System

Abstract

A system is provided.SOLUTION: A system comprising: means for detecting and blocking inappropriate content using production-based artificial intelligence; means for updating an inappropriate content list based on collected data; means for evaluating the appropriateness of content a user attempts to access based on the inappropriate content list; means for recommending educational content; means for collecting and enabling remote monitoring of user usage; and means for parents to provide feedback.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] As the information society evolves, children's opportunities to access the Internet are rapidly increasing, but so is the risk of accessing inappropriate content and malicious websites. This threatens children's security and privacy, and leads to problems such as digital addiction and increased screen time. There is a need for methods to mitigate these risks and make it easier for parents to manage their children's digital environments. There is also a need for methods to promote access to educational content and help children develop healthy online communication skills. [Means for solving the problem]

[0005] The present invention provides a system that includes means for detecting and blocking inappropriate content using generative artificial intelligence, means for updating an inappropriate content list based on collected data, means for evaluating the appropriateness of content a user attempts to access based on the list, means for recommending educational content, means for collecting and remotely monitoring user usage, and means for parents to provide feedback, thereby protecting children from inappropriate content, making it easier for parents to remotely monitor their digital environment, and promoting access to educational content and fostering healthy online communication skills.

[0006] "Generative AI" is an AI system that uses natural language processing and deep learning technologies to analyze content such as text, images, and videos, and determine and generate appropriate content.

[0007] "Inappropriate content" refers to websites, text, images, videos or other digital content that contains material that is harmful or offensive to children.

[0008] A "blocking means" is a component of a system that has the ability to detect attempted access to inappropriate content and limit or prevent access to that content.

[0009] "Database updating" refers to the process of regularly collecting the latest inappropriate content lists and related information to keep the database up to date.

[0010] A "means for assessing the appropriateness of content" is a system component that has the function of analyzing the content that a user attempts to access and determining whether that content is inappropriate.

[0011] An "educational content recommendation means" is a system component that has the ability to select and recommend appropriate and useful educational content based on a user's interests and learning style.

[0012] "Means for collecting user usage information" refers to system components that have the ability to collect and store usage data about applications and websites used by users.

[0013] The "means for enabling remote monitoring" is a system component that has the function of enabling real-time monitoring of the usage status of a user's smartphone from a remote location.

[0014] The "means for parental feedback" is a component of the system that allows parents to provide real-time comments and instructions on their children's smartphone usage. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0016] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0017] First, the terms used in the following description will be explained.

[0018] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0019] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0020] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0021] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0022] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0026] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0027] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0028] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0029] 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 of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0030] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0033] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0034] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0035] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0036] This invention is a system that uses generative artificial intelligence (AI) to detect and block inappropriate content and maintain a healthy digital environment for children. This system works in cooperation with different entities: a server, a device, and a user.

[0037] 1. Server-side implementation

[0038] Collecting and updating inappropriate content lists

[0039] