on the user's feedback. This allows for customizing the collection method to reflect the user's past feedback, thereby collecting more appropriate data. Some or all of the above-described processing by the collection unit may be performed using, or without, AI. For example, the collection unit can input the user's past feedback into AI, which then customizes the collection method.

[0069] When collecting data, the collection unit can prioritize collecting highly relevant data by taking into account the user's geographical location information. Examples of uses of geographical location information include, but are not limited to, GPS data and location information services. For example, if the user is in Tokyo, the collection unit can prioritize collecting data related to Tokyo. Furthermore, if the user is in Kyoto, the collection unit can prioritize collecting data related to Kyoto's traditional culture. For example, if the user is in Okinawa, the collection unit can prioritize collecting data related to Okinawa's natural landscapes. This allows for more relevant data to be collected by taking the user's geographical location information into account. Some or all of the above-described processing in the collection unit may be performed using, or without, AI. For example, the collection unit can input the user's geographical location information into AI, which can then automatically select highly relevant data.

[0070] When constructing an image model, the construction unit can adjust the level of detail of the model according to the importance of the data. Evaluation of the importance of the data includes, but is not limited to, criteria such as the reliability and relevance of the data. For example, the construction unit constructs a detailed image model based on data with high importance. The construction unit can also construct a simplified image model based on data with low importance. For example, the construction unit gradually adjusts the level of detail of the model according to the importance of the data. This allows an image model that reflects more important data to be constructed by adjusting the level of detail of the model based on the importance of the data. Some or all of the above-described processing in the construction unit may be performed using, or without, AI. For example, the construction unit can input the importance of the data into AI, which then adjusts the level of detail of the model.

[0071] When constructing an image model, the construction unit can apply different construction algorithms based on the category of data. Examples of data category classification include, but are not limited to, image category and text category. For example, the construction unit can apply a specific algorithm to landscape images to construct a detailed landscape model. The construction unit can also apply a different algorithm to portrait images to construct a detailed portrait model. For example, the construction unit selects an optimal algorithm for each category and constructs an image model. This allows for the construction of a more accurate image model by applying the optimal algorithm depending on the data category. Some or all of the above-described processing in the construction unit may be performed using, for example, AI, or may be performed without using AI. For example, the construction unit can input the data category into AI, which selects the optimal algorithm to construct an image model.

[0072] When constructing an image model, the construction unit can improve the accuracy of the model based on the user's past construction results. Use of past construction results includes, but is not limited to, the accuracy of past models and user feedback. For example, the construction unit improves the accuracy of the current model based on image models previously constructed by the user. The construction unit can also adjust the accuracy of the model by referring to the user's past feedback. For example, the construction unit analyzes the user's past construction results and adopts an optimal model construction method. This allows the accuracy of the model to be improved by referring to the user's past construction results. Some or all of the above-described processing in the construction unit may be performed using, for example, AI, or may be performed without using AI. For example, the construction unit can input the user's past construction results into AI, which can then improve the accuracy of the model.

[0073] When constructing an image model, the construction unit can determine the priority of models based on the time of data submission. Use of the time of data submission includes, but is not limited to, the priority of the latest data and the frequency of submission. For example, the construction unit constructs an image model by preferentially using the latest data. The construction unit can also use older data by lowering the priority. For example, the construction unit evaluates the importance of data based on the time of submission and determines the priority of models. By determining the priority of models based on the time of data submission, an image model that reflects the latest data preferentially can be constructed. Some or all of the above-described processing in the construction unit may be performed using, for example, AI, or may be performed without using AI. For example, the construction unit can input the time of data submission into AI, which then determines the priority of models.

[0074] When constructing an image model, the construction unit can adjust the order of models based on the relevance of the data. Evaluation of the relevance of the data includes, but is not limited to, criteria such as correlation analysis and relevance scores. For example, the construction unit constructs an image model by preferentially using highly relevant data. The construction unit can also postpone the use of less relevant data. For example, the construction unit adjusts the construction order of models based on the relevance of the data. In this way, by adjusting the order of models based on the relevance of the data, an image model that preferentially reflects more relevant data can be constructed. Some or all of the above-described processing in the construction unit may be performed using, for example, AI, or may be performed without using AI. For example, the construction unit can input the relevance of the data into AI, which can then adjust the order of the models.

[0075] When constructing an image model, the construction unit can adjust the use of technical terminology in the model based on the user's level of expertise. Evaluation of the level of expertise includes, but is not limited to, criteria such as qualifications and past experience. For example, the construction unit can construct an image model using detailed technical terminology for a user with high expertise. The construction unit can also construct an image model using simpler terminology for a user with low expertise. For example, the construction unit gradually adjusts the use of technical terminology in the model according to the user's level of expertise. This allows for the provision of an image model that is easier to understand by adjusting the use of technical terminology according to the user's level of expertise. Some or all of the above-described processing in the construction unit can be performed using, for example, AI, or without AI. For example, the construction unit can input the user's level of expertise into AI, which can then adjust the use of technical terminology.

[0076] The check unit can improve the accuracy of the bias check based on the interrelationships between images during the bias check. Evaluation of the interrelationships between images can be based on, but not limited to, criteria such as co-occurrence networks and correlation analysis. For example, the check unit can analyze the interrelationships between images to accurately check for the presence or absence of bias. The check unit can also adjust the criteria for the bias check by taking the interrelationships between images into account. For example, the check unit can evaluate the results of the bias check based on the interrelationships between images. This allows the accuracy of the bias check to be improved by taking the interrelationships between images into account. Some or all of the above-described processing in the check unit can be performed using, or without, AI. For example, the check unit can input image interrelationship data into AI, which can then improve the accuracy of the bias check.

[0077] When performing a bias check, the check unit can perform a check based on the attribute information of the image submitter. Use of the submitter's attribute information includes, but is not limited to, age, gender, and occupation. The check unit can, for example, perform a bias check taking into account the age and gender of the image submitter. The check unit can also perform a bias check taking into account the occupation and background of the image submitter. For example, the check unit adjusts the bias check criteria based on the attribute information of the image submitter. This allows for a more accurate bias check by taking into account the attribute information of the image submitter. Some or all of the above-described processing in the check unit can be performed using, for example, AI, or without AI. For example, the check unit can input the submitter's attribute information into AI, which can then perform the bias check.

[0078] During bias checks, the check unit can weight checks based on the frequency of image submission. Evaluation of submission frequency includes, but is not limited to, criteria such as the number of submissions and the interval between submissions. For example, the check unit can focus bias checks on images submitted frequently. The check unit can also perform lighter bias checks on images submitted infrequently. For example, the check unit can adjust the weighting of the bias check based on the submission frequency. This allows for more efficient bias checks by weighting the checks based on the frequency of image submission. Some or all of the above-described processing in the check unit can be performed using, for example, AI, or without AI. For example, the check unit can input submission frequency data into AI, which can then weight the bias check.

[0079] When performing a bias check, the check unit can perform a check based on the geographic distribution of the images. Evaluation of the geographic distribution can include, but is not limited to, criteria such as data distribution by region and geographic clusters. For example, the check unit can analyze the geographic distribution of the images and check for bias toward a specific region. The check unit can also adjust the criteria for the bias check based on the geographic distribution. For example, the check unit can evaluate the results of the bias check taking into account the geographic distribution. This makes it possible to check for bias toward a specific region by taking into account the geographic distribution of the images. Some or all of the above-described processing in the check unit can be performed using, or without, AI. For example, the check unit can input geographic distribution data into AI, which can then perform the bias check.

[0080] During the bias check, the check unit can improve the accuracy of the check by referring to related literature of the image. Reference to related literature includes, but is not limited to, criteria such as a literature database and the number of citations. For example, the check unit can refer to literature related to the image to accurately check for the presence or absence of bias. The check unit can also adjust the criteria for the bias check based on the related literature. For example, the check unit can evaluate the results of the bias check by referring to the related literature. Thus, by referring to the related literature of the image, the accuracy of the bias check can be improved. Some or all of the above-described processing in the check unit can be performed using, for example, AI, or can be performed without using AI. For example, the check unit can input related literature data into AI, which can improve the accuracy of the bias check.

[0081] When performing a bias check, the check unit can perform a check based on the market value of the image. Evaluation of market value includes, but is not limited to, criteria such as sales data and evaluation scores. For example, the check unit can focus the bias check on images with high market value. The check unit can also perform a lighter bias check on images with low market value. For example, the check unit can adjust the weighting of the bias check based on market value. This allows the bias check to be focused on more important images by taking the market value of the image into consideration. Some or all of the above-described processing in the check unit can be performed using, or without, AI. For example, the check unit can input market value data into AI, which then performs the bias check.

[0082] When presenting bias evidence, the presentation unit can optimize current evidence based on past evidence data. Use of past evidence data includes, but is not limited to, criteria such as past evaluation results and feedback. The presentation unit, for example, optimizes current bias evidence based on past evidence data. The presentation unit can also adjust the presentation method of bias evidence by referring to past evidence data. For example, the presentation unit analyzes past evidence data and presents optimal bias evidence. This allows current bias evidence to be optimized by referring to the past evidence data. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input past evidence data into AI, which then optimizes current evidence.

[0083] When presenting bias evidence, the presentation unit can improve the method of presenting the evidence based on user feedback. Improvements to the method of presenting the evidence include, but are not limited to, criteria such as a method of reflecting user feedback and adjusting the presentation format. For example, the presentation unit improves the method of presenting the bias evidence based on user feedback. The presentation unit can also adjust the level of detail of the evidence by reflecting user feedback. For example, the presentation unit improves the visual presentation method of the evidence by referring to user feedback. In this way, the presentation method of the evidence can be improved by reflecting user feedback. Some or all of the above-mentioned processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input user feedback into AI, which can improve the method of presenting the evidence.

[0084] When presenting bias evidence, the presentation unit can customize the method of presenting the evidence based on the user's past feedback. Customization of the method of presenting the evidence includes, but is not limited to, criteria such as a method of reflecting the user's past feedback and adjusting the presentation format. For example, the presentation unit customizes the method of presenting the evidence based on the user's past feedback. The presentation unit can also adjust the level of detail of the evidence by reflecting the user's past feedback. For example, the presentation unit customizes the visual presentation method of the evidence by referring to the user's past feedback. In this way, the method of presenting the evidence can be customized by reflecting the user's past feedback. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input the user's past feedback into AI, which can then customize the method of presenting the evidence.

[0085] When presenting bias evidence, the presentation unit can select an optimal evidence presentation method based on the user's geographical location information. Use of geographical location information includes, but is not limited to, GPS data, location information services, etc. For example, if the user is in Tokyo, the presentation unit can prioritize presenting bias evidence related to Tokyo. Furthermore, if the user is in Kyoto, the presentation unit can also prioritize presenting bias evidence related to Kyoto. For example, if the user is in Okinawa, the presentation unit can prioritize presenting bias evidence related to Okinawa. This allows for providing more relevant evidence by taking the user's geographical location information into consideration. Some or all of the above-described processing in the presentation unit may be performed using, or without, AI. For example, the presentation unit can input the user's geographical location information into AI, which can then select an optimal evidence presentation method.

[0086] When presenting bias evidence, the presentation unit can analyze the user's social media activity and suggest a method for presenting the evidence based on the activity. Analysis of social media activity includes, but is not limited to, criteria such as analysis of post content and analysis of followers. The presentation unit can present relevant bias evidence, for example, based on images and videos shared by the user on social media. The presentation unit can also analyze the user's social media posts to present bias evidence that may be of interest. For example, the presentation unit can present relevant bias evidence based on the activity of the user's friends on social media. This allows for analysis of the user's social media activity to provide more relevant evidence. Some or all of the above-described processing by the presentation unit can be performed using, for example, AI, or without AI. For example, the presentation unit can input the user's social media activity data into AI, which can then suggest a method for presenting evidence.

[0087] The processing flow of the first embodiment will be briefly explained below.

[0088] Step 1: The collection unit collects data specific to Japan. Data specific to Japan includes Japanese culture, social background, language, etc. The collection unit collects data including traditional Japanese festivals, landscapes, and scenes from everyday life. The collection unit also collects public data on the Internet and data provided by users. For example, it collects images related to Japan from image databases on the Internet and also collects image data provided by users. Step 2: The construction unit builds an image model specialized for Japan based on the data collected by the collection unit. The image model is realized using technologies such as deep learning models and convolutional neural networks (CNN). For example, an image model specialized for Japan can be built using deep learning technology and generative AI to learn data that reflects Japanese culture and social background. Step 3: The check unit performs a bias check based on discomfort on the image model constructed by the construction unit. The bias check is performed by evaluating whether the user feels uncomfortable with the generated image. For example, if a user feels uncomfortable with the generated image, that feedback is collected and fed back to the generation AI. Discomfort can also be assessed using questionnaire surveys and psychological evaluations. Step 4: The presentation unit presents the bias evidence obtained by the checking unit. The bias evidence is presented in the form of statistical data, user feedback, etc. For example, the bias evidence is published in the form of a report and presented visually using graphs and charts.

[0089] (Example 2) A system according to an embodiment of the present invention collects data specific to Japan, constructs a large-scale image model specific to Japan using a generative AI, performs a bias check based on discomfort, and presents bias evidence. This system collects data reflecting Japanese culture and social background and trains a generative AI to construct an image model specific to Japan. Next, it performs a bias check based on discomfort on the constructed image model and corrects the bias based on user feedback. Furthermore, it presents bias evidence and can seriously address issues of AI ethics. For example, the system collects data including traditional Japanese festivals, landscapes, and scenes from everyday life. Next, it constructs an image model specific to Japan using a generative AI. Next, it evaluates whether users feel uncomfortable with the images generated by the generative AI. Next, it corrects the bias based on user feedback. Next, it verifies whether bias exists in the images generated by the generative AI and publishes the results in a report. This allows the system to provide fair services specific to Japan and seriously address issues of AI ethics. This allows the system to address bias issues that arise with the evolution of generative AI and provide fair services specific to Japan. Furthermore, seriously addressing issues of AI ethics can earn the trust of society as a whole.

[0090] A bias check system according to an embodiment includes a collection unit, a construction unit, a check unit, and a presentation unit. The collection unit collects data specific to Japan. The data specific to Japan includes, but is not limited to, Japanese culture, social background, and language. The collection unit collects data including, for example, traditional Japanese festivals, landscapes, and scenes from everyday life. The collection unit can also collect public data on the Internet and data provided by users. For example, the collection unit collects images related to Japan from an image database on the Internet. The collection unit can also collect image data provided by users. The construction unit constructs an image model specific to Japan based on the data collected by the collection unit. The image model is realized using, for example, a deep learning model or a convolutional neural network (CNN), but is not limited to, these examples. For example, the construction unit constructs the image model specific to Japan using deep learning technology. The construction unit can also construct the image model using a convolutional neural network (CNN). The construction unit can also construct the image model using generative AI. For example, the construction unit constructs an image model specialized for Japan by having the generation AI learn data that reflects Japanese culture and social background. The check unit performs a bias check based on discomfort on the image model constructed by the construction unit. The bias check is performed, for example, by evaluating whether a user feels uncomfortable with the generated image, but is not limited to such an example. For example, if a user feels "uncomfortable" with the generated image, the check unit collects that feedback and feeds it back to the generation AI. The check unit can also evaluate discomfort using a questionnaire survey or psychological evaluation. For example, the check unit conducts a questionnaire survey to evaluate the user's discomfort. The check unit can also evaluate discomfort using psychological evaluation. The presentation unit presents bias evidence obtained by the check unit. The bias evidence is presented in the form of, for example, statistical data or user feedback, but is not limited to such an example. For example, the presentation unit publishes the bias evidence in the form of a report.The presentation unit can also visually present bias evidence using graphs or charts. For example, the presentation unit displays bias evidence as graphs or charts. This allows the bias check system according to the embodiment to perform bias checks using data specific to Japan and seriously address issues of AI ethics.

[0091] The collection unit can collect data based on Japanese culture and social background. Examples of Japanese culture and social background include, but are not limited to, traditional events and social customs. The collection unit can collect data including, for example, traditional Japanese festivals, landscapes, and scenes from everyday life. The collection unit can also collect public data on the Internet and data provided by users. For example, the collection unit can collect images related to Japan from an image database on the Internet. The collection unit can also collect image data provided by users. This allows for the collection of data reflecting Japanese culture and social background, thereby building a more accurate image model. Some or all of the above-described processing in the collection unit can be performed using, for example, AI, or can be performed without AI. For example, when collecting images related to Japan from an image database on the Internet, the collection unit can automatically select highly relevant images using AI.

[0092] The check unit can evaluate whether a user feels uncomfortable with an image generated by the generation AI. Evaluation of discomfort includes, but is not limited to, methods such as a questionnaire survey and psychological evaluation. For example, if a user feels uncomfortable with a generated image, the check unit collects that feedback and provides it to the generation AI. The check unit can also conduct a questionnaire survey to evaluate the user's discomfort. For example, the check unit can evaluate the user's discomfort with the generated image through a questionnaire survey. The check unit can also evaluate the discomfort using a psychological evaluation. For example, the check unit can evaluate the user's discomfort with the generated image through a psychological evaluation. This makes it easier to detect the presence of bias by evaluating the user's discomfort. Some or all of the above-described processing in the check unit can be performed using AI, for example, or without AI. For example, the check unit can input the results of a questionnaire survey into AI, which can then evaluate the discomfort.

[0093] The check unit can correct biases based on user feedback. Bias corrections include, but are not limited to, types such as gender bias and age bias. For example, the check unit corrects gender biases based on user feedback. The check unit can also correct age biases based on user feedback. For example, the check unit corrects gender biases in images generated by the generation AI based on user feedback. The check unit can also correct age biases in images generated by the generation AI based on user feedback. This allows for correcting biases based on user feedback to provide a fairer image model. Some or all of the above-described processing in the check unit may be performed using AI, for example, or may be performed without using AI. For example, the check unit can input user feedback into AI, which then corrects the biases.

[0094] The presentation unit can verify whether bias exists in the images generated by the generation AI and publish the results in the form of a report. Examples of report formats include, but are not limited to, text reports and graph reports. For example, the presentation unit publishes bias evidence as a text report. The presentation unit can also publish bias evidence as a graph report. For example, the presentation unit publishes a report that describes bias evidence in detail in text format. The presentation unit can also publish a report that visually displays bias evidence using graphs and charts. This makes it possible to verify the existence of bias and publish the results, thereby raising awareness of AI ethics issues. Some or all of the above-described processing in the presentation unit may be performed using AI, or may be performed without using AI. For example, the presentation unit can input bias evidence into AI, which then generates a report.

[0095] The presentation unit can provide information on issues related to AI ethics. Issues related to AI ethics include, but are not limited to, privacy protection and fairness. The presentation unit can provide, for example, information on privacy protection. The presentation unit can also provide information on fairness. For example, the presentation unit can provide guidelines on privacy protection for data collected by AI. The presentation unit can also provide information on fairness in images generated by AI. This can raise awareness of issues related to AI ethics and gain the trust of society as a whole. Some or all of the above-described processing in the presentation unit can be performed using AI, or can be performed without using AI. For example, the presentation unit can input information related to AI ethics into AI, which can then organize and provide the information.

[0096] The collection unit can estimate the user's emotions and adjust the timing of data collection based on the estimated emotions. Estimation of the user's emotions includes, but is not limited to, methods such as facial expression recognition and voice analysis. For example, the collection unit starts data collection when the user is relaxed and pauses collection when the user is stressed. The collection unit can also increase the frequency of data collection and collect more detailed data when the user is excited. For example, the collection unit increases the frequency of data collection and collects more detailed data when the user is excited. The collection unit can also refrain from data collection when the user is tired and resume it after the user has rested. For example, the collection unit refrains from data collection when the user is tired and resumes it after the user has rested. This allows for more appropriate data to be collected by adjusting the timing of data collection according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the collection unit may be performed using, for example, AI, or may be performed without using AI. For example, the collection unit may input the user's facial expression data into a generation AI, which may infer the emotion, and adjust the timing of data collection based on the result.

[0097] The collection unit can collect data based on the culture and social background of each region in Japan and construct a region-specific image model. Region-specific data includes, but is not limited to, events and ceremonies, architectural styles, and food cultures unique to each region. For example, the collection unit collects data on events and ceremonies unique to each region, such as the Hokkaido Snow Festival and Okinawa's Eisa. The collection unit can also collect data on architectural styles unique to each region, such as traditional townhouses in Kyoto and modern buildings in Tokyo. For example, the collection unit collects data on regional food cultures and traditional cuisines and reflects them in the image model. In this way, by collecting data unique to each region, a region-specific image model can be constructed. Some or all of the above-described processing by the collection unit may be performed using, for example, AI, or may be performed without AI. For example, the collection unit can input data on regional events and ceremonies into AI, which then automatically collects related data.

[0098] The collection unit can filter data according to seasonal events and occasions when collecting data. Examples of seasonal events and occasions include, but are not limited to, cherry blossom viewing and entrance ceremonies in spring, fireworks festivals and beach trips in summer, and Christmas and New Year's in winter. For example, the collection unit collects data on cherry blossom viewing and entrance ceremonies in spring and reflects the data in the image model. The collection unit can also collect data on fireworks festivals and beach trips in summer and reflect the data in the image model. For example, the collection unit collects data on Christmas and New Year's in winter and reflects the data in the image model. This allows for filtering data based on seasonal events and occasions, thereby collecting a wider variety of data. Some or all of the above-described processing in the collection unit may be performed using, or without, AI. For example, the collection unit can input data on seasonal events and occasions into AI, and the AI ​​can automatically filter related data.

[0099] When collecting data, the collection unit can customize the collection method based on the user's past feedback. Customizing the collection method includes, but is not limited to, for example, how to reflect the feedback and adjusting the collection process. For example, the collection unit prioritizes a specific data collection method based on feedback provided by the user in the past. The collection unit can also avoid a data collection method that the user has expressed discomfort with and adopt an alternative method. For example, the collection unit adjusts the frequency and timing of data collection based on the user's feedback. This allows for customizing the collection method to reflect the user's past feedback, thereby collecting more appropriate data. Some or all of the above-described processing by the collection unit may be performed using, or without, AI. For example, the collection unit can input the user's past feedback into AI, which then customizes the collection method.

[0100] The collection unit can estimate the user's emotions and determine the priority of data to be collected based on the emotions. Data prioritization can be based on criteria such as, but not limited to, importance scores and user interest levels. For example, when the user is relaxed, the collection unit can prioritize collecting detailed data. Furthermore, when the user is stressed, the collection unit can prioritize collecting only important data. For example, when the user is excited, the collection unit can prioritize collecting interesting data. This allows for data prioritization based on the user's emotions, resulting in the prioritized collection of more important data. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the collection unit can be performed using, for example, AI, or without AI. For example, the collection unit can input the user's emotion data into a generation AI, which can then prioritize the data.

[0101] When collecting data, the collection unit can prioritize collecting highly relevant data by taking into account the user's geographical location information. Examples of uses of geographical location information include, but are not limited to, GPS data and location information services. For example, if the user is in Tokyo, the collection unit can prioritize collecting data related to Tokyo. Furthermore, if the user is in Kyoto, the collection unit can prioritize collecting data related to Kyoto's traditional culture. For example, if the user is in Okinawa, the collection unit can prioritize collecting data related to Okinawa's natural landscapes. This allows for more relevant data to be collected by taking the user's geographical location information into account. Some or all of the above-described processing in the collection unit may be performed using, or without, AI. For example, the collection unit can input the user's geographical location information into AI, which can then automatically select highly relevant data.

[0102] The collection unit may analyze the user's social media activity during data collection and collect relevant data based on the activity. Examples of analyzing social media activity include, but are not limited to, analyzing posts and followers. The collection unit may collect relevant data based on, for example, images and videos shared by the user on social media. The collection unit may also analyze the user's social media posts to collect data that may be of interest to the user. For example, the collection unit may collect relevant data based on the activities of the user's friends on social media. This allows for the collection of more relevant data by analyzing the user's social media activity. Some or all of the above-described processing by the collection unit may be performed using, for example, AI, or may be performed without using AI. For example, the collection unit may input the user's social media activity data into AI, which then automatically collects the relevant data.

[0103] The construction unit can estimate the user's emotions and adjust the image model construction method based on the user's emotions. Adjusting the image model construction method includes, but is not limited to, selecting an algorithm and adjusting parameters. For example, when the user is relaxed, the construction unit can construct a detailed image model. Furthermore, when the user is in a hurry, the construction unit can construct a simplified image model. For example, when the user is excited, the construction unit can construct an image model including visually stimulating elements. This allows for the construction of a more appropriate image model by adjusting the image model construction method according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the construction unit can be performed using, for example, an AI, or without an AI. For example, the construction unit can input the user's emotion data into the generation AI, which can then adjust the image model construction method.

[0104] When constructing an image model, the construction unit can adjust the level of detail of the model according to the importance of the data. Evaluation of the importance of the data includes, but is not limited to, criteria such as the reliability and relevance of the data. For example, the construction unit constructs a detailed image model based on data with high importance. The construction unit can also construct a simplified image model based on data with low importance. For example, the construction unit gradually adjusts the level of detail of the model according to the importance of the data. This allows an image model that reflects more important data to be constructed by adjusting the level of detail of the model based on the importance of the data. Some or all of the above-described processing in the construction unit may be performed using, or without, AI. For example, the construction unit can input the importance of the data into AI, which then adjusts the level of detail of the model.

[0105] When constructing an image model, the construction unit can apply different construction algorithms based on the category of data. Examples of data category classification include, but are not limited to, image category and text category. For example, the construction unit can apply a specific algorithm to landscape images to construct a detailed landscape model. The construction unit can also apply a different algorithm to portrait images to construct a detailed portrait model. For example, the construction unit selects an optimal algorithm for each category and constructs an image model. This allows for the construction of a more accurate image model by applying the optimal algorithm depending on the data category. Some or all of the above-described processing in the construction unit may be performed using, for example, AI, or may be performed without using AI. For example, the construction unit can input the data category into AI, which selects the optimal algorithm to construct an image model.

[0106] When constructing an image model, the construction unit can improve the accuracy of the model based on the user's past construction results. Use of past construction results includes, but is not limited to, the accuracy of past models and user feedback. For example, the construction unit improves the accuracy of the current model based on image models previously constructed by the user. The construction unit can also adjust the accuracy of the model by referring to the user's past feedback. For example, the construction unit analyzes the user's past construction results and adopts an optimal model construction method. This allows the accuracy of the model to be improved by referring to the user's past construction results. Some or all of the above-described processing in the construction unit may be performed using, for example, AI, or may be performed without using AI. For example, the construction unit can input the user's past construction results into AI, which can then improve the accuracy of the model.

[0107] The construction unit can estimate the user's emotion and adjust the length of the image model based on the emotion. Adjustment of the length of the image model can include, but is not limited to, the number of model layers and the number of parameters. For example, if the user is in a hurry, the construction unit can build a short image model. Also, if the user is relaxed, the construction unit can build a detailed, long image model. For example, if the user is excited, the construction unit can build an image model that includes visually stimulating elements. This allows for the construction of a more appropriate image model by adjusting the length of the image model according to the user's emotion. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the construction unit can be performed using, for example, AI, or without AI. For example, the construction unit can input the user's emotion data into the generation AI, which can then adjust the length of the image model.

[0108] When constructing an image model, the construction unit can determine the priority of models based on the time of data submission. Use of the time of data submission includes, but is not limited to, the priority of the latest data and the frequency of submission. For example, the construction unit constructs an image model by preferentially using the latest data. The construction unit can also use older data by lowering the priority. For example, the construction unit evaluates the importance of data based on the time of submission and determines the priority of models. By determining the priority of models based on the time of data submission, an image model that reflects the latest data preferentially can be constructed. Some or all of the above-described processing in the construction unit may be performed using, for example, AI, or may be performed without using AI. For example, the construction unit can input the time of data submission into AI, which then determines the priority of models.

[0109] When constructing an image model, the construction unit can adjust the order of models based on the relevance of the data. Evaluation of the relevance of the data includes, but is not limited to, criteria such as correlation analysis and relevance scores. For example, the construction unit constructs an image model by preferentially using highly relevant data. The construction unit can also postpone the use of less relevant data. For example, the construction unit adjusts the construction order of models based on the relevance of the data. In this way, by adjusting the order of models based on the relevance of the data, an image model that preferentially reflects more relevant data can be constructed. Some or all of the above-described processing in the construction unit may be performed using, for example, AI, or may be performed without using AI. For example, the construction unit can input the relevance of the data into AI, which can then adjust the order of the models.

[0110] When constructing an image model, the construction unit can adjust the use of technical terminology in the model based on the user's level of expertise. Evaluation of the level of expertise includes, but is not limited to, criteria such as qualifications and past experience. For example, the construction unit can construct an image model using detailed technical terminology for a user with high expertise. The construction unit can also construct an image model using simpler terminology for a user with low expertise. For example, the construction unit gradually adjusts the use of technical terminology in the model according to the user's level of expertise. This allows for the provision of an image model that is easier to understand by adjusting the use of technical terminology according to the user's level of expertise. Some or all of the above-described processing in the construction unit can be performed using, for example, AI, or without AI. For example, the construction unit can input the user's level of expertise into AI, which can then adjust the use of technical terminology.

[0111] The check unit can estimate the user's emotions and adjust the bias check criteria based on the emotions. Adjusting the bias check criteria includes, but is not limited to, setting an evaluation score or a reference value. For example, if the user is relaxed, the check unit can perform a detailed bias check. Furthermore, if the user is in a hurry, the check unit can also perform a simplified bias check. For example, if the user is excited, the check unit can perform a bias check that includes visually stimulating elements. This allows for adjusting the bias check criteria according to the user's emotions, thereby enabling a more appropriate bias check. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the check unit can be performed using, for example, an AI, or without an AI. For example, the check unit can input the user's emotion data into the generation AI, which can then adjust the bias check criteria.

[0112] The check unit can improve the accuracy of the bias check based on the interrelationships between images during the bias check. Evaluation of the interrelationships between images can be based on, but not limited to, criteria such as co-occurrence networks and correlation analysis. For example, the check unit can analyze the interrelationships between images to accurately check for the presence or absence of bias. The check unit can also adjust the criteria for the bias check by taking the interrelationships between images into account. For example, the check unit can evaluate the results of the bias check based on the interrelationships between images. This allows the accuracy of the bias check to be improved by taking the interrelationships between images into account. Some or all of the above-described processing in the check unit can be performed using, or without, AI. For example, the check unit can input image interrelationship data into AI, which can then improve the accuracy of the bias check.

[0113] When performing a bias check, the check unit can perform a check based on the attribute information of the image submitter. Use of the submitter's attribute information includes, but is not limited to, age, gender, and occupation. The check unit can, for example, perform a bias check taking into account the age and gender of the image submitter. The check unit can also perform a bias check taking into account the occupation and background of the image submitter. For example, the check unit adjusts the bias check criteria based on the attribute information of the image submitter. This allows for a more accurate bias check by taking into account the attribute information of the image submitter. Some or all of the above-described processing in the check unit can be performed using, for example, AI, or without AI. For example, the check unit can input the submitter's attribute information into AI, which can then perform the bias check.

[0114] During bias checks, the check unit can weight checks based on the frequency of image submission. Evaluation of submission frequency includes, but is not limited to, criteria such as the number of submissions and the interval between submissions. For example, the check unit can focus bias checks on images submitted frequently. The check unit can also perform lighter bias checks on images submitted infrequently. For example, the check unit can adjust the weighting of the bias check based on the submission frequency. This allows for more efficient bias checks by weighting the checks based on the frequency of image submission. Some or all of the above-described processing in the check unit can be performed using, for example, AI, or without AI. For example, the check unit can input submission frequency data into AI, which can then weight the bias check.

[0115] The check unit can estimate the user's emotions and adjust the order in which the bias check results are displayed based on the user's emotions. Examples of adjustments to the order in which the results are displayed include, but are not limited to, criteria such as order of importance or chronological order. For example, if the user is relaxed, the check unit can prioritize displaying detailed bias check results. Furthermore, if the user is in a hurry, the check unit can prioritize displaying important bias check results. For example, if the user is excited, the check unit can prioritize displaying visually stimulating bias check results. This allows for more appropriate information to be provided by adjusting the order in which the bias check results are displayed according to the user's emotions. The emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the check unit can be performed using, for example, an AI, or without an AI. For example, the check unit can input the user's emotion data into the generation AI and adjust the order in which the generation AI displays the bias check results.

[0116] When performing a bias check, the check unit can perform a check based on the geographic distribution of the images. Evaluation of the geographic distribution can include, but is not limited to, criteria such as data distribution by region and geographic clusters. For example, the check unit can analyze the geographic distribution of the images and check for bias toward a specific region. The check unit can also adjust the criteria for the bias check based on the geographic distribution. For example, the check unit can evaluate the results of the bias check taking into account the geographic distribution. This makes it possible to check for bias toward a specific region by taking into account the geographic distribution of the images. Some or all of the above-described processing in the check unit can be performed using, or without, AI. For example, the check unit can input geographic distribution data into AI, which can then perform the bias check.

[0117] During the bias check, the check unit can improve the accuracy of the check by referring to related literature of the image. Reference to related literature includes, but is not limited to, criteria such as a literature database and the number of citations. For example, the check unit can refer to literature related to the image to accurately check for the presence or absence of bias. The check unit can also adjust the criteria for the bias check based on the related literature. For example, the check unit can evaluate the results of the bias check by referring to the related literature. Thus, by referring to the related literature of the image, the accuracy of the bias check can be improved. Some or all of the above-described processing in the check unit can be performed using, for example, AI, or can be performed without using AI. For example, the check unit can input related literature data into AI, which can improve the accuracy of the bias check.

[0118] When performing a bias check, the check unit can perform a check based on the market value of the image. Evaluation of market value includes, but is not limited to, criteria such as sales data and evaluation scores. For example, the check unit can focus the bias check on images with high market value. The check unit can also perform a lighter bias check on images with low market value. For example, the check unit can adjust the weighting of the bias check based on market value. This allows the bias check to be focused on more important images by taking the market value of the image into consideration. Some or all of the above-described processing in the check unit can be performed using, or without, AI. For example, the check unit can input market value data into AI, which then performs the bias check.

[0119] The presentation unit can estimate the user's emotions and adjust the presentation method of bias evidence based on the emotions. Examples of adjustment of the presentation method of bias evidence include, but are not limited to, text format, graph format, and the like. For example, if the user is relaxed, the presentation unit can present detailed bias evidence. Furthermore, if the user is in a hurry, the presentation unit can present simplified bias evidence. For example, if the user is excited, the presentation unit can present visually stimulating bias evidence. This allows for more appropriate information to be provided by adjusting the presentation method of bias evidence according to the user's emotions. The emotion estimation is realized using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, for example, a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to these examples. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or without AI. For example, the presentation unit can input the user's emotion data into the generation AI, which can then adjust the presentation method of bias evidence.

[0120] When presenting bias evidence, the presentation unit can optimize current evidence based on past evidence data. Use of past evidence data includes, but is not limited to, criteria such as past evaluation results and feedback. The presentation unit, for example, optimizes current bias evidence based on past evidence data. The presentation unit can also adjust the presentation method of bias evidence by referring to past evidence data. For example, the presentation unit analyzes past evidence data and presents optimal bias evidence. This allows current bias evidence to be optimized by referring to the past evidence data. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input past evidence data into AI, which then optimizes current evidence.

