system

Abstract

We provide the system. [Solution] A means for presenting an image recognition task displayed on a terminal including display units, which requires the user to label images based on a specific task, A means for receiving the labeling information selected by the user and storing the labeling information, A means of training an artificial intelligence model using stored labeling information, A means of improving the accuracy of recognition of specific cultures and landscapes using a trained artificial intelligence model, A system that includes this.

Description

Technical Field

[0004] , ,

[0005] , , ,

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is 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 an explanation of a character of the chatbot, 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] Conventional CAPTCHA systems mainly utilize images generated abroad, and there have been few opportunities to deepen the understanding ability of artificial intelligence for Japanese culture and landscapes. For this reason, it is difficult to improve the recognition accuracy of artificial intelligence in Japan as a culture-specific type, and furthermore, there is a problem that the results of the labeling work are biased towards the development of AI research in other countries and contribute to it.

Means for Solving the Problems

[0005] This invention provides a CAPTCHA system that facilitates image recognition based on specific tasks, enabling users to label images related to unique Japanese culture and landscapes. This allows an artificial intelligence model to be trained using the stored labeling information, improving its recognition accuracy for Japanese culture and landscapes. Furthermore, by analyzing the stored labeling information and adjusting the parameters of the artificial intelligence model as needed, further improvements in accuracy can be achieved.

[0006] "Users" refers to individuals who use the system to work on image recognition tasks.

[0007] A "display unit" refers to a mechanism or software installed on a terminal that presents an image recognition task to the user.

[0008] An "image recognition task" refers to a task presented to a user, requiring them to select or label images based on a specific culture or landscape.

[0009] "Labeling information" refers to the data selected or entered by the user in the image recognition task.

[0010] An "artificial intelligence model" refers to an algorithm or program that learns from data and improves the accuracy of image recognition for specific cultures or landscapes.

[0011] "Parameters" refer to adjustable settings or conditions that an artificial intelligence model uses when learning.

[0012] "Culture" refers to the traditions and customs unique to a country or region, as well as the associated visual elements.

[0013] "Landscape" refers to the appearance or scenery formed by the natural and man-made elements of a particular area. [Brief explanation of the drawing]

[0014] [Figure 1]It is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

MODE FOR CARRYING OUT THE INVENTION

[0015] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0016] First, the terms used in the following description will be explained.

[0017] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units 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), and the like.

[0018] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0019] In the following embodiments, the numbered storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, and the like.

[0020] In the following embodiments, the numbered communication I / F (Interface) is an interface including a communication processor and 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), or Bluetooth (registered trademark), and the like.

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0022] [First Embodiment]

[0023] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0024] As shown in Figure 1, the 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.

[0025] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0027] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0028] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0030] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] As shown in Figure 2, in the data processing device 12, specific processing 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" related to the technology of this 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 processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0032] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0033] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0034] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0035] This invention improves the accuracy of artificial intelligence models through a CAPTCHA system that presents users with image recognition challenges focused on specific cultures or landscapes. This system is implemented through an overall process including a user terminal, a server, and an artificial intelligence model.

[0036] When this system is running, if a user accesses a specific webpage, their device will display an image recognition task as a CAPTCHA widget. The displayed images are related to cultural content of a specific country or region, and the user is given a specific task. For example, based on the instruction "Select the image that contains Mount Fuji," the user will select the appropriate image from the presented images.

[0037] After selection, the terminal sends the result to the server. The server stores the received labeling information in a database. This stored data is used to train an artificial intelligence model. The server periodically retrieves the labeling data and uses it in the model's learning process to improve recognition accuracy. The server also evaluates the model's performance and adjusts parameters as needed to continuously optimize the model's performance.

[0038] As a concrete example, consider a scenario where a user labels images of traditional Japanese festivals. In this case, the user is presented with images containing festival features (e.g., taiko drums or portable shrines) and instructed to select the appropriate image. Once the user completes their selection, the information is sent to a server, and the stored data is used to train an AI model. Through this process, the AI ​​becomes able to recognize festival photos with greater accuracy.

[0039] This invention is expected to deepen AI's understanding of Japan's unique culture and landscapes, enabling diverse applications.

[0040] The following describes the processing flow.

[0041] Step 1:

[0042] When the device detects the user's access to a webpage, it displays a CAPTCHA widget. The device then sends a request for the image set to the CAPTCHA server.

[0043] Step 2:

[0044] When the server receives a request, it randomly selects images related to Japanese culture or landscapes. It then sends the selected images and the corresponding task (e.g., "Please select images that include Mount Fuji") to the device.

[0045] Step 3:

[0046] The user follows the instructions displayed on the device and selects the appropriate image from the presented options. Once the task is completed, the user presses the "Submit" button.

[0047] Step 4:

[0048] The device sends information about the image selected by the user to the server as data. This data includes the ID and timestamp of the selected image.

[0049] Step 5:

[0050] The server stores the received labeling data in a database. This data is later used to train an artificial intelligence model.

[0051] Step 6:

[0052] The server periodically retrieves labeling information from the database and uses this information to train the artificial intelligence model. Through the training process, the AI ​​model improves its image recognition accuracy.

[0053] Step 7:

[0054] The server evaluates the model and continuously optimizes its performance by adjusting parameters as needed. This allows the AI ​​to more accurately identify cultural elements unique to Japan.

[0055] (Example 1)

[0056] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0057] Conventional machine learning models exhibit high accuracy in general image recognition tasks, but their accuracy can decline when it comes to specific ethnic characteristics or natural landscapes. Efficient data collection and model training methods for improving the recognition accuracy of images containing such cultural and geographical features remain a challenge that has not yet been sufficiently developed.

[0058] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0059] In this invention, the server includes means for presenting an image recognition task displayed by a device including a display unit, which requires the user to label images based on a specific task; means for receiving and storing the labeling information selected by the user; and means for training a machine learning model using the stored labeling information. This makes it possible to improve the recognition accuracy of the machine learning model for specific ethnicities or natural landscapes.

[0060] "User" refers to a person who uses the system to label images based on the assigned task.

[0061] A "task" refers to the instructions presented to the user, in which they must select an image based on specific conditions.

[0062] A "display unit" refers to a part of the device used by a user to present an image recognition task, specifically the hardware or software used to display images.

[0063] "Device" refers to a platform that presents tasks to users, comprising hardware or software components including a display unit.

[0064] "Labeling information" refers to images and related data selected by users based on a specific task.

[0065] "Storage" refers to the process of securely saving received labeling information in a database or storage device.

[0066] A "machine learning model" refers to an algorithm or computational process designed to improve recognition accuracy for a specific task using collected data.

[0067] "Ethnicity" refers to characteristics and elements based on a particular culture or social background.

[0068] "Natural landscape" refers to natural visual features that include specific geographical or environmental characteristics.

[0069] "Recognition accuracy" refers to the ability of a machine learning model to accurately interpret input data and provide the correct results for a specific task.

[0070] This invention improves the accuracy of image recognition for specific ethnicities or natural landscapes using a system that includes a user, a terminal, and a server. Specifically, the system involves a series of processes in which the terminal presents an image recognition task to the user, the user sends the labeling information to the server, and the server uses this information to train a machine learning model.

[0071] The device can take the form of an internet-connected computer or smartphone. The device is equipped with a display unit for showing image recognition tasks to the user and communicates with a server to provide image data. The displayed images possess specific cultural or geographical characteristics.

[0072] The user selects an image based on a task presented on the device. For example, if the user receives the prompt "Select an image showing a traditional Japanese festival," they will select an image related to a Japanese festival. The selected labeling information is securely stored in a database.

[0073] The server uses the received labeling information to train a generative AI model, improving the model's recognition accuracy. This process utilizes machine learning algorithms, incorporates the labeling information as a new dataset, and periodically evaluates and adjusts the model's performance.

[0074] As a concrete example, when a user labels images related to festivals, the server feeds that data to a machine learning model, which is then trained to recognize festival images with high accuracy. An example of this prompt would be, "Please select images that include Japanese drums or portable shrines."

[0075] Through this system, machine learning models can more accurately recognize specific cultural and geographical objects, enabling their use in a variety of applications.

[0076] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0077] Step 1:

[0078] When a user accesses a webpage, the device launches a CAPTCHA widget. It receives the user's request as input and displays the CAPTCHA widget as output. Based on this access, the device sends a request to the server, requesting image data.

[0079] Step 2:

[0080] The server receives requests from terminals and sends image data necessary for the image recognition task to the terminals. It receives data containing the content of the requested task as input and sends image data to the terminals as output. The server retrieves a set of images from the database that match the specified task.

[0081] Step 3:

[0082] The terminal uses image data received from the server to present an image recognition task to the user. It takes the received image data as input and displays the task to the user as output. The terminal performs specific actions to display images and instructions on the screen to help the user visually understand the task.

[0083] Step 4:

[0084] The user selects the image that best matches the task from the presented images. As input, the user visually reviews multiple images, and as output, retrieves information about the selected image. The user then clicks or taps the image that is most appropriate based on the task.

[0085] Step 5:

[0086] The terminal sends the user's selection results to the server. It receives identification information for the selected image as input and sends that information to the server as output. The terminal packages the selection in the correct format and sends it to the server as an HTTP request.

[0087] Step 6:

[0088] The server stores the received labeling information in a database. It receives user-selected labeling information as input and stores that information in the database as output. The server then appropriately places the received data in the storage system and updates the index.

[0089] Step 7:

[0090] The server trains a generative AI model using stored labeling data. It takes stored data as input and generates updated parameters for the model as output. The server applies machine learning algorithms and feeds data back to the model to facilitate learning.

[0091] Step 8:

[0092] The server evaluates the model's performance and adjusts parameters as needed. It analyzes the output data of the trained model as input and creates an optimized model as output. The server then evaluates the model using test data to improve accuracy and adjusts the learning rate and other parameters.

[0093] (Application Example 1)

[0094] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0095] In recent years, there has been a growing need to accurately identify specific regional cultures and natural landscapes, but conventional artificial intelligence models struggle to fully understand such cultural elements. This problem hinders the provision of accurate information and the improvement of tourist experiences, particularly in the tourism and regional development industries. Therefore, there is a need to provide systems that can utilize cultural and geographical information more effectively.

[0096] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0097] In this invention, the server includes means for presenting an image recognition task, which is presented by an information processing terminal including a display device, in which the user is asked to label images based on a specific task; means for training a machine learning model using stored labeling information; and means for analyzing captured images and classifying whether the images have specific features. This makes it possible to recognize images related to specific regional cultures or natural landscapes with high accuracy and to provide cultural or geographical information based on the results.

[0098] An "information processing terminal" is an electronic device used by users to process data through an interface, and includes smartphones and tablets.

[0099] A "machine learning model" is an algorithm that automatically learns a specific task based on a vast amount of data, and is applied to image recognition and natural language processing.

[0100] A "display device" is a device used to present information visually, and displays and monitors fall into this category.

[0101] "Labeling information" refers to information that indicates the meaning associated with a particular image or data, and serves a role in classification and categorization.

[0102] "Cultural information" refers to traditions, customs, and historical knowledge unique to a particular region or society, and is an important element in tourism and education.

[0103] "Geographic information" refers to information that indicates the location, characteristics, and relationships of a specific geographical area, and is essential for spatial awareness and mapping services.

[0104] The system for carrying out this invention comprises an information processing terminal, a server, and a machine learning model. The information processing terminal is a terminal operated by the user, and a smartphone is an example of this. The terminal includes a display device that presents image recognition tasks related to local culture or natural landscapes based on a specific task. When the user labels the task displayed on the terminal, the result is sent to the server.

[0105] The server is responsible for storing the received labeling information in electronic memory. Furthermore, it uses this data to train a machine learning model. The machine learning model also functions as a generative AI model, improving its recognition accuracy for specific cultures and landscapes by learning from the labeled data. This model is built using TENSORFLOW®.

[0106] When a user takes an image using their device, the captured image is analyzed to identify whether it contains specific cultural and geographical features. Based on the identification results, relevant cultural or geographical information is displayed on the device. For example, if traditional Japanese pottery is recognized, information about its history and manufacturing methods will be displayed in real time.

[0107] As a concrete example, when a tourist photographs an exhibit at a traditional festival, the server and a machine learning model can collaborate to provide relevant cultural information. An example of a prompt to input to the generative AI model is, "Use an image taken at the most recent festival and generate relevant cultural information." This allows the system to support a deeper cultural understanding and enhance understanding of the region.

[0108] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0109] Step 1:

[0110] A user takes an image using an information processing terminal. The input is a raw image obtained through the camera. This image may belong to a local culture or natural landscape. Here, the terminal uses camera hardware to capture and store the image data.

[0111] Step 2:

[0112] The device analyzes captured images and identifies whether they possess predefined features. The input is the image data obtained in step 1. This process uses an image analysis algorithm to resize the image to a specific size and perform data processing such as normalizing the color tone. The output is a feature list based on the analysis results.

[0113] Step 3:

[0114] Based on the analysis results, the device associates images with cultural or geographical information. The input is the feature list obtained in step 2. A generative AI model is used to generate a prompt sentence. This prompt sentence serves as a queue for retrieving relevant cultural information. The output is the prompt sentence and the associated information.

[0115] Step 4:

[0116] The terminal sends the prompt and related information to the server. The input is the prompt and related information generated in step 3. The terminal transfers the data to the server using the network interface. The output is the prompt and information stored on the server.

[0117] Step 5:

[0118] The server uses the received prompt message to utilize a generative AI model to generate detailed cultural information. The input is the data received in step 4. The server uses the generative AI model to perform data calculations to improve the accuracy and detail of the relevant information. The output is the generated detailed cultural information.

[0119] Step 6:

[0120] The server distributes the generated cultural information to the user's information processing terminal. The input is the detailed cultural information generated in step 5. The server uses the network interface to transmit the information to the terminal. The output is the cultural information displayed on the user's terminal.

[0121] Step 7:

[0122] The user reviews and understands the cultural information displayed on the device. The input is the cultural information received in step 6. The user views the displayed content and deepens their understanding of the local culture. The output is the user's improved knowledge.

[0123] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0124] This invention provides a system that improves the accuracy of an artificial intelligence model and the user experience through an image recognition task combined with an emotion engine that recognizes user emotions. This system includes a user terminal, a server, an emotion engine, and an artificial intelligence model.

[0125] When a user accesses a specific webpage, their device displays a CAPTCHA widget presenting an image recognition challenge. This widget incorporates an emotion engine that analyzes the user's emotions from their facial expressions and voice as they tackle the challenge. For example, if the user shows a positive response, the difficulty of the challenge may be adjusted to present a more satisfying challenge.

[0126] When a user selects an image, the device sends emotion data along with the selection to the server. The server stores the received data in a database and uses the emotion data, along with labeling information, to train an AI model. This allows the AI ​​model to understand changes in the user's emotions and improve the accuracy of image recognition accordingly.

[0127] For example, when a user labels images of traditional Japanese festivals, the system detects smiles and expressions of surprise to engage the user, and then presents related images and tasks one after another. Through this process, active user participation is encouraged, the quality of the labeling improves, and the recognition accuracy of the AI ​​model increases.

[0128] Furthermore, emotional data is used not only for training models but also for providing personalized user experiences. Thus, this embodiment of the present invention aims to leverage an emotional engine to extend the capabilities of AI technology and improve the user experience.

[0129] The following describes the processing flow.

[0130] Step 1:

[0131] When a user accesses a webpage, their device displays a CAPTCHA widget. This widget presents the user with questions related to Japanese culture and landscapes.

[0132] Step 2:

[0133] As soon as the user begins working on a task, the device activates an emotion engine that analyzes the user's emotions in real time from their facial expressions and voice. This information is used to adjust the difficulty of the task and improve the user experience.

[0134] Step 3:

[0135] The user selects the correct image from the presented images. During the selection process, the emotion engine records the user's facial expressions and reactions and prepares to send this data to the server.

[0136] Step 4:

[0137] The device sends the labeling information selected by the user and the emotion data acquired by the emotion engine to the server. This data includes the image ID and the user's emotional state.

[0138] Step 5:

[0139] The server stores the received labeling and sentiment data in a database. This information is later used as a training dataset for the AI ​​model.

[0140] Step 6:

[0141] The server periodically retrieves labeling and sentiment data from the database and uses it to train an artificial intelligence model. The training process takes into account the user's emotional state and aims to improve the accuracy of recognizing specific cultures and landscapes.

[0142] Step 7:

[0143] The server evaluates the trained model and optimizes its performance by adjusting parameters as needed. This optimization further improves accuracy by leveraging user sentiment data.

[0144] (Example 2)

[0145] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0146] Conventional image recognition systems only require simple labeling without considering user emotions, making it difficult to improve the user experience or maximize the recognition accuracy of artificial intelligence models. Furthermore, there is a lack of mechanisms to utilize emotional data when improving the accuracy of identifying specific cultures or landscapes.

[0147] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0148] In this invention, the server includes means for receiving user-selected labeling information and sentiment data and storing the information on a recording medium; means for training an artificial intelligence model using the stored labeling information and sentiment data; and means for improving the recognition accuracy for specific cultures and landscapes using the trained artificial intelligence model. This not only improves the accuracy of image recognition by leveraging the user's emotions, but also enhances the user experience by providing personalized tasks and enables effective training of the artificial intelligence model.

[0149] A "user" is any individual or organization that can use this system.

[0150] An "image recognition task" is a task in which a user selects the correct image from multiple displayed images based on specific criteria.

[0151] A "display device" is a component of a terminal used to present information visually, and its role is to show images or tasks to the user.

[0152] An "emotion analysis device" is a device that analyzes a user's facial expressions and voice to estimate and evaluate their emotional state.

[0153] "Labeling information" refers to information related to the identification and classification of images selected by the user based on the task.

[0154] "Emotional data" refers to information about the user's emotional state, including data acquired by an emotion analysis device.

[0155] A "recording medium" is a physical or electronic medium that can store information.

[0156] An "artificial intelligence model" is an algorithm or program designed to analyze large amounts of data and perform a specific task.

[0157] "Discrimination accuracy" is a measure of an artificial intelligence model's ability to accurately classify or recognize given data.

[0158] "Regionally specific culture" refers to a culture that exhibits characteristics such as traditions, customs, arts, and lifestyles within a particular geographical area.

[0159] This invention is a system that utilizes user emotion data to improve the accuracy of an artificial intelligence model through an image recognition task. The system comprises a user terminal, a server, an emotion analysis device, and an artificial intelligence model.

[0160] When a user accesses a specific webpage, the device presents an image recognition task via a display. The device incorporates an emotion analysis system that analyzes the user's facial expressions and voice while they work on the task, acquiring emotion data. Emotion analysis utilizes hardware such as a facial recognition camera and microphone.

[0161] When a user selects an image, the device sends the selected image's labeling information along with the acquired sentiment data to the server. The server stores this information on a recording medium and uses it to train an artificial intelligence model. The server analyzes the received labeling information and sentiment data to train the AI ​​model and improve its recognition accuracy.

[0162] For example, when a user is labeling images related to a festival, if the emotion analysis device detects the user's smile, the server adjusts the difficulty of the next task presented, offering a task that is more interesting to the user. This process allows the user to engage with the task more actively and also improves the model's recognition accuracy.

[0163] A concrete example of a prompt would be a text-based instruction such as, "What image is best to present when the user is smiling?" Based on this instruction, the generative AI model generates the optimal output according to the user's emotions. This improves the user experience and enables more sophisticated image recognition.

[0164] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0165] Step 1:

[0166] A user accesses a web page

[0167] The user accesses a specific webpage from a web browser using their device. At this point, the input is the user's access request. The output is the target webpage displayed on the device.

[0168] Step 2:

[0169] The device displays a CAPTCHA widget.

[0170] The device displays a CAPTCHA widget embedded in the webpage. This widget has an integrated sentiment analyzer and is ready for processing. The input is the data from the webpage, and the output is the CAPTCHA widget with the sentiment analyzer.

[0171] Step 3:

[0172] Users tackle image recognition challenges.

[0173] The user works on a presented image recognition task and selects images that meet the given criteria. At this point, the input is the presented set of images, and the output is information about the image selected by the user.

[0174] Step 4:

[0175] The device acquires emotional data.

[0176] The device acquires the user's facial expressions and voice using an emotion analysis device and stores them as emotion data. Specifically, it uses a camera and microphone to capture and analyze the user's emotions. The input is the user's real-time facial expressions and voice, and the output is the analyzed emotion data.

[0177] Step 5:

[0178] The device sends data to the server.

[0179] The device sends the image and sentiment data selected by the user to the server. A secure communication protocol is used for transmission. The input is the user's selected data and sentiment data, and the output is the completion of the data transmission to the server.

[0180] Step 6:

[0181] The server stores and processes the data.

[0182] The server stores the received data on a recording medium and analyzes the sentiment data and labeling information. Here, it processes the data into a dataset suitable for training an AI model. The input is the received data, and the output is the stored database entries and the training dataset.

[0183] Step 7:

[0184] The server trains the AI ​​model.

[0185] The server trains an AI model based on stored data to improve its recognition accuracy. The input is the training dataset, and the output is the updated AI model.

[0186] Step 8:

[0187] The system provides feedback

[0188] The system adjusts the difficulty level of the next task displayed and presents images and tasks that match the user's emotional state. The input is an updated AI model and the user's emotional information, and the output is a customized next task.

[0189] (Application Example 2)

[0190] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0191] Conventional image recognition systems failed to fully capture user interest and satisfaction because they provided uniform tasks without considering user emotions. Furthermore, the training of artificial intelligence models lacked sufficient use of emotional data to improve the user experience. This resulted in challenges to overall system performance and personalization.

[0192] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0193] In this invention, the server includes means for the user to label images based on a specific task and transmit them along with sentiment data; means for storing the received labeling information and sentiment data and using them to train an artificial intelligence model; and means for dynamically presenting images and tasks based on the user's preferences using the trained model. This enables a personalized experience that takes into account the user's emotions and improved accuracy in image recognition.

[0194] A "display device" is a device used to visually present information to users, and has the function of displaying images or tasks.

[0195] "Emotional data" refers to data that indicates the emotional state of a user, and includes information obtained from facial expressions, voice, and other sources.

[0196] "Labeling information" refers to the identification information of an image as determined by the user, and is data used to indicate the content and category of the image.

[0197] An "artificial intelligence model" is an algorithm trained using machine learning techniques and is used for tasks such as image recognition.

[0198] A "personalized user experience" is an experience optimized based on the individual preferences and emotions of the user, providing content and services that are tailored to each individual user.

[0199] The system of this invention presents users with a task of labeling images with emotional data, trains an artificial intelligence model using the collected data, and provides a personalized user experience. The implementation details of this system are described below.

[0200] The server displays images through a display device such as smart glasses or a smartphone, presenting the user with a specific task. Emotion analysis software, such as EmotionAPI, is used to acquire emotional data from the user's facial expressions and voice. This emotional data is captured in real time by hardware such as cameras and microphones and transmitted to the server.

[0201] The server uses the received labeling information and sentiment data to train an artificial intelligence model using machine learning libraries such as TensorFlow and Keras. This model is then used to improve the personalized user experience based on the collected data. For example, suppose a smile is detected when a user is browsing a specific item in a virtual fashion store. The system recognizes this as a positive reaction and recommends related products to maximize the user's purchase intent.

[0202] By utilizing generative AI models and using prompts such as, "Present similar items to products the user is interested in, based on the user's sentiment data, in real time," a richer and more customized user experience can be achieved.

[0203] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0204] Step 1:

[0205] The device presents the user with an image recognition task. The user views the task through a display device such as smart glasses or a smartphone. Multiple images related to the task are displayed at this time. The input consists of images and task information, and the output is the displayed image. The device retrieves task information from its internal memory and presents it to the user as a component.

[0206] Step 2:

[0207] The user selects an image according to the task. This selection is based on the user's decision, and the selected image is generated as labeling information. The input is the image selected by the user, and the output is the labeling information. Based on the user's selection, the terminal stores this information in cache memory.

[0208] Step 3:

[0209] The device launches emotion analysis software and acquires emotion data from the user's facial expressions and voice. Data is collected in real time via the camera and microphone and analyzed using the EmotionAPI. The input is the user's facial expressions and voice data, and the output is emotion data. The device temporarily stores the analyzed emotion data.

[0210] Step 4:

[0211] The terminal sends the acquired labeling information and sentiment data to the server. This data is used in the next analysis step. The input is the labeling information and sentiment data, and the output is the data sent to the server. The terminal sends information to the server as data packets over the network.