[0121] When presenting bias evidence, the presentation unit can improve the method of presenting the evidence based on user feedback. Improvements to the method of presenting the evidence include, but are not limited to, criteria such as a method of reflecting user feedback and adjusting the presentation format. For example, the presentation unit improves the method of presenting the bias evidence based on user feedback. The presentation unit can also adjust the level of detail of the evidence by reflecting user feedback. For example, the presentation unit improves the visual presentation method of the evidence by referring to user feedback. In this way, the presentation method of the evidence can be improved by reflecting user feedback. Some or all of the above-mentioned processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input user feedback into AI, which can improve the method of presenting the evidence.

[0122] When presenting bias evidence, the presentation unit can customize the method of presenting the evidence based on the user's past feedback. Customization of the method of presenting the evidence includes, but is not limited to, criteria such as a method of reflecting the user's past feedback and adjusting the presentation format. For example, the presentation unit customizes the method of presenting the evidence based on the user's past feedback. The presentation unit can also adjust the level of detail of the evidence by reflecting the user's past feedback. For example, the presentation unit customizes the visual presentation method of the evidence by referring to the user's past feedback. In this way, the method of presenting the evidence can be customized by reflecting the user's past feedback. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input the user's past feedback into AI, which can then customize the method of presenting the evidence.

[0123] The presentation unit can estimate the user's emotions and prioritize bias evidence based on the emotions. The prioritization of bias evidence can be based on criteria such as, but not limited to, importance scores and user interest levels. For example, if the user is relaxed, the presentation unit can prioritize detailed bias evidence. Furthermore, if the user is in a hurry, the presentation unit can prioritize important bias evidence. For example, if the user is excited, the presentation unit can prioritize visually stimulating bias evidence. This prioritizes bias evidence based on the user's emotions, allowing for more important information to be provided preferentially. The emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the presentation unit can be performed using, for example, an AI. For example, the presentation unit can input the user's emotion data into the generation AI, which can then prioritize the bias evidence.

[0124] When presenting bias evidence, the presentation unit can select an optimal evidence presentation method based on the user's geographical location information. Use of geographical location information includes, but is not limited to, GPS data, location information services, etc. For example, if the user is in Tokyo, the presentation unit can prioritize presenting bias evidence related to Tokyo. Furthermore, if the user is in Kyoto, the presentation unit can also prioritize presenting bias evidence related to Kyoto. For example, if the user is in Okinawa, the presentation unit can prioritize presenting bias evidence related to Okinawa. This allows for providing more relevant evidence by taking the user's geographical location information into consideration. Some or all of the above-described processing in the presentation unit may be performed using, or without, AI. For example, the presentation unit can input the user's geographical location information into AI, which can then select an optimal evidence presentation method.

[0125] When presenting bias evidence, the presentation unit can analyze the user's social media activity and suggest a method for presenting the evidence based on the activity. Analysis of social media activity includes, but is not limited to, criteria such as analysis of post content and analysis of followers. The presentation unit can present relevant bias evidence, for example, based on images and videos shared by the user on social media. The presentation unit can also analyze the user's social media posts to present bias evidence that may be of interest. For example, the presentation unit can present relevant bias evidence based on the activity of the user's friends on social media. This allows for analysis of the user's social media activity to provide more relevant evidence. Some or all of the above-described processing by the presentation unit can be performed using, for example, AI, or without AI. For example, the presentation unit can input the user's social media activity data into AI, which can then suggest a method for presenting evidence.

[0126] When presenting bias evidence, the presentation unit can customize the method of presenting the evidence based on the user's past feedback. Customization of the method of presenting the evidence includes, but is not limited to, criteria such as a method of reflecting the user's past feedback and adjusting the presentation format. For example, the presentation unit customizes the method of presenting the evidence based on the user's past feedback. The presentation unit can also adjust the level of detail of the evidence by reflecting the user's past feedback. For example, the presentation unit customizes the visual presentation method of the evidence by referring to the user's past feedback. In this way, the method of presenting the evidence can be customized by reflecting the user's past feedback. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input the user's past feedback into AI, which can then customize the method of presenting the evidence.

[0127] When performing a bias check, the check unit can perform a check based on the market value of the image. Evaluation of market value includes, but is not limited to, criteria such as sales data and evaluation scores. For example, the check unit can focus the bias check on images with high market value. The check unit can also perform a lighter bias check on images with low market value. For example, the check unit can adjust the weighting of the bias check based on market value. This allows the bias check to be focused on more important images by taking the market value of the image into consideration. Some or all of the above-described processing in the check unit can be performed using, or without, AI. For example, the check unit can input market value data into AI, which then performs the bias check.

[0128] The presentation unit can estimate the user's emotions and adjust the presentation method of bias evidence based on the emotions. Examples of adjustment of the presentation method of bias evidence include, but are not limited to, text format, graph format, and the like. For example, if the user is relaxed, the presentation unit can present detailed bias evidence. Furthermore, if the user is in a hurry, the presentation unit can present simplified bias evidence. For example, if the user is excited, the presentation unit can present visually stimulating bias evidence. This allows for more appropriate information to be provided by adjusting the presentation method of bias evidence according to the user's emotions. The emotion estimation is realized using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, for example, a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to these examples. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or without AI. For example, the presentation unit can input the user's emotion data into the generation AI, which can then adjust the presentation method of bias evidence.

[0129] When presenting bias evidence, the presentation unit can optimize current evidence based on past evidence data. Use of past evidence data includes, but is not limited to, criteria such as past evaluation results and feedback. The presentation unit, for example, optimizes current bias evidence based on past evidence data. The presentation unit can also adjust the presentation method of bias evidence by referring to past evidence data. For example, the presentation unit analyzes past evidence data and presents optimal bias evidence. This allows current bias evidence to be optimized by referring to the past evidence data. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input past evidence data into AI, which then optimizes current evidence.

[0130] When presenting bias evidence, the presentation unit can improve the method of presenting the evidence based on user feedback. Improvements to the method of presenting the evidence include, but are not limited to, criteria such as a method of reflecting user feedback and adjusting the presentation format. For example, the presentation unit improves the method of presenting the bias evidence based on user feedback. The presentation unit can also adjust the level of detail of the evidence by reflecting user feedback. For example, the presentation unit improves the visual presentation method of the evidence by referring to user feedback. In this way, the presentation method of the evidence can be improved by reflecting user feedback. Some or all of the above-mentioned processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input user feedback into AI, which can improve the method of presenting the evidence.

[0131] When presenting bias evidence, the presentation unit can customize the method of presenting the evidence based on the user's past feedback. Customization of the method of presenting the evidence includes, but is not limited to, criteria such as a method of reflecting the user's past feedback and adjusting the presentation format. For example, the presentation unit customizes the method of presenting the evidence based on the user's past feedback. The presentation unit can also adjust the level of detail of the evidence by reflecting the user's past feedback. For example, the presentation unit customizes the visual presentation method of the evidence by referring to the user's past feedback. In this way, the method of presenting the evidence can be customized by reflecting the user's past feedback. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input the user's past feedback into AI, which can then customize the method of presenting the evidence.

[0132] The presentation unit can estimate the user's emotions and prioritize bias evidence based on the emotions. The prioritization of bias evidence can be based on criteria such as, but not limited to, importance scores and user interest levels. For example, if the user is relaxed, the presentation unit can prioritize detailed bias evidence. Furthermore, if the user is in a hurry, the presentation unit can prioritize important bias evidence. For example, if the user is excited, the presentation unit can prioritize visually stimulating bias evidence. This prioritizes bias evidence based on the user's emotions, allowing for more important information to be provided preferentially. The emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the presentation unit can be performed using, for example, an AI. For example, the presentation unit can input the user's emotion data into the generation AI, which can then prioritize the bias evidence.

[0133] When presenting bias evidence, the presentation unit can select an optimal evidence presentation method based on the user's geographical location information. Use of geographical location information includes, but is not limited to, GPS data, location information services, etc. For example, if the user is in Tokyo, the presentation unit can prioritize presenting bias evidence related to Tokyo. Furthermore, if the user is in Kyoto, the presentation unit can also prioritize presenting bias evidence related to Kyoto. For example, if the user is in Okinawa, the presentation unit can prioritize presenting bias evidence related to Okinawa. This allows for providing more relevant evidence by taking the user's geographical location information into consideration. Some or all of the above-described processing in the presentation unit may be performed using, or without, AI. For example, the presentation unit can input the user's geographical location information into AI, which can then select an optimal evidence presentation method.

[0134] When presenting bias evidence, the presentation unit can analyze the user's social media activity and suggest a method for presenting the evidence based on the activity. Analysis of social media activity includes, but is not limited to, criteria such as analysis of post content and analysis of followers. The presentation unit can present relevant bias evidence, for example, based on images and videos shared by the user on social media. The presentation unit can also analyze the user's social media posts to present bias evidence that may be of interest. For example, the presentation unit can present relevant bias evidence based on the activity of the user's friends on social media. This allows for analysis of the user's social media activity to provide more relevant evidence. Some or all of the above-described processing by the presentation unit can be performed using, for example, AI, or without AI. For example, the presentation unit can input the user's social media activity data into AI, which can then suggest a method for presenting evidence.

[0135] When presenting bias evidence, the presentation unit can customize the method of presenting the evidence by reflecting the user's past feedback. Customization of the method of presenting the evidence includes, but is not limited to, criteria such as a method of reflecting the user's past feedback and adjusting the presentation format. For example, the presentation unit customizes the method of presenting the evidence based on the user's past feedback. The presentation unit can also adjust the level of detail of the evidence by reflecting the user's past feedback. For example, the presentation unit customizes the visual presentation method of the evidence by referring to the user's past feedback. In this way, the method of presenting the evidence can be customized by reflecting the user's past feedback. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input the user's past feedback into AI, which can then customize the method of presenting the evidence. === Hard Collateral 1-1 === Each of the multiple elements, including the collection unit, construction unit, check unit, and presentation unit, described above, is realized, for example, by at least one of the smart device 14 and the data processing device 12. For example, the collection unit collects data specific to Japan using the camera 42 and communication I / F 44 of the smart device 14. The construction unit is realized, for example, by the specific processing unit 290 of the data processing device 12 and constructs an image model specific to Japan based on the collected data. The check unit is realized, for example, by the control unit 46A of the smart device 14 and performs a bias check based on discomfort based on user feedback. The presentation unit is realized, for example, by the specific processing unit 290 of the data processing device 12 and publishes bias evidence in report format. === Hard Collateral 1-2 === Each of the multiple elements, including the collection unit, construction unit, check unit, and presentation unit, described above, is realized, for example, by at least one of the smart glasses 214 and the data processing device 12. For example, the collection unit collects data specific to Japan using the camera 42 and communication I / F 44 of the smart glasses 214. The construction unit is realized, for example, by the specific processing unit 290 of the data processing device 12 and constructs an image model specific to Japan based on the collected data. The check unit is realized, for example, by the control unit 46A of the smart glasses 214 and performs a bias check based on discomfort based on user feedback. The presentation unit is realized, for example, by the specific processing unit 290 of the data processing device 12 and publishes bias evidence in report format. === Hard Collateral 1-3 === Each of the multiple elements including the collection unit, construction unit, check unit, and presentation unit described above is realized, for example, by at least one of the headset type terminal 314 and the data processing device 12. For example, the collection unit collects data specialized for Japan using the camera 42 and communication I / F 44 of the headset type terminal 314. The construction unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and constructs an image model specialized for Japan based on the collected data. The check unit is realized, for example, by the control unit 46A of the headset type terminal 314, and performs a bias check based on discomfort on the basis of user feedback. The presentation unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and publishes bias evidence in the form of a report. === Hard Collateral 1-4 === Each of the multiple elements including the collection unit, construction unit, check unit, and presentation unit described above is realized, for example, by at least one of the robot 414 and the data processing device 12. For example, the collection unit collects data specific to Japan using the camera 42 and communication I / F 44 of the robot 414. The construction unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and constructs an image model specific to Japan based on the collected data. The check unit is realized, for example, by the control unit 46A of the robot 414, and performs a bias check based on discomfort based on user feedback. The presentation unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and publishes bias evidence in the form of a report.

[0136] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0137] The collection unit can estimate the user's emotions and adjust the timing of data collection based on the emotions. For example, if the user is relaxed, data collection can be started, and if the user is stressed, data collection can be paused. The collection unit can also increase the frequency of data collection and collect more detailed data if the user is excited. For example, if the user is excited, the collection unit can increase the frequency of data collection and collect more detailed data. The collection unit can also refrain from data collection if the user is tired and resume it after the user has rested. For example, if the user is tired, the collection unit can refrain from data collection and resume it after the user has rested. This allows for more appropriate data to be collected by adjusting the timing of data collection according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the collection unit can be performed using, for example, an AI, or without an AI. For example, the collection unit can input the user's facial expression data into the generation AI, which can then estimate the emotion and adjust the timing of data collection based on the results.

[0138] The collection unit can collect data based on the culture and social background of each region in Japan and construct a region-specific image model. Region-specific data includes, but is not limited to, events and ceremonies, architectural styles, and food cultures unique to each region. For example, the collection unit collects data on events and ceremonies unique to each region, such as the Hokkaido Snow Festival and Okinawa's Eisa. The collection unit can also collect data on architectural styles unique to each region, such as traditional townhouses in Kyoto and modern buildings in Tokyo. For example, the collection unit collects data on regional food cultures and traditional cuisines and reflects them in the image model. In this way, by collecting data unique to each region, a region-specific image model can be constructed. Some or all of the above-described processing by the collection unit may be performed using, for example, AI, or may be performed without AI. For example, the collection unit can input data on regional events and ceremonies into AI, which then automatically collects related data.

[0139] The collection unit can filter data according to seasonal events and occasions when collecting data. Examples of seasonal events and occasions include, but are not limited to, cherry blossom viewing and entrance ceremonies in spring, fireworks festivals and beach trips in summer, and Christmas and New Year's in winter. For example, the collection unit collects data on cherry blossom viewing and entrance ceremonies in spring and reflects the data in the image model. The collection unit can also collect data on fireworks festivals and beach trips in summer and reflect the data in the image model. For example, the collection unit collects data on Christmas and New Year's in winter and reflects the data in the image model. This allows for filtering data based on seasonal events and occasions, thereby collecting a wider variety of data. Some or all of the above-described processing in the collection unit may be performed using, or without, AI. For example, the collection unit can input data on seasonal events and occasions into AI, and the AI ​​can automatically filter related data.

[0140] When collecting data, the collection unit can customize the collection method based on the user's past feedback. Customizing the collection method includes, but is not limited to, for example, how to reflect the feedback and adjusting the collection process. For example, the collection unit prioritizes a specific data collection method based on feedback provided by the user in the past. The collection unit can also avoid a data collection method that the user has expressed discomfort with and adopt an alternative method. For example, the collection unit adjusts the frequency and timing of data collection based on the user's feedback. This allows for customizing the collection method to reflect the user's past feedback, thereby collecting more appropriate data. Some or all of the above-described processing by the collection unit may be performed using, or without, AI. For example, the collection unit can input the user's past feedback into AI, which then customizes the collection method.

[0141] The collection unit can estimate the user's emotions and determine the priority of data to be collected based on the emotions. Data prioritization can be based on criteria such as, but not limited to, importance scores and user interest levels. For example, when the user is relaxed, the collection unit can prioritize collecting detailed data. Furthermore, when the user is stressed, the collection unit can prioritize collecting only important data. For example, when the user is excited, the collection unit can prioritize collecting interesting data. This allows for data prioritization based on the user's emotions, resulting in the prioritized collection of more important data. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the collection unit can be performed using, for example, AI, or without AI. For example, the collection unit can input the user's emotion data into a generation AI, which can then prioritize the data.

[0142] When collecting data, the collection unit can prioritize collecting highly relevant data by taking into account the user's geographical location information. Examples of uses of geographical location information include, but are not limited to, GPS data and location information services. For example, if the user is in Tokyo, the collection unit can prioritize collecting data related to Tokyo. Furthermore, if the user is in Kyoto, the collection unit can prioritize collecting data related to Kyoto's traditional culture. For example, if the user is in Okinawa, the collection unit can prioritize collecting data related to Okinawa's natural landscapes. This allows for more relevant data to be collected by taking the user's geographical location information into account. Some or all of the above-described processing in the collection unit may be performed using, or without, AI. For example, the collection unit can input the user's geographical location information into AI, which can then automatically select highly relevant data.

[0143] The construction unit can estimate the user's emotions and adjust the image model construction method based on the user's emotions. Adjusting the image model construction method includes, but is not limited to, selecting an algorithm and adjusting parameters. For example, when the user is relaxed, the construction unit can construct a detailed image model. Furthermore, when the user is in a hurry, the construction unit can construct a simplified image model. For example, when the user is excited, the construction unit can construct an image model including visually stimulating elements. This allows for the construction of a more appropriate image model by adjusting the image model construction method according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the construction unit can be performed using, for example, an AI, or without an AI. For example, the construction unit can input the user's emotion data into the generation AI, which can then adjust the image model construction method.

[0144] When constructing an image model, the construction unit can adjust the level of detail of the model according to the importance of the data. Evaluation of the importance of the data includes, but is not limited to, criteria such as the reliability and relevance of the data. For example, the construction unit constructs a detailed image model based on data with high importance. The construction unit can also construct a simplified image model based on data with low importance. For example, the construction unit gradually adjusts the level of detail of the model according to the importance of the data. This allows an image model that reflects more important data to be constructed by adjusting the level of detail of the model based on the importance of the data. Some or all of the above-described processing in the construction unit may be performed using, or without, AI. For example, the construction unit can input the importance of the data into AI, which then adjusts the level of detail of the model.