[0212] Step 5:

[0213] The server trains an artificial intelligence model using the received labeling and sentiment data. This model is built using TensorFlow and Keras, and optimizes itself by identifying anomalies based on the dataset. The input is the received labeling and sentiment data, and the output is the updated artificial intelligence model. The server allocates computing resources to perform high-speed data calculations.

[0214] Step 6:

[0215] The server uses a trained artificial intelligence model to generate personalized recommendations for the user. This selects content that is appropriate for the user's next action. The input is the trained AI model, and the output is the recommended content. The server leverages the generative AI model to adjust the prompt output in real time.

[0216] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0217] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0218] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0219] [Second Embodiment]

[0220] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0221] As shown in Figure 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.

[0222] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0223] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0224] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0225] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0226] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0227] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0228] The specific processing program 56 is an example of a "program" relating to the technology of this 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 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.

[0229] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0230] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0231] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0232] This invention improves the accuracy of artificial intelligence models through a CAPTCHA system that presents users with image recognition challenges focused on specific cultures or landscapes. This system is implemented through an overall process including a user terminal, a server, and an artificial intelligence model.

[0233] When this system is running, if a user accesses a specific webpage, their device will display an image recognition task as a CAPTCHA widget. The displayed images are related to cultural content of a specific country or region, and the user is given a specific task. For example, based on the instruction "Select the image that contains Mount Fuji," the user will select the appropriate image from the presented images.

[0234] After selection, the terminal sends the result to the server. The server stores the received labeling information in a database. This stored data is used to train an artificial intelligence model. The server periodically retrieves the labeling data and uses it in the model's learning process to improve recognition accuracy. The server also evaluates the model's performance and adjusts parameters as needed to continuously optimize the model's performance.

[0235] As a concrete example, consider a scenario where a user labels images of traditional Japanese festivals. In this case, the user is presented with images containing festival features (e.g., taiko drums or portable shrines) and instructed to select the appropriate image. Once the user completes their selection, the information is sent to a server, and the stored data is used to train an AI model. Through this process, the AI ​​becomes able to recognize festival photos with greater accuracy.

[0236] This invention is expected to deepen AI's understanding of Japan's unique culture and landscapes, enabling diverse applications.

[0237] The following describes the processing flow.

[0238] Step 1:

[0239] When the device detects the user's access to a webpage, it displays a CAPTCHA widget. The device then sends a request for the image set to the CAPTCHA server.

[0240] Step 2:

[0241] When the server receives a request, it randomly selects images related to Japanese culture or landscapes. It then sends the selected images and the corresponding task (e.g., "Please select images that include Mount Fuji") to the device.

[0242] Step 3:

[0243] The user follows the instructions displayed on the device and selects the appropriate image from the presented options. Once the task is completed, the user presses the "Submit" button.

[0244] Step 4:

[0245] The device sends information about the image selected by the user to the server as data. This data includes the ID and timestamp of the selected image.

[0246] Step 5:

[0247] The server stores the received labeling data in a database. This data is later used to train an artificial intelligence model.

[0248] Step 6:

[0249] The server periodically retrieves labeling information from the database and uses this information to train the artificial intelligence model. Through the training process, the AI ​​model improves its image recognition accuracy.

[0250] Step 7:

[0251] The server evaluates the model and continuously optimizes its performance by adjusting parameters as needed. This allows the AI ​​to more accurately identify cultural elements unique to Japan.

[0252] (Example 1)

[0253] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0254] Conventional machine learning models exhibit high accuracy in general image recognition tasks, but their accuracy can decline when it comes to specific ethnic characteristics or natural landscapes. Efficient data collection and model training methods for improving the recognition accuracy of images containing such cultural and geographical features remain a challenge that has not yet been sufficiently developed.

[0255] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0256] In this invention, the server includes means for presenting an image recognition task displayed by a device including a display unit, which requires the user to label images based on a specific task; means for receiving and storing the labeling information selected by the user; and means for training a machine learning model using the stored labeling information. This makes it possible to improve the recognition accuracy of the machine learning model for specific ethnicities or natural landscapes.

[0257] "User" refers to a person who uses the system to label images based on the assigned task.

[0258] A "task" refers to the instructions presented to the user, in which they must select an image based on specific conditions.

[0259] A "display unit" refers to a part of the device used by a user to present an image recognition task, specifically the hardware or software used to display images.

[0260] "Device" refers to a platform that presents tasks to users, comprising hardware or software components including a display unit.

[0261] "Labeling information" refers to images and related data selected by users based on a specific task.

[0262] "Storage" refers to the process of securely saving received labeling information in a database or storage device.

[0263] A "machine learning model" refers to an algorithm or computational process designed to improve recognition accuracy for a specific task using collected data.

[0264] "Ethnicity" refers to characteristics and elements based on a particular culture or social background.

[0265] "Natural landscape" refers to natural visual features that include specific geographical or environmental characteristics.

[0266] "Recognition accuracy" refers to the ability of a machine learning model to accurately interpret input data and provide the correct results for a specific task.

[0267] This invention improves the accuracy of image recognition for specific ethnicities or natural landscapes using a system that includes a user, a terminal, and a server. Specifically, the system involves a series of processes in which the terminal presents an image recognition task to the user, the user sends the labeling information to the server, and the server uses this information to train a machine learning model.

[0268] The device can take the form of an internet-connected computer or smartphone. The device is equipped with a display unit for showing image recognition tasks to the user and communicates with a server to provide image data. The displayed images possess specific cultural or geographical characteristics.

[0269] The user selects an image based on a task presented on the device. For example, if the user receives the prompt "Select an image showing a traditional Japanese festival," they will select an image related to a Japanese festival. The selected labeling information is securely stored in a database.

[0270] The server uses the received labeling information to train a generative AI model, improving the model's recognition accuracy. This process utilizes machine learning algorithms, incorporates the labeling information as a new dataset, and periodically evaluates and adjusts the model's performance.

[0271] As a concrete example, when a user labels images related to festivals, the server feeds that data to a machine learning model, which is then trained to recognize festival images with high accuracy. An example of this prompt would be, "Please select images that include Japanese drums or portable shrines."

[0272] Through this system, machine learning models can more accurately recognize specific cultural and geographical objects, enabling their use in a variety of applications.

[0273] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0274] Step 1:

[0275] When a user accesses a webpage, the device launches a CAPTCHA widget. It receives the user's request as input and displays the CAPTCHA widget as output. Based on this access, the device sends a request to the server, requesting image data.

[0276] Step 2:

[0277] The server receives requests from terminals and sends image data necessary for the image recognition task to the terminals. It receives data containing the content of the requested task as input and sends image data to the terminals as output. The server retrieves a set of images from the database that match the specified task.

[0278] Step 3:

[0279] The terminal presents an image recognition task to the user using the image data received from the server. As input, it captures the received image data, and as output, it displays the task to the user. The terminal performs specific operations of displaying the image and instructions on the screen to visually enable the user to understand the task.

[0280] Step 4:

[0281] The user selects an image that matches the task from the presented images. As input, the user visually checks multiple images, and as output, obtains information on the selected image. The user performs an operation of clicking or tapping on the most appropriate image based on the task.

[0282] Step 5:

[0283] The terminal sends the user's selection result to the server. As input, it receives the identification information of the selected image, and as output, it sends that information to the server. The terminal packages the selection content in an accurate format and performs the operation of sending it to the server as an HTTP request.

[0284] Step 6:

[0285] The server saves the received labeling information in the database. As input, it receives the labeling information selected by the user, and as output, it stores that information in the database. The server appropriately places the received data in the storage system and performs the operation of updating the index.

[0286] Step 7:

[0287] The server trains a generative AI model using the saved labeling data. As input, it captures the saved data, and as output, generates updated parameters of the model. The server applies a machine learning algorithm and performs the operation of feeding data back to the model to promote learning.

[0288] Step 8:

[0289] The server evaluates the model's performance and adjusts parameters as needed. It analyzes the output data of the trained model as input and creates an optimized model as output. The server then evaluates the model using test data to improve accuracy and adjusts the learning rate and other parameters.

[0290] (Application Example 1)

[0291] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0292] In recent years, there has been a growing need to accurately identify specific regional cultures and natural landscapes, but conventional artificial intelligence models struggle to fully understand such cultural elements. This problem hinders the provision of accurate information and the improvement of tourist experiences, particularly in the tourism and regional development industries. Therefore, there is a need to provide systems that can utilize cultural and geographical information more effectively.

[0293] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0294] In this invention, the server includes means for presenting an image recognition task, which is presented by an information processing terminal including a display device, in which the user is asked to label images based on a specific task; means for training a machine learning model using stored labeling information; and means for analyzing captured images and classifying whether the images have specific features. This makes it possible to recognize images related to specific regional cultures or natural landscapes with high accuracy and to provide cultural or geographical information based on the results.

[0295] An "information processing terminal" is an electronic device used by users to process data through an interface, and includes smartphones and tablets.

[0296] A "machine learning model" is an algorithm that automatically learns a specific task based on a vast amount of data, and is applied to image recognition and natural language processing.

[0297] A "display device" is a device used to present information visually, and displays and monitors fall into this category.

[0298] "Labeling information" refers to information that indicates the meaning associated with a particular image or data, and serves a role in classification and categorization.

[0299] "Cultural information" refers to traditions, customs, and historical knowledge unique to a particular region or society, and is an important element in tourism and education.

[0300] "Geographic information" refers to information that indicates the location, characteristics, and relationships of a specific geographical area, and is essential for spatial awareness and mapping services.

[0301] The system for carrying out this invention comprises an information processing terminal, a server, and a machine learning model. The information processing terminal is a terminal operated by the user, and a smartphone is an example of this. The terminal includes a display device that presents image recognition tasks related to local culture or natural landscapes based on a specific task. When the user labels the task displayed on the terminal, the result is sent to the server.

[0302] The server is responsible for storing the received labeling information in electronic storage. Furthermore, it uses this data to train a machine learning model. This machine learning model also functions as a generative AI model, improving its recognition accuracy for specific cultures and landscapes by learning from the labeled data. This model is built using TensorFlow.

[0303] When a user takes a picture through a terminal, the captured image is analyzed to identify whether it has specific cultural and geographical features. Based on the identification result, relevant cultural or geographical information is presented to the terminal. For example, when traditional Japanese pottery is recognized, information about its history and manufacturing method is displayed in real time.

[0304] As a specific example, when a tourist takes a picture of a display of a traditional festival, the server and the machine learning model can cooperate to provide the corresponding cultural information. An example of a prompt sentence input to the generative AI model is "Please generate cultural information related to the image taken at the latest festival using the image." This enables the system to support a deeper cultural understanding and enhance the understanding of the region.

[0305] The flow of the specific process in Application Example 1 will be described using FIG. 12.

[0306] Step 1:

[0307] The user takes a picture using an information processing terminal. The input is the raw image obtained through the camera. This image may belong to the local culture or natural scenery. Here, the terminal uses the camera hardware to capture and save the image data.

[0308] Step 2:

[0309] The terminal analyzes the captured image to identify whether it has pre-defined features. The input is the image data obtained in Step 1. In this process, an image analysis algorithm is used to perform data processing such as resizing the image to a specific size and normalizing the color tone. The output is a feature list based on the analysis result.

[0310] Step 3:

[0311] Based on the analysis results, the device associates images with cultural or geographical information. The input is the feature list obtained in step 2. A generative AI model is used to generate a prompt sentence. This prompt sentence serves as a queue for retrieving relevant cultural information. The output is the prompt sentence and the associated information.

[0312] Step 4:

[0313] The terminal sends the prompt and related information to the server. The input is the prompt and related information generated in step 3. The terminal transfers the data to the server using the network interface. The output is the prompt and information stored on the server.

[0314] Step 5:

[0315] The server uses the received prompt message to utilize a generative AI model to generate detailed cultural information. The input is the data received in step 4. The server uses the generative AI model to perform data calculations to improve the accuracy and detail of the relevant information. The output is the generated detailed cultural information.

[0316] Step 6:

[0317] The server distributes the generated cultural information to the user's information processing terminal. The input is the detailed cultural information generated in step 5. The server uses the network interface to transmit the information to the terminal. The output is the cultural information displayed on the user's terminal.

[0318] Step 7:

[0319] The user reviews and understands the cultural information displayed on the device. The input is the cultural information received in step 6. The user views the displayed content and deepens their understanding of the local culture. The output is the user's improved knowledge.

[0320] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0321] This invention provides a system that improves the accuracy of an artificial intelligence model and the user experience through an image recognition task combined with an emotion engine that recognizes user emotions. This system includes a user terminal, a server, an emotion engine, and an artificial intelligence model.