[0145] When constructing an image model, the construction unit can apply different construction algorithms based on the category of data. Examples of data category classification include, but are not limited to, image category and text category. For example, the construction unit can apply a specific algorithm to landscape images to construct a detailed landscape model. The construction unit can also apply a different algorithm to portrait images to construct a detailed portrait model. For example, the construction unit selects an optimal algorithm for each category and constructs an image model. This allows for the construction of a more accurate image model by applying the optimal algorithm depending on the data category. Some or all of the above-described processing in the construction unit may be performed using, for example, AI, or may be performed without using AI. For example, the construction unit can input the data category into AI, which selects the optimal algorithm to construct an image model.

[0146] When constructing an image model, the construction unit can improve the accuracy of the model based on the user's past construction results. Use of past construction results includes, but is not limited to, the accuracy of past models and user feedback. For example, the construction unit improves the accuracy of the current model based on image models previously constructed by the user. The construction unit can also adjust the accuracy of the model by referring to the user's past feedback. For example, the construction unit analyzes the user's past construction results and adopts an optimal model construction method. This allows the accuracy of the model to be improved by referring to the user's past construction results. Some or all of the above-described processing in the construction unit may be performed using, for example, AI, or may be performed without using AI. For example, the construction unit can input the user's past construction results into AI, which can then improve the accuracy of the model.

[0147] The construction unit can estimate the user's emotion and adjust the length of the image model based on the emotion. Adjustment of the length of the image model can include, but is not limited to, the number of model layers and the number of parameters. For example, if the user is in a hurry, the construction unit can build a short image model. Also, if the user is relaxed, the construction unit can build a detailed, long image model. For example, if the user is excited, the construction unit can build an image model that includes visually stimulating elements. This allows for the construction of a more appropriate image model by adjusting the length of the image model according to the user's emotion. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the construction unit can be performed using, for example, AI, or without AI. For example, the construction unit can input the user's emotion data into the generation AI, which can then adjust the length of the image model.

[0148] When constructing an image model, the construction unit can determine the priority of models based on the time of data submission. Use of the time of data submission includes, but is not limited to, the priority of the latest data and the frequency of submission. For example, the construction unit constructs an image model by preferentially using the latest data. The construction unit can also use older data by lowering the priority. For example, the construction unit evaluates the importance of data based on the time of submission and determines the priority of models. By determining the priority of models based on the time of data submission, an image model that reflects the latest data preferentially can be constructed. Some or all of the above-described processing in the construction unit may be performed using, for example, AI, or may be performed without using AI. For example, the construction unit can input the time of data submission into AI, which then determines the priority of models.

[0149] When constructing an image model, the construction unit can adjust the order of models based on the relevance of the data. Evaluation of the relevance of the data includes, but is not limited to, criteria such as correlation analysis and relevance scores. For example, the construction unit constructs an image model by preferentially using highly relevant data. The construction unit can also postpone the use of less relevant data. For example, the construction unit adjusts the construction order of models based on the relevance of the data. In this way, by adjusting the order of models based on the relevance of the data, an image model that preferentially reflects more relevant data can be constructed. Some or all of the above-described processing in the construction unit may be performed using, for example, AI, or may be performed without using AI. For example, the construction unit can input the relevance of the data into AI, which can then adjust the order of the models.

[0150] When constructing an image model, the construction unit can adjust the use of technical terminology in the model based on the user's level of expertise. Evaluation of the level of expertise includes, but is not limited to, criteria such as qualifications and past experience. For example, the construction unit can construct an image model using detailed technical terminology for a user with high expertise. The construction unit can also construct an image model using simpler terminology for a user with low expertise. For example, the construction unit gradually adjusts the use of technical terminology in the model according to the user's level of expertise. This allows for the provision of an image model that is easier to understand by adjusting the use of technical terminology according to the user's level of expertise. Some or all of the above-described processing in the construction unit can be performed using, for example, AI, or without AI. For example, the construction unit can input the user's level of expertise into AI, which can then adjust the use of technical terminology.

[0151] The check unit can estimate the user's emotions and adjust the bias check criteria based on the emotions. Adjusting the bias check criteria includes, but is not limited to, setting an evaluation score or a reference value. For example, if the user is relaxed, the check unit can perform a detailed bias check. Furthermore, if the user is in a hurry, the check unit can also perform a simplified bias check. For example, if the user is excited, the check unit can perform a bias check that includes visually stimulating elements. This allows for adjusting the bias check criteria according to the user's emotions, thereby enabling a more appropriate bias check. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the check unit can be performed using, for example, an AI, or without an AI. For example, the check unit can input the user's emotion data into the generation AI, which can then adjust the bias check criteria.

[0152] The check unit can improve the accuracy of the bias check based on the interrelationships between images during the bias check. Evaluation of the interrelationships between images can be based on, but not limited to, criteria such as co-occurrence networks and correlation analysis. For example, the check unit can analyze the interrelationships between images to accurately check for the presence or absence of bias. The check unit can also adjust the criteria for the bias check by taking the interrelationships between images into account. For example, the check unit can evaluate the results of the bias check based on the interrelationships between images. This allows the accuracy of the bias check to be improved by taking the interrelationships between images into account. Some or all of the above-described processing in the check unit can be performed using, or without, AI. For example, the check unit can input image interrelationship data into AI, which can then improve the accuracy of the bias check.

[0153] When performing a bias check, the check unit can perform a check based on the attribute information of the image submitter. Use of the submitter's attribute information includes, but is not limited to, age, gender, and occupation. The check unit can, for example, perform a bias check taking into account the age and gender of the image submitter. The check unit can also perform a bias check taking into account the occupation and background of the image submitter. For example, the check unit adjusts the bias check criteria based on the attribute information of the image submitter. This allows for a more accurate bias check by taking into account the attribute information of the image submitter. Some or all of the above-described processing in the check unit can be performed using, for example, AI, or without AI. For example, the check unit can input the submitter's attribute information into AI, which can then perform the bias check.

[0154] During bias checks, the check unit can weight checks based on the frequency of image submission. Evaluation of submission frequency includes, but is not limited to, criteria such as the number of submissions and the interval between submissions. For example, the check unit can focus bias checks on images submitted frequently. The check unit can also perform lighter bias checks on images submitted infrequently. For example, the check unit can adjust the weighting of the bias check based on the submission frequency. This allows for more efficient bias checks by weighting the checks based on the frequency of image submission. Some or all of the above-described processing in the check unit can be performed using, for example, AI, or without AI. For example, the check unit can input submission frequency data into AI, which can then weight the bias check.

[0155] The check unit can estimate the user's emotions and adjust the order in which the bias check results are displayed based on the user's emotions. Examples of adjustments to the order in which the results are displayed include, but are not limited to, criteria such as order of importance or chronological order. For example, if the user is relaxed, the check unit can prioritize displaying detailed bias check results. Furthermore, if the user is in a hurry, the check unit can prioritize displaying important bias check results. For example, if the user is excited, the check unit can prioritize displaying visually stimulating bias check results. This allows for more appropriate information to be provided by adjusting the order in which the bias check results are displayed according to the user's emotions. The emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the check unit can be performed using, for example, an AI, or without an AI. For example, the check unit can input the user's emotion data into the generation AI and adjust the order in which the generation AI displays the bias check results.

[0156] When performing a bias check, the check unit can perform a check based on the geographic distribution of the images. Evaluation of the geographic distribution can include, but is not limited to, criteria such as data distribution by region and geographic clusters. For example, the check unit can analyze the geographic distribution of the images and check for bias toward a specific region. The check unit can also adjust the criteria for the bias check based on the geographic distribution. For example, the check unit can evaluate the results of the bias check taking into account the geographic distribution. This makes it possible to check for bias toward a specific region by taking into account the geographic distribution of the images. Some or all of the above-described processing in the check unit can be performed using, or without, AI. For example, the check unit can input geographic distribution data into AI, which can then perform the bias check.

[0157] During the bias check, the check unit can improve the accuracy of the check by referring to related literature of the image. Reference to related literature includes, but is not limited to, criteria such as a literature database and the number of citations. For example, the check unit can refer to literature related to the image to accurately check for the presence or absence of bias. The check unit can also adjust the criteria for the bias check based on the related literature. For example, the check unit can evaluate the results of the bias check by referring to the related literature. Thus, by referring to the related literature of the image, the accuracy of the bias check can be improved. Some or all of the above-described processing in the check unit can be performed using, for example, AI, or can be performed without using AI. For example, the check unit can input related literature data into AI, which can improve the accuracy of the bias check.

[0158] When performing a bias check, the check unit can perform a check based on the market value of the image. Evaluation of market value includes, but is not limited to, criteria such as sales data and evaluation scores. For example, the check unit can focus the bias check on images with high market value. The check unit can also perform a lighter bias check on images with low market value. For example, the check unit can adjust the weighting of the bias check based on market value. This allows the bias check to be focused on more important images by taking the market value of the image into consideration. Some or all of the above-described processing in the check unit can be performed using, or without, AI. For example, the check unit can input market value data into AI, which then performs the bias check.

[0159] The presentation unit can estimate the user's emotions and adjust the presentation method of bias evidence based on the emotions. Examples of adjustment of the presentation method of bias evidence include, but are not limited to, text format, graph format, and the like. For example, if the user is relaxed, the presentation unit can present detailed bias evidence. Furthermore, if the user is in a hurry, the presentation unit can present simplified bias evidence. For example, if the user is excited, the presentation unit can present visually stimulating bias evidence. This allows for more appropriate information to be provided by adjusting the presentation method of bias evidence according to the user's emotions. The emotion estimation is realized using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, for example, a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to these examples. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or without AI. For example, the presentation unit can input the user's emotion data into the generation AI, which can then adjust the presentation method of bias evidence.

[0160] When presenting bias evidence, the presentation unit can optimize current evidence based on past evidence data. Use of past evidence data includes, but is not limited to, criteria such as past evaluation results and feedback. The presentation unit, for example, optimizes current bias evidence based on past evidence data. The presentation unit can also adjust the presentation method of bias evidence by referring to past evidence data. For example, the presentation unit analyzes past evidence data and presents optimal bias evidence. This allows current bias evidence to be optimized by referring to the past evidence data. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input past evidence data into AI, which then optimizes current evidence.

[0161] When presenting bias evidence, the presentation unit can improve the method of presenting the evidence based on user feedback. Improvements to the method of presenting the evidence include, but are not limited to, criteria such as a method of reflecting user feedback and adjusting the presentation format. For example, the presentation unit improves the method of presenting the bias evidence based on user feedback. The presentation unit can also adjust the level of detail of the evidence by reflecting user feedback. For example, the presentation unit improves the visual presentation method of the evidence by referring to user feedback. In this way, the presentation method of the evidence can be improved by reflecting user feedback. Some or all of the above-mentioned processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input user feedback into AI, which can improve the method of presenting the evidence.

[0162] When presenting bias evidence, the presentation unit can customize the method of presenting the evidence based on the user's past feedback. Customization of the method of presenting the evidence includes, but is not limited to, criteria such as a method of reflecting the user's past feedback and adjusting the presentation format. For example, the presentation unit customizes the method of presenting the evidence based on the user's past feedback. The presentation unit can also adjust the level of detail of the evidence by reflecting the user's past feedback. For example, the presentation unit customizes the visual presentation method of the evidence by referring to the user's past feedback. In this way, the method of presenting the evidence can be customized by reflecting the user's past feedback. Some or all of the above-described processing in the presentation unit may be performed using, for example, AI, or may be performed without using AI. For example, the presentation unit can input the user's past feedback into AI, which can then customize the method of presenting the evidence.

[0163] The presentation unit can estimate the user's emotions and prioritize bias evidence based on the emotions. The prioritization of bias evidence can be based on criteria such as, but not limited to, importance scores and user interest levels. For example, if the user is relaxed, the presentation unit can prioritize detailed bias evidence. Furthermore, if the user is in a hurry, the presentation unit can prioritize important bias evidence. For example, if the user is excited, the presentation unit can prioritize visually stimulating bias evidence. This prioritizes bias evidence based on the user's emotions, allowing for more important information to be provided preferentially. The emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the presentation unit can be performed using, for example, an AI. For example, the presentation unit can input the user's emotion data into the generation AI, which can then prioritize the bias evidence.

[0164] When presenting bias evidence, the presentation unit can select an optimal evidence presentation method based on the user's geographical location information. Use of geographical location information includes, but is not limited to, GPS data, location information services, etc. For example, if the user is in Tokyo, the presentation unit can prioritize presenting bias evidence related to Tokyo. Furthermore, if the user is in Kyoto, the presentation unit can also prioritize presenting bias evidence related to Kyoto. For example, if the user is in Okinawa, the presentation unit can prioritize presenting bias evidence related to Okinawa. This allows for providing more relevant evidence by taking the user's geographical location information into consideration. Some or all of the above-described processing in the presentation unit may be performed using, or without, AI. For example, the presentation unit can input the user's geographical location information into AI, which can then select an optimal evidence presentation method.

[0165] When presenting bias evidence, the presentation unit can analyze the user's social media activity and suggest a method for presenting the evidence based on the activity. Analysis of social media activity includes, but is not limited to, criteria such as analysis of post content and analysis of followers. The presentation unit can present relevant bias evidence, for example, based on images and videos shared by the user on social media. The presentation unit can also analyze the user's social media posts to present bias evidence that may be of interest. For example, the presentation unit can present relevant bias evidence based on the activity of the user's friends on social media. This allows for analysis of the user's social media activity to provide more relevant evidence. Some or all of the above-described processing by the presentation unit can be performed using, for example, AI, or without AI. For example, the presentation unit can input the user's social media activity data into AI, which can then suggest a method for presenting evidence.

[0166] The processing flow of the second embodiment will be briefly explained below.

[0167] Step 1: The collection unit collects data specific to Japan. Data specific to Japan includes Japanese culture, social background, language, etc. The collection unit collects data including traditional Japanese festivals, landscapes, and scenes from everyday life. The collection unit also collects public data on the Internet and data provided by users. For example, it collects images related to Japan from image databases on the Internet and also collects image data provided by users. Step 2: The construction unit builds an image model specialized for Japan based on the data collected by the collection unit. The image model is realized using technologies such as deep learning models and convolutional neural networks (CNN). For example, an image model specialized for Japan can be built using deep learning technology and generative AI to learn data that reflects Japanese culture and social background. Step 3: The check unit performs a bias check based on discomfort on the image model constructed by the construction unit. The bias check is performed by evaluating whether the user feels uncomfortable with the generated image. For example, if a user feels uncomfortable with the generated image, that feedback is collected and fed back to the generation AI. Discomfort can also be assessed using questionnaire surveys and psychological evaluations. Step 4: The presentation unit presents the bias evidence obtained by the checking unit. The bias evidence is presented in the form of statistical data, user feedback, etc. For example, the bias evidence is published in the form of a report and presented visually using graphs and charts.

[0168] 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.

[0169] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AIs include the data generation model 58, such as a neural network model (e.g., a neural network model), and a neural network model (e.g., a neural network model). The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating speech, text data indicating text, and image data indicating an image is also input to the data generation model 58. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0170] 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.

[0171] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0172] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0173] 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0174] 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.

[0175] 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.

[0176] 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.

[0177] 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).

[0178] 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.

[0179] 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.

[0180] 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.

[0181] 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.

[0182] 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.

[0183] 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.

[0184] 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.

[0185] 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.

[0186] 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.

[0187] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0188] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0189] 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.

[0190] 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.

[0191] 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.

[0192] 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.

[0193] 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).

[0194] 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.

[0195] 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.

[0196] 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.

[0197] 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.

[0198] In the headset type terminal 314, the identification process is performed by the processor 46. A identification program 60 is stored in the storage 50. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as a control unit 46A in accordance with the identification program 60 executed on the RAM 48. Note that the headset type terminal 314 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the identification processing unit 290 using these models.

[0199] 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.

[0200] 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.

[0201] 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.

[0202] 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.

[0203] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0204] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0205] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0206] 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.

[0207] 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.

[0208] 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.

[0209] 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).

[0210] 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.

[0211] 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.

[0212] 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.

[0213] 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.

[0214] 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.

[0215] In the robot 414, the processor 46 performs the identification process. The storage 50 stores the identification program 60. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as the control unit 46A in accordance with the identification program 60 executed on the RAM 48. The robot 414 also has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform the same process as the identification processing unit 290 using these models.

[0216] 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.

[0217] 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.

[0218] 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.

[0219] 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.

[0220] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0221] 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.

[0222] 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.

[0223] 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.

[0224] 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).

[0225] 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.

[0226] 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."

[0227] 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.

[0228] 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.

[0229] 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.

[0230] 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.

[0231] 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.

[0232] 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.

[0233] 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.

[0234] 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.

[0235] 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.

[0236] 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.

[0237] 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.

[0238] 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.

[0239] [Explanation of symbols]

[0240] 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 collection department that collects data specific to Japan, a construction unit that constructs an image model specialized for Japan based on the data collected by the collection unit; a check unit that performs a bias check to evaluate discomfort with respect to the image model constructed by the construction unit; a presentation unit that presents bias evidence obtained by the check unit. A system characterized by:

2. The collecting unit Collecting data based on Japanese culture and social background 2. The system of claim 1.

3. The checking unit Evaluate whether the user feels uncomfortable with the image generated by the AI 2. The system of claim 1.

4. The checking unit Correct bias based on user feedback 2. The system of claim 1.

5. The presentation unit We will verify whether bias exists in the images generated by the AI ​​and publish the results in a report format.

2. The system of claim 1.

6. The presentation unit Providing information on AI ethics issues 2. The system of claim 1.

7. The collecting unit Inferring user emotions and adjusting the timing of data collection based on those emotions 2. The system of claim 1.

8. The collecting unit Collect data based on the culture and social background of each region in Japan and build region-specific image models 2. The system of claim 1.

Images