[0322] When a user accesses a specific webpage, their device displays a CAPTCHA widget presenting an image recognition challenge. This widget incorporates an emotion engine that analyzes the user's emotions from their facial expressions and voice as they tackle the challenge. For example, if the user shows a positive response, the difficulty of the challenge may be adjusted to present a more satisfying challenge.

[0323] When a user selects an image, the device sends emotion data along with the selection to the server. The server stores the received data in a database and uses the emotion data, along with labeling information, to train an AI model. This allows the AI ​​model to understand changes in the user's emotions and improve the accuracy of image recognition accordingly.

[0324] For example, when a user labels images of traditional Japanese festivals, the system detects smiles and expressions of surprise to engage the user, and then presents related images and tasks one after another. Through this process, active user participation is encouraged, the quality of the labeling improves, and the recognition accuracy of the AI ​​model increases.

[0325] Furthermore, emotional data is used not only for training models but also for providing personalized user experiences. Thus, this embodiment of the present invention aims to leverage an emotional engine to extend the capabilities of AI technology and improve the user experience.

[0326] The following describes the processing flow.

[0327] Step 1:

[0328] When a user accesses a webpage, their device displays a CAPTCHA widget. This widget presents the user with questions related to Japanese culture and landscapes.

[0329] Step 2:

[0330] As soon as the user begins working on a task, the device activates an emotion engine that analyzes the user's emotions in real time from their facial expressions and voice. This information is used to adjust the difficulty of the task and improve the user experience.

[0331] Step 3:

[0332] The user selects the correct image from the presented images. During the selection process, the emotion engine records the user's facial expressions and reactions and prepares to send this data to the server.

[0333] Step 4:

[0334] The device sends the labeling information selected by the user and the emotion data acquired by the emotion engine to the server. This data includes the image ID and the user's emotional state.

[0335] Step 5:

[0336] The server stores the received labeling and sentiment data in a database. This information is later used as a training dataset for the AI ​​model.

[0337] Step 6:

[0338] The server periodically retrieves labeling and sentiment data from the database and uses it to train an artificial intelligence model. The training process takes into account the user's emotional state and aims to improve the accuracy of recognizing specific cultures and landscapes.

[0339] Step 7:

[0340] The server evaluates the trained model and optimizes its performance by adjusting parameters as needed. This optimization further improves accuracy by leveraging user sentiment data.

[0341] (Example 2)

[0342] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0343] Conventional image recognition systems only require simple labeling without considering user emotions, making it difficult to improve the user experience or maximize the recognition accuracy of artificial intelligence models. Furthermore, there is a lack of mechanisms to utilize emotional data when improving the accuracy of identifying specific cultures or landscapes.

[0344] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0345] In this invention, the server includes means for receiving user-selected labeling information and sentiment data and storing the information on a recording medium; means for training an artificial intelligence model using the stored labeling information and sentiment data; and means for improving the recognition accuracy for specific cultures and landscapes using the trained artificial intelligence model. This not only improves the accuracy of image recognition by leveraging the user's emotions, but also enhances the user experience by providing personalized tasks and enables effective training of the artificial intelligence model.

[0346] A "user" is any individual or organization that can use this system.

[0347] An "image recognition task" is a task in which a user selects the correct image from multiple displayed images based on specific criteria.

[0348] A "display device" is a component of a terminal used to present information visually, and its role is to show images or tasks to the user.

[0349] An "emotion analysis device" is a device that analyzes a user's facial expressions and voice to estimate and evaluate their emotional state.

[0350] "Labeling information" refers to information related to the identification and classification of images selected by the user based on the task.

[0351] "Emotional data" refers to information about the user's emotional state, including data acquired by an emotion analysis device.

[0352] A "recording medium" is a physical or electronic medium that can store information.

[0353] An "artificial intelligence model" is an algorithm or program designed to analyze large amounts of data and perform a specific task.

[0354] "Discrimination accuracy" is a measure of an artificial intelligence model's ability to accurately classify or recognize given data.

[0355] "Regionally specific culture" refers to a culture that exhibits characteristics such as traditions, customs, arts, and lifestyles within a particular geographical area.

[0356] This invention is a system that utilizes user emotion data to improve the accuracy of an artificial intelligence model through an image recognition task. The system comprises a user terminal, a server, an emotion analysis device, and an artificial intelligence model.

[0357] When a user accesses a specific webpage, the device presents an image recognition task via a display. The device incorporates an emotion analysis system that analyzes the user's facial expressions and voice while they work on the task, acquiring emotion data. Emotion analysis utilizes hardware such as a facial recognition camera and microphone.

[0358] When a user selects an image, the device sends the selected image's labeling information along with the acquired sentiment data to the server. The server stores this information on a recording medium and uses it to train an artificial intelligence model. The server analyzes the received labeling information and sentiment data to train the AI ​​model and improve its recognition accuracy.

[0359] For example, when a user is labeling images related to a festival, if the emotion analysis device detects the user's smile, the server adjusts the difficulty of the next task presented, offering a task that is more interesting to the user. This process allows the user to engage with the task more actively and also improves the model's recognition accuracy.

[0360] A concrete example of a prompt might be a text-based instruction such as, "What image is best to present when the user is smiling?" Based on this instruction, the generative AI model generates the optimal output according to the user's emotions. This improves the user experience and enables more sophisticated image recognition.

[0361] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0362] Step 1:

[0363] A user accesses a web page

[0364] The user accesses a specific webpage from a web browser using their device. At this point, the input is the user's access request. The output is the target webpage displayed on the device.

[0365] Step 2:

[0366] The device displays a CAPTCHA widget.

[0367] The device displays a CAPTCHA widget embedded in the webpage. This widget has an integrated sentiment analyzer and is ready for processing. The input is the data from the webpage, and the output is the CAPTCHA widget with the sentiment analyzer.

[0368] Step 3:

[0369] Users tackle image recognition challenges.

[0370] The user works on a presented image recognition task and selects images that meet the given criteria. At this point, the input is the presented set of images, and the output is information about the image selected by the user.

[0371] Step 4:

[0372] The device acquires emotional data.

[0373] The device acquires the user's facial expressions and voice using an emotion analysis device and stores them as emotion data. Specifically, it uses a camera and microphone to capture and analyze the user's emotions. The input is the user's real-time facial expressions and voice, and the output is the analyzed emotion data.

[0374] Step 5:

[0375] The device sends data to the server.

[0376] The device sends the image and sentiment data selected by the user to the server. A secure communication protocol is used for transmission. The input is the user's selected data and sentiment data, and the output is the completion of the data transmission to the server.

[0377] Step 6:

[0378] The server stores and processes the data.

[0379] The server stores the received data on a recording medium and analyzes the sentiment data and labeling information. Here, it processes the data into a dataset suitable for training an AI model. The input is the received data, and the output is the stored database entries and the training dataset.

[0380] Step 7:

[0381] The server trains the AI ​​model.

[0382] The server trains an AI model based on stored data to improve its recognition accuracy. The input is the training dataset, and the output is the updated AI model.

[0383] Step 8:

[0384] The system provides feedback

[0385] The system adjusts the difficulty level of the next task displayed and presents images and tasks that match the user's emotional state. The input is an updated AI model and the user's emotional information, and the output is a customized next task.

[0386] (Application Example 2)

[0387] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0388] Conventional image recognition systems failed to fully capture user interest and satisfaction because they provided uniform tasks without considering user emotions. Furthermore, the training of artificial intelligence models lacked sufficient use of emotional data to improve the user experience. This resulted in challenges to overall system performance and personalization.

[0389] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0390] In this invention, the server includes means for the user to label images based on a specific task and transmit them along with sentiment data; means for storing the received labeling information and sentiment data and using them to train an artificial intelligence model; and means for dynamically presenting images and tasks based on the user's preferences using the trained model. This enables a personalized experience that takes into account the user's emotions and improved accuracy in image recognition.

[0391] A "display device" is a device used to visually present information to users, and has the function of displaying images or tasks.

[0392] "Emotional data" refers to data that indicates the emotional state of a user, and includes information obtained from facial expressions, voice, and other sources.

[0393] "Labeling information" refers to the identification information of an image as determined by the user, and is data used to indicate the content and category of the image.

[0394] An "artificial intelligence model" is an algorithm trained using machine learning techniques and is used for tasks such as image recognition.

[0395] A "personalized user experience" is an experience optimized based on the individual preferences and emotions of the user, providing content and services that are tailored to each individual user.

[0396] The system of this invention presents users with a task of labeling images with emotional data, trains an artificial intelligence model using the collected data, and provides a personalized user experience. The implementation details of this system are described below.

[0397] The server displays images through a display device such as smart glasses or a smartphone, presenting the user with a specific task. Emotion analysis software, such as EmotionAPI, is used to acquire emotional data from the user's facial expressions and voice. This emotional data is captured in real time by hardware such as cameras and microphones and transmitted to the server.

[0398] The server uses the received labeling information and sentiment data to train an artificial intelligence model using machine learning libraries such as TensorFlow and Keras. This model is then used to improve the personalized user experience based on the collected data. For example, suppose a smile is detected when a user is browsing a specific item in a virtual fashion store. The system recognizes this as a positive reaction and recommends related products to maximize the user's purchase intent.

[0399] By utilizing generative AI models and using prompts such as, "Present similar items to products the user is interested in, based on the user's sentiment data, in real time," a richer and more customized user experience can be achieved.

[0400] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0401] Step 1:

[0402] The device presents the user with an image recognition task. The user views the task through a display device such as smart glasses or a smartphone. Multiple images related to the task are displayed at this time. The input consists of images and task information, and the output is the displayed image. The device retrieves task information from its internal memory and presents it to the user as a component.

[0403] Step 2:

[0404] The user selects an image according to the task. This selection is based on the user's decision, and the selected image is generated as labeling information. The input is the image selected by the user, and the output is the labeling information. Based on the user's selection, the terminal stores this information in cache memory.

[0405] Step 3:

[0406] The device launches emotion analysis software and acquires emotion data from the user's facial expressions and voice. Data is collected in real time via the camera and microphone and analyzed using the EmotionAPI. The input is the user's facial expressions and voice data, and the output is emotion data. The device temporarily stores the analyzed emotion data.

[0407] Step 4:

[0408] The terminal sends the acquired labeling information and sentiment data to the server. This data is used in the next analysis step. The input is the labeling information and sentiment data, and the output is the data sent to the server. The terminal sends information to the server as data packets over the network.

[0409] Step 5:

[0410] The server trains an artificial intelligence model using the received labeling and sentiment data. This model is built using TensorFlow and Keras, and optimizes itself by identifying anomalies based on the dataset. The input is the received labeling and sentiment data, and the output is the updated artificial intelligence model. The server allocates computing resources to perform high-speed data calculations.

[0411] Step 6:

[0412] The server uses a trained artificial intelligence model to generate personalized recommendations for the user. This selects content that is appropriate for the user's next action. The input is the trained AI model, and the output is the recommended content. The server leverages the generative AI model to adjust the prompt output in real time.

[0413] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0414] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0415] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0416] [Third Embodiment]

[0417] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0418] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0419] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0420] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0421] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0422] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0423] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0424] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0425] The specific processing program 56 is an example of a "program" relating to the technology of this 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 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.

[0426] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0427] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0428] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0429] This invention improves the accuracy of artificial intelligence models through a CAPTCHA system that presents users with image recognition challenges focused on specific cultures or landscapes. This system is implemented through an overall process including a user terminal, a server, and an artificial intelligence model.

[0430] When this system is running, if a user accesses a specific webpage, their device will display an image recognition task as a CAPTCHA widget. The displayed images are related to cultural content of a specific country or region, and the user is given a specific task. For example, based on the instruction "Select the image that contains Mount Fuji," the user will select the appropriate image from the presented images.

[0431] After selection, the terminal sends the result to the server. The server stores the received labeling information in a database. This stored data is used to train an artificial intelligence model. The server periodically retrieves the labeling data and uses it in the model's learning process to improve recognition accuracy. The server also evaluates the model's performance and adjusts parameters as needed to continuously optimize the model's performance.

[0432] As a concrete example, consider a scenario where a user labels images of traditional Japanese festivals. In this case, the user is presented with images containing festival features (e.g., taiko drums or portable shrines) and instructed to select the appropriate image. Once the user completes their selection, the information is sent to a server, and the stored data is used to train an AI model. Through this process, the AI ​​becomes able to recognize festival photos with greater accuracy.

[0433] This invention is expected to deepen AI's understanding of Japan's unique culture and landscapes, enabling diverse applications.

[0434] The following describes the processing flow.

[0435] Step 1:

[0436] When the device detects the user's access to a webpage, it displays a CAPTCHA widget. The device then sends a request for the image set to the CAPTCHA server.

[0437] Step 2:

[0438] When the server receives a request, it randomly selects images related to Japanese culture or landscapes. It then sends the selected images and the corresponding task (e.g., "Please select images that include Mount Fuji") to the device.

[0439] Step 3:

[0440] The user follows the instructions displayed on the device and selects the appropriate image from the presented options. Once the task is completed, the user presses the "Submit" button.

[0441] Step 4:

[0442] The device sends information about the image selected by the user to the server as data. This data includes the ID and timestamp of the selected image.

[0443] Step 5:

[0444] The server stores the received labeling data in a database. This data is later used to train an artificial intelligence model.

[0445] Step 6:

[0446] The server periodically retrieves labeling information from the database and uses this information to train the artificial intelligence model. Through the training process, the AI ​​model improves its image recognition accuracy.

[0447] Step 7:

[0448] The server evaluates the model and continuously optimizes its performance by adjusting parameters as needed. This allows the AI ​​to more accurately identify cultural elements unique to Japan.

[0449] (Example 1)

[0450] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0451] Conventional machine learning models exhibit high accuracy in general image recognition tasks, but their accuracy can decline when it comes to specific ethnic characteristics or natural landscapes. Efficient data collection and model training methods for improving the recognition accuracy of images containing such cultural and geographical features remain a challenge that has not yet been sufficiently developed.

[0452] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0453] In this invention, the server includes means for presenting an image recognition task displayed by a device including a display unit, which requires the user to label images based on a specific task; means for receiving and storing the labeling information selected by the user; and means for training a machine learning model using the stored labeling information. This makes it possible to improve the recognition accuracy of the machine learning model for specific ethnicities or natural landscapes.

[0454] "User" refers to a person who uses the system to label images based on the assigned task.

[0455] A "task" refers to the instructions presented to the user, in which they must select an image based on specific conditions.

[0456] A "display unit" refers to a part of the device used by a user to present an image recognition task, specifically the hardware or software used to display images.

[0457] "Device" refers to a platform that presents tasks to users, comprising hardware or software components including a display unit.

[0458] "Labeling information" refers to images and related data selected by users based on a specific task.

[0459] "Storage" refers to the process of securely saving received labeling information in a database or storage device.

[0460] A "machine learning model" refers to an algorithm or computational process designed to improve recognition accuracy for a specific task using collected data.

[0461] "Ethnicity" refers to characteristics and elements based on a particular culture or social background.

[0462] "Natural landscape" refers to natural visual features that include specific geographical or environmental characteristics.

[0463] "Recognition accuracy" refers to the ability of a machine learning model to accurately interpret input data and provide the correct results for a specific task.

[0464] This invention improves the accuracy of image recognition for specific ethnicities or natural landscapes using a system that includes a user, a terminal, and a server. Specifically, the system involves a series of processes in which the terminal presents an image recognition task to the user, the user sends the labeling information to the server, and the server uses this information to train a machine learning model.

[0465] The device can take the form of an internet-connected computer or smartphone. The device is equipped with a display unit for showing image recognition tasks to the user and communicates with a server to provide image data. The displayed images possess specific cultural or geographical characteristics.

[0466] The user selects an image based on a task presented on the device. For example, if the user receives the prompt "Select an image showing a traditional Japanese festival," they will select an image related to a Japanese festival. The selected labeling information is securely stored in a database.

[0467] The server uses the received labeling information to train a generative AI model, improving the model's recognition accuracy. This process utilizes machine learning algorithms, incorporates the labeling information as a new dataset, and periodically evaluates and adjusts the model's performance.

[0468] As a concrete example, when a user labels images related to festivals, the server feeds that data to a machine learning model, which is then trained to recognize festival images with high accuracy. An example of this prompt would be, "Please select images that include Japanese drums or portable shrines."

[0469] Through this system, machine learning models can more accurately recognize specific cultural and geographical objects, enabling their use in a variety of applications.

[0470] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0471] Step 1:

[0472] When a user accesses a webpage, the device launches a CAPTCHA widget. It receives the user's request as input and displays the CAPTCHA widget as output. Based on this access, the device sends a request to the server, requesting image data.

[0473] Step 2:

[0474] The server receives requests from terminals and sends image data necessary for the image recognition task to the terminals. It receives data containing the content of the requested task as input and sends image data to the terminals as output. The server retrieves a set of images from the database that match the specified task.

[0475] Step 3:

[0476] The terminal uses image data received from the server to present an image recognition task to the user. It takes the received image data as input and displays the task to the user as output. The terminal performs specific actions to display images and instructions on the screen to help the user visually understand the task.

[0477] Step 4:

[0478] The user selects the image that best matches the task from the presented images. As input, the user visually reviews multiple images, and as output, retrieves information about the selected image. The user then clicks or taps the image that is most appropriate based on the task.

[0479] Step 5:

[0480] The terminal sends the user's selection results to the server. It receives identification information for the selected image as input and sends that information to the server as output. The terminal packages the selection in the correct format and sends it to the server as an HTTP request.

[0481] Step 6:

[0482] The server stores the received labeling information in a database. It receives user-selected labeling information as input and stores that information in the database as output. The server then appropriately places the received data in the storage system and updates the index.

[0483] Step 7:

[0484] The server trains a generative AI model using stored labeling data. It takes stored data as input and generates updated parameters for the model as output. The server applies machine learning algorithms and feeds data back to the model to facilitate learning.

[0485] Step 8:

[0486] The server evaluates the model's performance and adjusts parameters as needed. It analyzes the output data of the trained model as input and creates an optimized model as output. The server then evaluates the model using test data to improve accuracy and adjusts the learning rate and other parameters.

[0487] (Application Example 1)

[0488] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0489] In recent years, there has been a growing need to accurately identify specific regional cultures and natural landscapes, but conventional artificial intelligence models struggle to fully understand such cultural elements. This problem hinders the provision of accurate information and the improvement of tourist experiences, particularly in the tourism and regional development industries. Therefore, there is a need to provide systems that can utilize cultural and geographical information more effectively.

[0490] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0491] In this invention, the server includes means for presenting an image recognition task, which is presented by an information processing terminal including a display device, in which the user is asked to label images based on a specific task; means for training a machine learning model using stored labeling information; and means for analyzing captured images and classifying whether the images have specific features. This makes it possible to recognize images related to specific regional cultures or natural landscapes with high accuracy and to provide cultural or geographical information based on the results.

[0492] An "information processing terminal" is an electronic device used by users to process data through an interface, and includes smartphones and tablets.

[0493] A "machine learning model" is an algorithm that automatically learns a specific task based on a vast amount of data, and is applied to image recognition and natural language processing.

[0494] A "display device" is a device used to present information visually, and displays and monitors fall into this category.

[0495] "Labeling information" refers to information that indicates the meaning associated with a particular image or data, and serves a role in classification and categorization.

[0496] "Cultural information" refers to traditions, customs, and historical knowledge unique to a particular region or society, and is an important element in tourism and education.

[0497] "Geographic information" refers to information that indicates the location, characteristics, and relationships of a specific geographical area, and is essential for spatial awareness and mapping services.

[0498] The system for carrying out this invention comprises an information processing terminal, a server, and a machine learning model. The information processing terminal is a terminal operated by the user, and a smartphone is an example of this. The terminal includes a display device that presents image recognition tasks related to local culture or natural landscapes based on a specific task. When the user labels the task displayed on the terminal, the result is sent to the server.

[0499] The server is responsible for storing the received labeling information in electronic storage. Furthermore, it uses this data to train a machine learning model. This machine learning model also functions as a generative AI model, improving its recognition accuracy for specific cultures and landscapes by learning from the labeled data. This model is built using TensorFlow.

[0500] When a user takes an image using their device, the captured image is analyzed to identify whether it contains specific cultural and geographical features. Based on the identification results, relevant cultural or geographical information is displayed on the device. For example, if traditional Japanese pottery is recognized, information about its history and manufacturing methods will be displayed in real time.

[0501] As a concrete example, when a tourist photographs an exhibit at a traditional festival, the server and a machine learning model can collaborate to provide relevant cultural information. An example of a prompt to input to the generative AI model is, "Use an image taken at the most recent festival and generate relevant cultural information." This allows the system to support a deeper cultural understanding and enhance understanding of the region.

[0502] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0503] Step 1:

[0504] A user takes an image using an information processing terminal. The input is a raw image obtained through the camera. This image may belong to a local culture or natural landscape. Here, the terminal uses camera hardware to capture and store the image data.

[0505] Step 2:

[0506] The device analyzes captured images and identifies whether they possess predefined features. The input is the image data obtained in step 1. This process uses an image analysis algorithm to resize the image to a specific size and perform data processing such as normalizing the color tone. The output is a feature list based on the analysis results.

[0507] Step 3:

[0508] Based on the analysis results, the device associates images with cultural or geographical information. The input is the feature list obtained in step 2. A generative AI model is used to generate a prompt sentence. This prompt sentence serves as a queue for retrieving relevant cultural information. The output is the prompt sentence and the associated information.

[0509] Step 4:

[0510] The terminal sends the prompt and related information to the server. The input is the prompt and related information generated in step 3. The terminal transfers the data to the server using the network interface. The output is the prompt and information stored on the server.

[0511] Step 5:

[0512] The server uses the received prompt message to utilize a generative AI model to generate detailed cultural information. The input is the data received in step 4. The server uses the generative AI model to perform data calculations to improve the accuracy and detail of the relevant information. The output is the generated detailed cultural information.

[0513] Step 6:

[0514] The server distributes the generated cultural information to the user's information processing terminal. The input is the detailed cultural information generated in step 5. The server uses the network interface to transmit the information to the terminal. The output is the cultural information displayed on the user's terminal.

[0515] Step 7:

[0516] The user reviews and understands the cultural information displayed on the device. The input is the cultural information received in step 6. The user views the displayed content and deepens their understanding of the local culture. The output is the user's improved knowledge.

[0517] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0518] This invention provides a system that improves the accuracy of an artificial intelligence model and the user experience through an image recognition task combined with an emotion engine that recognizes user emotions. This system includes a user terminal, a server, an emotion engine, and an artificial intelligence model.

[0519] When a user accesses a specific webpage, their device displays a CAPTCHA widget presenting an image recognition challenge. This widget incorporates an emotion engine that analyzes the user's emotions from their facial expressions and voice as they tackle the challenge. For example, if the user shows a positive response, the difficulty of the challenge may be adjusted to present a more satisfying challenge.

[0520] When a user selects an image, the device sends emotion data along with the selection to the server. The server stores the received data in a database and uses the emotion data, along with labeling information, to train an AI model. This allows the AI ​​model to understand changes in the user's emotions and improve the accuracy of image recognition accordingly.

[0521] For example, when a user labels images of traditional Japanese festivals, the system detects smiles and expressions of surprise to engage the user, and then presents related images and tasks one after another. Through this process, active user participation is encouraged, the quality of the labeling improves, and the recognition accuracy of the AI ​​model increases.

[0522] Furthermore, emotional data is used not only for training models but also for providing personalized user experiences. Thus, this embodiment of the present invention aims to leverage an emotional engine to extend the capabilities of AI technology and improve the user experience.

[0523] The following describes the processing flow.

[0524] Step 1:

[0525] When a user accesses a webpage, their device displays a CAPTCHA widget. This widget presents the user with questions related to Japanese culture and landscapes.

[0526] Step 2:

[0527] As soon as the user begins working on a task, the device activates an emotion engine that analyzes the user's emotions in real time from their facial expressions and voice. This information is used to adjust the difficulty of the task and improve the user experience.

[0528] Step 3:

[0529] The user selects the correct image from the presented images. During the selection process, the emotion engine records the user's facial expressions and reactions and prepares to send this data to the server.

[0530] Step 4:

[0531] The device sends the labeling information selected by the user and the emotion data acquired by the emotion engine to the server. This data includes the image ID and the user's emotional state.

[0532] Step 5:

[0533] The server stores the received labeling and sentiment data in a database. This information is later used as a training dataset for the AI ​​model.

[0534] Step 6:

[0535] The server periodically retrieves labeling and sentiment data from the database and uses it to train an artificial intelligence model. The training process takes into account the user's emotional state and aims to improve the accuracy of recognizing specific cultures and landscapes.

[0536] Step 7:

[0537] The server evaluates the trained model and optimizes its performance by adjusting parameters as needed. This optimization further improves accuracy by leveraging user sentiment data.

[0538] (Example 2)

[0539] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0540] Conventional image recognition systems only require simple labeling without considering user emotions, making it difficult to improve the user experience or maximize the recognition accuracy of artificial intelligence models. Furthermore, there is a lack of mechanisms to utilize emotional data when improving the accuracy of identifying specific cultures or landscapes.

[0541] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0542] In this invention, the server includes means for receiving user-selected labeling information and sentiment data and storing the information on a recording medium; means for training an artificial intelligence model using the stored labeling information and sentiment data; and means for improving the recognition accuracy for specific cultures and landscapes using the trained artificial intelligence model. This not only improves the accuracy of image recognition by leveraging the user's emotions, but also enhances the user experience by providing personalized tasks and enables effective training of the artificial intelligence model.

[0543] A "user" is any individual or organization that can use this system.

[0544] An "image recognition task" is a task in which a user selects the correct image from multiple displayed images based on specific criteria.

[0545] A "display device" is a component of a terminal used to present information visually, and its role is to show images or tasks to the user.

[0546] An "emotion analysis device" is a device that analyzes a user's facial expressions and voice to estimate and evaluate their emotional state.

[0547] "Labeling information" refers to information related to the identification and classification of images selected by the user based on the task.

[0548] "Emotional data" refers to information about the user's emotional state, including data acquired by an emotion analysis device.

[0549] A "recording medium" is a physical or electronic medium that can store information.

[0550] An "artificial intelligence model" is an algorithm or program designed to analyze large amounts of data and perform a specific task.

[0551] "Discrimination accuracy" is a measure of an artificial intelligence model's ability to accurately classify or recognize given data.

[0552] "Regionally specific culture" refers to a culture that exhibits characteristics such as traditions, customs, arts, and lifestyles within a particular geographical area.

[0553] This invention is a system that utilizes user emotion data to improve the accuracy of an artificial intelligence model through an image recognition task. The system comprises a user terminal, a server, an emotion analysis device, and an artificial intelligence model.

[0554] When a user accesses a specific webpage, the device presents an image recognition task via a display. The device incorporates an emotion analysis system that analyzes the user's facial expressions and voice while they work on the task, acquiring emotion data. Emotion analysis utilizes hardware such as a facial recognition camera and microphone.

[0555] When a user selects an image, the device sends the selected image's labeling information along with the acquired sentiment data to the server. The server stores this information on a recording medium and uses it to train an artificial intelligence model. The server analyzes the received labeling information and sentiment data to train the AI ​​model and improve its recognition accuracy.

[0556] For example, when a user is labeling images related to a festival, if the emotion analysis device detects the user's smile, the server adjusts the difficulty of the next task presented, offering a task that is more interesting to the user. This process allows the user to engage with the task more actively and also improves the model's recognition accuracy.

[0557] A concrete example of a prompt might be a text-based instruction such as, "What image is best to present when the user is smiling?" Based on this instruction, the generative AI model generates the optimal output according to the user's emotions. This improves the user experience and enables more sophisticated image recognition.

[0558] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0559] Step 1:

[0560] A user accesses a web page

[0561] The user accesses a specific webpage from a web browser using their device. At this point, the input is the user's access request. The output is the target webpage displayed on the device.

[0562] Step 2:

[0563] The device displays a CAPTCHA widget.

[0564] The device displays a CAPTCHA widget embedded in the webpage. This widget has an integrated sentiment analyzer and is ready for processing. The input is the data from the webpage, and the output is the CAPTCHA widget with the sentiment analyzer.

[0565] Step 3:

[0566] Users tackle image recognition challenges.

[0567] The user works on a presented image recognition task and selects images that meet the given criteria. At this point, the input is the presented set of images, and the output is information about the image selected by the user.

[0568] Step 4:

[0569] The device acquires emotional data.

[0570] The device acquires the user's facial expressions and voice using an emotion analysis device and stores them as emotion data. Specifically, it uses a camera and microphone to capture and analyze the user's emotions. The input is the user's real-time facial expressions and voice, and the output is the analyzed emotion data.

[0571] Step 5:

[0572] The device sends data to the server.

[0573] The device sends the image and sentiment data selected by the user to the server. A secure communication protocol is used for transmission. The input is the user's selected data and sentiment data, and the output is the completion of the data transmission to the server.

[0574] Step 6:

[0575] The server stores and processes the data.

[0576] The server stores the received data on a recording medium and analyzes the sentiment data and labeling information. Here, it processes the data into a dataset suitable for training an AI model. The input is the received data, and the output is the stored database entries and the training dataset.

[0577] Step 7:

[0578] The server trains the AI ​​model.

[0579] The server trains an AI model based on stored data to improve its recognition accuracy. The input is the training dataset, and the output is the updated AI model.

[0580] Step 8:

[0581] The system provides feedback

[0582] The system adjusts the difficulty level of the next task displayed and presents images and tasks that match the user's emotional state. The input is an updated AI model and the user's emotional information, and the output is a customized next task.

[0583] (Application Example 2)

[0584] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0585] Conventional image recognition systems failed to fully capture user interest and satisfaction because they provided uniform tasks without considering user emotions. Furthermore, the training of artificial intelligence models lacked sufficient use of emotional data to improve the user experience. This resulted in challenges to overall system performance and personalization.

[0586] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0587] In this invention, the server includes means for the user to label images based on a specific task and transmit them along with sentiment data; means for storing the received labeling information and sentiment data and using them to train an artificial intelligence model; and means for dynamically presenting images and tasks based on the user's preferences using the trained model. This enables a personalized experience that takes into account the user's emotions and improved accuracy in image recognition.

[0588] A "display device" is a device used to visually present information to users, and has the function of displaying images or tasks.

[0589] "Emotional data" refers to data that indicates the emotional state of a user, and includes information obtained from facial expressions, voice, and other sources.

[0590] "Labeling information" refers to the identification information of an image as determined by the user, and is data used to indicate the content and category of the image.

[0591] An "artificial intelligence model" is an algorithm trained using machine learning techniques and is used for tasks such as image recognition.

[0592] A "personalized user experience" is an experience optimized based on the individual preferences and emotions of the user, providing content and services that are tailored to each individual user.

[0593] The system of this invention presents users with a task of labeling images with emotional data, trains an artificial intelligence model using the collected data, and provides a personalized user experience. The implementation details of this system are described below.

[0594] The server displays images through a display device such as smart glasses or a smartphone, presenting the user with a specific task. Emotion analysis software, such as EmotionAPI, is used to acquire emotional data from the user's facial expressions and voice. This emotional data is captured in real time by hardware such as cameras and microphones and transmitted to the server.

[0595] The server uses the received labeling information and sentiment data to train an artificial intelligence model using machine learning libraries such as TensorFlow and Keras. This model is then used to improve the personalized user experience based on the collected data. For example, suppose a smile is detected when a user is browsing a specific item in a virtual fashion store. The system recognizes this as a positive reaction and recommends related products to maximize the user's purchase intent.

[0596] By utilizing generative AI models and using prompts such as, "Present similar items to products the user is interested in, based on the user's sentiment data, in real time," a richer and more customized user experience can be achieved.

[0597] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0598] Step 1:

[0599] The device presents the user with an image recognition task. The user views the task through a display device such as smart glasses or a smartphone. Multiple images related to the task are displayed at this time. The input consists of images and task information, and the output is the displayed image. The device retrieves task information from its internal memory and presents it to the user as a component.

[0600] Step 2:

[0601] The user selects an image according to the task. This selection is based on the user's decision, and the selected image is generated as labeling information. The input is the image selected by the user, and the output is the labeling information. Based on the user's selection, the terminal stores this information in cache memory.

[0602] Step 3:

[0603] The device launches emotion analysis software and acquires emotion data from the user's facial expressions and voice. Data is collected in real time via the camera and microphone and analyzed using the EmotionAPI. The input is the user's facial expressions and voice data, and the output is emotion data. The device temporarily stores the analyzed emotion data.

[0604] Step 4:

[0605] The terminal sends the acquired labeling information and sentiment data to the server. This data is used in the next analysis step. The input is the labeling information and sentiment data, and the output is the data sent to the server. The terminal sends information to the server as data packets over the network.

[0606] Step 5:

[0607] The server trains an artificial intelligence model using the received labeling and sentiment data. This model is built using TensorFlow and Keras, and optimizes itself by identifying anomalies based on the dataset. The input is the received labeling and sentiment data, and the output is the updated artificial intelligence model. The server allocates computing resources to perform high-speed data calculations.

[0608] Step 6:

[0609] The server uses a trained artificial intelligence model to generate personalized recommendations for the user. This selects content that is appropriate for the user's next action. The input is the trained AI model, and the output is the recommended content. The server leverages the generative AI model to adjust the prompt output in real time.

[0610] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0611] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0612] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0613] [Fourth Embodiment]

[0614] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0615] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0616] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0617] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0618] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0619] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0620] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0621] The controlled 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0622] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0623] The specific processing program 56 is an example of a "program" relating to the technology of this 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 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.

[0624] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0625] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0626] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0627] This invention improves the accuracy of artificial intelligence models through a CAPTCHA system that presents users with image recognition challenges focused on specific cultures or landscapes. This system is implemented through an overall process including a user terminal, a server, and an artificial intelligence model.

[0628] When this system is running, if a user accesses a specific webpage, their device will display an image recognition task as a CAPTCHA widget. The displayed images are related to cultural content of a specific country or region, and the user is given a specific task. For example, based on the instruction "Select the image that contains Mount Fuji," the user will select the appropriate image from the presented images.

[0629] After selection, the terminal sends the result to the server. The server stores the received labeling information in a database. This stored data is used to train an artificial intelligence model. The server periodically retrieves the labeling data and uses it in the model's learning process to improve recognition accuracy. The server also evaluates the model's performance and adjusts parameters as needed to continuously optimize the model's performance.

[0630] As a concrete example, consider a scenario where a user labels images of traditional Japanese festivals. In this case, the user is presented with images containing festival features (e.g., taiko drums or portable shrines) and instructed to select the appropriate image. Once the user completes their selection, the information is sent to a server, and the stored data is used to train an AI model. Through this process, the AI ​​becomes able to recognize festival photos with greater accuracy.

[0631] This invention is expected to deepen AI's understanding of Japan's unique culture and landscapes, enabling diverse applications.

[0632] The following describes the processing flow.

[0633] Step 1:

[0634] When the device detects the user's access to a webpage, it displays a CAPTCHA widget. The device then sends a request for the image set to the CAPTCHA server.

[0635] Step 2:

[0636] When the server receives a request, it randomly selects images related to Japanese culture or landscapes. It then sends the selected images and the corresponding task (e.g., "Please select images that include Mount Fuji") to the device.

[0637] Step 3:

[0638] The user follows the instructions displayed on the device and selects the appropriate image from the presented options. Once the task is completed, the user presses the "Submit" button.

[0639] Step 4:

[0640] The device sends information about the image selected by the user to the server as data. This data includes the ID and timestamp of the selected image.

[0641] Step 5:

[0642] The server stores the received labeling data in a database. This data is later used to train an artificial intelligence model.

[0643] Step 6:

[0644] The server periodically retrieves labeling information from the database and uses this information to train the artificial intelligence model. Through the training process, the AI ​​model improves its image recognition accuracy.

[0645] Step 7:

[0646] The server evaluates the model and continuously optimizes its performance by adjusting parameters as needed. This allows the AI ​​to more accurately identify cultural elements unique to Japan.

[0647] (Example 1)

[0648] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0649] Conventional machine learning models exhibit high accuracy in general image recognition tasks, but their accuracy can decline when it comes to specific ethnic characteristics or natural landscapes. Efficient data collection and model training methods for improving the recognition accuracy of images containing such cultural and geographical features remain a challenge that has not yet been sufficiently developed.

[0650] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0651] In this invention, the server includes means for presenting an image recognition task displayed by a device including a display unit, which requires the user to label images based on a specific task; means for receiving and storing the labeling information selected by the user; and means for training a machine learning model using the stored labeling information. This makes it possible to improve the recognition accuracy of the machine learning model for specific ethnicities or natural landscapes.

[0652] "User" refers to a person who uses the system to label images based on the assigned task.

[0653] A "task" refers to the instructions presented to the user, in which they must select an image based on specific conditions.

[0654] A "display unit" refers to a part of the device used by a user to present an image recognition task, specifically the hardware or software used to display images.

[0655] "Device" refers to a platform that presents tasks to users, comprising hardware or software components including a display unit.

[0656] "Labeling information" refers to images and related data selected by users based on a specific task.

[0657] "Storage" refers to the process of securely saving received labeling information in a database or storage device.

[0658] A "machine learning model" refers to an algorithm or computational process designed to improve recognition accuracy for a specific task using collected data.

[0659] "Ethnicity" refers to characteristics and elements based on a particular culture or social background.

[0660] "Natural landscape" refers to natural visual features that include specific geographical or environmental characteristics.

[0661] "Recognition accuracy" refers to the ability of a machine learning model to accurately interpret input data and provide the correct results for a specific task.

[0662] This invention improves the accuracy of image recognition for specific ethnicities or natural landscapes using a system that includes a user, a terminal, and a server. Specifically, the system involves a series of processes in which the terminal presents an image recognition task to the user, the user sends the labeling information to the server, and the server uses this information to train a machine learning model.

[0663] The device can take the form of an internet-connected computer or smartphone. The device is equipped with a display unit for showing image recognition tasks to the user and communicates with a server to provide image data. The displayed images possess specific cultural or geographical characteristics.

[0664] The user selects an image based on a task presented on the device. For example, if the user receives the prompt "Select an image showing a traditional Japanese festival," they will select an image related to a Japanese festival. The selected labeling information is securely stored in a database.

[0665] The server uses the received labeling information to train a generative AI model, improving the model's recognition accuracy. This process utilizes machine learning algorithms, incorporates the labeling information as a new dataset, and periodically evaluates and adjusts the model's performance.

[0666] As a concrete example, when a user labels images related to festivals, the server feeds that data to a machine learning model, which is then trained to recognize festival images with high accuracy. An example of this prompt would be, "Please select images that include Japanese drums or portable shrines."

[0667] Through this system, machine learning models can more accurately recognize specific cultural and geographical objects, enabling their use in a variety of applications.

[0668] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0669] Step 1:

[0670] When a user accesses a webpage, the device launches a CAPTCHA widget. It receives the user's request as input and displays the CAPTCHA widget as output. Based on this access, the device sends a request to the server, requesting image data.

[0671] Step 2:

[0672] The server receives requests from terminals and sends image data necessary for the image recognition task to the terminals. It receives data containing the content of the requested task as input and sends image data to the terminals as output. The server retrieves a set of images from the database that match the specified task.

[0673] Step 3:

[0674] The terminal uses image data received from the server to present an image recognition task to the user. It takes the received image data as input and displays the task to the user as output. The terminal performs specific actions to display images and instructions on the screen to help the user visually understand the task.

[0675] Step 4:

[0676] The user selects the image that best matches the task from the presented images. As input, the user visually reviews multiple images, and as output, retrieves information about the selected image. The user then clicks or taps the image that is most appropriate based on the task.

[0677] Step 5:

[0678] The terminal sends the user's selection results to the server. It receives identification information for the selected image as input and sends that information to the server as output. The terminal packages the selection in the correct format and sends it to the server as an HTTP request.

[0679] Step 6:

[0680] The server stores the received labeling information in a database. It receives user-selected labeling information as input and stores that information in the database as output. The server then appropriately places the received data in the storage system and updates the index.

[0681] Step 7:

[0682] The server trains a generative AI model using stored labeling data. It takes stored data as input and generates updated parameters for the model as output. The server applies machine learning algorithms and feeds data back to the model to facilitate learning.

[0683] Step 8:

[0684] The server evaluates the model's performance and adjusts parameters as needed. It analyzes the output data of the trained model as input and creates an optimized model as output. The server then evaluates the model using test data to improve accuracy and adjusts the learning rate and other parameters.

[0685] (Application Example 1)

[0686] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0687] In recent years, there has been a growing need to accurately identify specific regional cultures and natural landscapes, but conventional artificial intelligence models struggle to fully understand such cultural elements. This problem hinders the provision of accurate information and the improvement of tourist experiences, particularly in the tourism and regional development industries. Therefore, there is a need to provide systems that can utilize cultural and geographical information more effectively.

[0688] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0689] In this invention, the server includes means for presenting an image recognition task, which is presented by an information processing terminal including a display device, in which the user is asked to label images based on a specific task; means for training a machine learning model using stored labeling information; and means for analyzing captured images and classifying whether the images have specific features. This makes it possible to recognize images related to specific regional cultures or natural landscapes with high accuracy and to provide cultural or geographical information based on the results.

[0690] An "information processing terminal" is an electronic device used by users to process data through an interface, and includes smartphones and tablets.

[0691] A "machine learning model" is an algorithm that automatically learns a specific task based on a vast amount of data, and is applied to image recognition and natural language processing.

[0692] A "display device" is a device used to present information visually, and displays and monitors fall into this category.

[0693] "Labeling information" refers to information that indicates the meaning associated with a particular image or data, and serves a role in classification and categorization.

[0694] "Cultural information" refers to traditions, customs, and historical knowledge unique to a particular region or society, and is an important element in tourism and education.

[0695] "Geographic information" refers to information that indicates the location, characteristics, and relationships of a specific geographical area, and is essential for spatial awareness and mapping services.

[0696] The system for carrying out this invention comprises an information processing terminal, a server, and a machine learning model. The information processing terminal is a terminal operated by the user, and a smartphone is an example of this. The terminal includes a display device that presents image recognition tasks related to local culture or natural landscapes based on a specific task. When the user labels the task displayed on the terminal, the result is sent to the server.

[0697] The server is responsible for storing the received labeling information in electronic storage. Furthermore, it uses this data to train a machine learning model. This machine learning model also functions as a generative AI model, improving its recognition accuracy for specific cultures and landscapes by learning from the labeled data. This model is built using TensorFlow.

[0698] When a user takes an image using their device, the captured image is analyzed to identify whether it contains specific cultural and geographical features. Based on the identification results, relevant cultural or geographical information is displayed on the device. For example, if traditional Japanese pottery is recognized, information about its history and manufacturing methods will be displayed in real time.

[0699] As a concrete example, when a tourist photographs an exhibit at a traditional festival, the server and a machine learning model can collaborate to provide relevant cultural information. An example of a prompt to input to the generative AI model is, "Use an image taken at the most recent festival and generate relevant cultural information." This allows the system to support a deeper cultural understanding and enhance understanding of the region.

[0700] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0701] Step 1:

[0702] A user takes an image using an information processing terminal. The input is a raw image obtained through the camera. This image may belong to a local culture or natural landscape. Here, the terminal uses camera hardware to capture and store the image data.

[0703] Step 2:

[0704] The device analyzes captured images and identifies whether they possess predefined features. The input is the image data obtained in step 1. This process uses an image analysis algorithm to resize the image to a specific size and perform data processing such as normalizing the color tone. The output is a feature list based on the analysis results.

[0705] Step 3:

[0706] Based on the analysis results, the device associates images with cultural or geographical information. The input is the feature list obtained in step 2. A generative AI model is used to generate a prompt sentence. This prompt sentence serves as a queue for retrieving relevant cultural information. The output is the prompt sentence and the associated information.

[0707] Step 4:

[0708] The terminal sends the prompt and related information to the server. The input is the prompt and related information generated in step 3. The terminal transfers the data to the server using the network interface. The output is the prompt and information stored on the server.

[0709] Step 5:

[0710] The server uses the received prompt message to utilize a generative AI model to generate detailed cultural information. The input is the data received in step 4. The server uses the generative AI model to perform data calculations to improve the accuracy and detail of the relevant information. The output is the generated detailed cultural information.

[0711] Step 6:

[0712] The server distributes the generated cultural information to the user's information processing terminal. The input is the detailed cultural information generated in step 5. The server uses the network interface to transmit the information to the terminal. The output is the cultural information displayed on the user's terminal.

[0713] Step 7:

[0714] The user reviews and understands the cultural information displayed on the device. The input is the cultural information received in step 6. The user views the displayed content and deepens their understanding of the local culture. The output is the user's improved knowledge.

[0715] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0716] This invention provides a system that improves the accuracy of an artificial intelligence model and the user experience through an image recognition task combined with an emotion engine that recognizes user emotions. This system includes a user terminal, a server, an emotion engine, and an artificial intelligence model.

[0717] When a user accesses a specific webpage, their device displays a CAPTCHA widget presenting an image recognition challenge. This widget incorporates an emotion engine that analyzes the user's emotions from their facial expressions and voice as they tackle the challenge. For example, if the user shows a positive response, the difficulty of the challenge may be adjusted to present a more satisfying challenge.

[0718] When a user selects an image, the device sends emotion data along with the selection to the server. The server stores the received data in a database and uses the emotion data, along with labeling information, to train an AI model. This allows the AI ​​model to understand changes in the user's emotions and improve the accuracy of image recognition accordingly.

[0719] For example, when a user labels images of traditional Japanese festivals, the system detects smiles and expressions of surprise to engage the user, and then presents related images and tasks one after another. Through this process, active user participation is encouraged, the quality of the labeling improves, and the recognition accuracy of the AI ​​model increases.

[0720] Furthermore, emotional data is used not only for training models but also for providing personalized user experiences. Thus, this embodiment of the present invention aims to leverage an emotional engine to extend the capabilities of AI technology and improve the user experience.

[0721] The following describes the processing flow.

[0722] Step 1:

[0723] When a user accesses a webpage, their device displays a CAPTCHA widget. This widget presents the user with questions related to Japanese culture and landscapes.

[0724] Step 2:

[0725] As soon as the user begins working on a task, the device activates an emotion engine that analyzes the user's emotions in real time from their facial expressions and voice. This information is used to adjust the difficulty of the task and improve the user experience.

[0726] Step 3:

[0727] The user selects the correct image from the presented images. During the selection process, the emotion engine records the user's facial expressions and reactions and prepares to send this data to the server.

[0728] Step 4:

[0729] The device sends the labeling information selected by the user and the emotion data acquired by the emotion engine to the server. This data includes the image ID and the user's emotional state.

[0730] Step 5:

[0731] The server stores the received labeling and sentiment data in a database. This information is later used as a training dataset for the AI ​​model.

[0732] Step 6:

[0733] The server periodically retrieves labeling and sentiment data from the database and uses it to train an artificial intelligence model. The training process takes into account the user's emotional state and aims to improve the accuracy of recognizing specific cultures and landscapes.

[0734] Step 7:

[0735] The server evaluates the trained model and optimizes its performance by adjusting parameters as needed. This optimization further improves accuracy by leveraging user sentiment data.

[0736] (Example 2)

[0737] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0738] Conventional image recognition systems only require simple labeling without considering user emotions, making it difficult to improve the user experience or maximize the recognition accuracy of artificial intelligence models. Furthermore, there is a lack of mechanisms to utilize emotional data when improving the accuracy of identifying specific cultures or landscapes.

[0739] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0740] In this invention, the server includes means for receiving user-selected labeling information and sentiment data and storing the information on a recording medium; means for training an artificial intelligence model using the stored labeling information and sentiment data; and means for improving the recognition accuracy for specific cultures and landscapes using the trained artificial intelligence model. This not only improves the accuracy of image recognition by leveraging the user's emotions, but also enhances the user experience by providing personalized tasks and enables effective training of the artificial intelligence model.

[0741] A "user" is any individual or organization that can use this system.

[0742] An "image recognition task" is a task in which a user selects the correct image from multiple displayed images based on specific criteria.

[0743] A "display device" is a component of a terminal used to present information visually, and its role is to show images or tasks to the user.

[0744] An "emotion analysis device" is a device that analyzes a user's facial expressions and voice to estimate and evaluate their emotional state.

[0745] "Labeling information" refers to information related to the identification and classification of images selected by the user based on the task.

[0746] "Emotional data" refers to information about the user's emotional state, including data acquired by an emotion analysis device.

[0747] A "recording medium" is a physical or electronic medium that can store information.

[0748] An "artificial intelligence model" is an algorithm or program designed to analyze large amounts of data and perform a specific task.

[0749] "Discrimination accuracy" is a measure of an artificial intelligence model's ability to accurately classify or recognize given data.

[0750] "Regionally specific culture" refers to a culture that exhibits characteristics such as traditions, customs, arts, and lifestyles within a particular geographical area.

[0751] This invention is a system that utilizes user emotion data to improve the accuracy of an artificial intelligence model through an image recognition task. The system comprises a user terminal, a server, an emotion analysis device, and an artificial intelligence model.

[0752] When a user accesses a specific webpage, the device presents an image recognition task via a display. The device incorporates an emotion analysis system that analyzes the user's facial expressions and voice while they work on the task, acquiring emotion data. Emotion analysis utilizes hardware such as a facial recognition camera and microphone.

[0753] When a user selects an image, the device sends the selected image's labeling information along with the acquired sentiment data to the server. The server stores this information on a recording medium and uses it to train an artificial intelligence model. The server analyzes the received labeling information and sentiment data to train the AI ​​model and improve its recognition accuracy.

[0754] For example, when a user is labeling images related to a festival, if the emotion analysis device detects the user's smile, the server adjusts the difficulty of the next task presented, offering a task that is more interesting to the user. This process allows the user to engage with the task more actively and also improves the model's recognition accuracy.

[0755] A concrete example of a prompt might be a text-based instruction such as, "What image is best to present when the user is smiling?" Based on this instruction, the generative AI model generates the optimal output according to the user's emotions. This improves the user experience and enables more sophisticated image recognition.

[0756] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0757] Step 1:

[0758] A user accesses a web page

[0759] The user accesses a specific webpage from a web browser using their device. At this point, the input is the user's access request. The output is the target webpage displayed on the device.

[0760] Step 2:

[0761] The device displays a CAPTCHA widget.

[0762] The device displays a CAPTCHA widget embedded in the webpage. This widget has an integrated sentiment analyzer and is ready for processing. The input is the data from the webpage, and the output is the CAPTCHA widget with the sentiment analyzer.

[0763] Step 3:

[0764] Users tackle image recognition challenges.

[0765] The user works on a presented image recognition task and selects images that meet the given criteria. At this point, the input is the presented set of images, and the output is information about the image selected by the user.

[0766] Step 4:

[0767] The device acquires emotional data.

[0768] The device acquires the user's facial expressions and voice using an emotion analysis device and stores them as emotion data. Specifically, it uses a camera and microphone to capture and analyze the user's emotions. The input is the user's real-time facial expressions and voice, and the output is the analyzed emotion data.

[0769] Step 5:

[0770] The device sends data to the server.

[0771] The device sends the image and sentiment data selected by the user to the server. A secure communication protocol is used for transmission. The input is the user's selected data and sentiment data, and the output is the completion of the data transmission to the server.

[0772] Step 6:

[0773] The server stores and processes the data.

[0774] The server stores the received data on a recording medium and analyzes the sentiment data and labeling information. Here, it processes the data into a dataset suitable for training an AI model. The input is the received data, and the output is the stored database entries and the training dataset.

[0775] Step 7:

[0776] The server trains the AI ​​model.

[0777] The server trains an AI model based on stored data to improve its recognition accuracy. The input is the training dataset, and the output is the updated AI model.

[0778] Step 8:

[0779] The system provides feedback

[0780] The system adjusts the difficulty level of the next task displayed and presents images and tasks that match the user's emotional state. The input is an updated AI model and the user's emotional information, and the output is a customized next task.

[0781] (Application Example 2)

[0782] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0783] Conventional image recognition systems failed to fully capture user interest and satisfaction because they provided uniform tasks without considering user emotions. Furthermore, the training of artificial intelligence models lacked sufficient use of emotional data to improve the user experience. This resulted in challenges to overall system performance and personalization.

[0784] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0785] In this invention, the server includes means for the user to label images based on a specific task and transmit them along with sentiment data; means for storing the received labeling information and sentiment data and using them to train an artificial intelligence model; and means for dynamically presenting images and tasks based on the user's preferences using the trained model. This enables a personalized experience that takes into account the user's emotions and improved accuracy in image recognition.

[0786] A "display device" is a device used to visually present information to users, and has the function of displaying images or tasks.

[0787] "Emotional data" refers to data that indicates the emotional state of a user, and includes information obtained from facial expressions, voice, and other sources.

[0788] "Labeling information" refers to the identification information of an image as determined by the user, and is data used to indicate the content and category of the image.

[0789] An "artificial intelligence model" is an algorithm trained using machine learning techniques and is used for tasks such as image recognition.

[0790] A "personalized user experience" is an experience optimized based on the individual preferences and emotions of the user, providing content and services that are tailored to each individual user.

[0791] The system of this invention presents users with a task of labeling images with emotional data, trains an artificial intelligence model using the collected data, and provides a personalized user experience. The implementation details of this system are described below.

[0792] The server displays images through a display device such as smart glasses or a smartphone, presenting the user with a specific task. Emotion analysis software, such as EmotionAPI, is used to acquire emotional data from the user's facial expressions and voice. This emotional data is captured in real time by hardware such as cameras and microphones and transmitted to the server.

[0793] The server uses the received labeling information and sentiment data to train an artificial intelligence model using machine learning libraries such as TensorFlow and Keras. This model is then used to improve the personalized user experience based on the collected data. For example, suppose a smile is detected when a user is browsing a specific item in a virtual fashion store. The system recognizes this as a positive reaction and recommends related products to maximize the user's purchase intent.

[0794] By utilizing generative AI models and using prompts such as, "Present similar items to products the user is interested in, based on the user's sentiment data, in real time," a richer and more customized user experience can be achieved.

[0795] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0796] Step 1:

[0797] The device presents the user with an image recognition task. The user views the task through a display device such as smart glasses or a smartphone. Multiple images related to the task are displayed at this time. The input consists of images and task information, and the output is the displayed image. The device retrieves task information from its internal memory and presents it to the user as a component.

[0798] Step 2:

[0799] The user selects an image according to the task. This selection is based on the user's decision, and the selected image is generated as labeling information. The input is the image selected by the user, and the output is the labeling information. Based on the user's selection, the terminal stores this information in cache memory.

[0800] Step 3:

[0801] The device launches emotion analysis software and acquires emotion data from the user's facial expressions and voice. Data is collected in real time via the camera and microphone and analyzed using the EmotionAPI. The input is the user's facial expressions and voice data, and the output is emotion data. The device temporarily stores the analyzed emotion data.

[0802] Step 4:

[0803] The terminal sends the acquired labeling information and sentiment data to the server. This data is used in the next analysis step. The input is the labeling information and sentiment data, and the output is the data sent to the server. The terminal sends information to the server as data packets over the network.

[0804] Step 5:

[0805] The server trains an artificial intelligence model using the received labeling and sentiment data. This model is built using TensorFlow and Keras, and optimizes itself by identifying anomalies based on the dataset. The input is the received labeling and sentiment data, and the output is the updated artificial intelligence model. The server allocates computing resources to perform high-speed data calculations.

[0806] Step 6:

[0807] The server uses a trained artificial intelligence model to generate personalized recommendations for the user. This selects content that is appropriate for the user's next action. The input is the trained AI model, and the output is the recommended content. The server leverages the generative AI model to adjust the prompt output in real time.

[0808] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0809] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0810] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0811] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0812] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0813] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0814] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0815] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0816] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0817] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0818] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0819] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0820] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0821] 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.

[0822] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0823] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0824] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0825] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0826] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0827] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0828] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[0829] The following is further disclosed regarding the embodiments described above.

[0830] (Claim 1)

[0831] [Means for presenting an image recognition task displayed on a terminal including display units, which requires the user to label images based on a specific task,

[0832] [Means for receiving the labeling information selected by the user and storing the labeling information,

[0833] [Methods for training an artificial intelligence model using stored labeling information,

[0834] [Methods for improving the accuracy of recognition of specific cultures and landscapes using trained artificial intelligence models,

[0835] A system that includes this.

[0836] (Claim 2)

[0837] [Means for analyzing the stored labeling information and evaluating the performance of the artificial intelligence model,

[0838] [The system according to claim 1, which adjusts the parameters of the artificial intelligence model based on the evaluation.

[0839] (Claim 3)

[0840] [The system according to claim 1, wherein the display unit is configured to display an image relating to the culture of a country.

[0841] "Example 1"

[0842] (Claim 1)

[0843] [Means for presenting an image recognition task displayed by a device including a display unit, which requires the user to label images based on a specific task,

[0844] [Means for receiving the labeling information selected by the user and storing the labeling information,

[0845] [Methods for training machine learning models using stored labeling information,

[0846] [Methods for improving the accuracy of recognition of specific ethnicities or natural landscapes using trained machine learning models,

[0847] A system that includes this.

[0848] (Claim 2)

[0849] [Means for analyzing the stored labeling information and evaluating the performance of the machine learning model,

[0850] [The system according to claim 1, which modifies the characteristics of a machine learning model based on the evaluation.

[0851] (Claim 3)

[0852] [The system according to claim 1, wherein the display unit is configured to display images related to a culture specific to the region.

[0853] "Application Example 1"

[0854] (Claim 1)

[0855] [Means for presenting an image recognition task, which is presented by an information processing terminal including a display device, that requires the user to label images based on a specific task,

[0856] [Means for receiving the labeling information selected by the user and storing the labeling information in an electronic storage device,

[0857] [Methods for training machine learning models using stored labeling information,

[0858] [Methods for improving the accuracy of recognition of specific regional cultures and natural landscapes using trained machine learning models,

[0859] [Means for analyzing captured images and classifying whether the images have specific features,

[0860] [Means for presenting cultural or geographical information to the user's information processing terminal based on identified images,

[0861] A system that includes this.

[0862] (Claim 2)

[0863] [Means for analyzing the stored labeling information and evaluating the performance of the machine learning model,

[0864] [The system according to claim 1, which adjusts the numerical parameters of the machine learning model based on the evaluation.

[0865] (Claim 3)

[0866] [The system according to claim 1, wherein the display device is configured to display digital images relating to a unique local culture.

[0867] "Example 2 of combining an emotion engine"

[0868] (Claim 1)

[0869] [Means for displaying an image recognition task presented by a terminal including a display device, which requires the user to label images based on a specific task,

[0870] [Methods for acquiring user emotional data using an emotion analysis device and adjusting the difficulty level of tasks based on said data,

[0871] [Means for receiving labeling information and sentiment data selected by the user and storing the information on a recording medium,

[0872] [Methods for training an artificial intelligence model using stored labeling information and sentiment data,

[0873] [Methods for improving the accuracy of identification of specific cultures and landscapes using trained artificial intelligence models,

[0874] A system that includes this.

[0875] (Claim 2)

[0876] [Means for analyzing the stored labeling information and sentiment data and for evaluating the performance of the artificial intelligence model,

[0877] [The system according to claim 1, which adjusts the components of the artificial intelligence model based on the evaluation.

[0878] (Claim 3)

[0879] [The system according to claim 1, wherein the display device is configured to display images relating to a culture specific to a region.

[0880] "Application example 2 when combining with an emotional engine"

[0881] (Claim 1)

[0882] [Means for presenting an image recognition task displayed by a device including a display device, which requires the user to label images based on a specific task,

[0883] [Means for receiving the labeling information and sentiment data selected by the user and storing the data,

[0884] [Methods for training an artificial intelligence model using stored labeling information and sentiment data,

[0885] [Means for improving recognition accuracy and personalized user experience for specific cultures and landscapes using trained artificial intelligence models,

[0886] A system that includes this.

[0887] (Claim 2)

[0888] [Means for analyzing the stored labeling information and sentiment data and for evaluating the performance of the artificial intelligence model,

[0889] [The system according to claim 1, which adjusts the parameters of an artificial intelligence model based on the evaluation and proposes tasks and products suitable for the user.

[0890] (Claim 3)

[0891] [The system according to claim 1, wherein the display device is configured to display images relating to a culture specific to the region and to adjust the recommended content by taking into account emotional responses in order to attract the user's interest. [Explanation of Symbols]

[0892] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for presenting an image recognition task displayed on a terminal including display units, which requires the user to label images based on a specific task, A means for receiving the labeling information selected by the user and storing the labeling information, A method for training an artificial intelligence model using stored labeling information, A means of improving the accuracy of recognition of specific cultures and landscapes using a trained artificial intelligence model, A system that includes this.

2. A means for analyzing the stored labeling information and evaluating the performance of the artificial intelligence model, The system according to claim 1, which adjusts the parameters of the artificial intelligence model based on the evaluation.

3. The system according to claim 1, wherein the display unit is set to display an image relating to the culture unique to a country.

Images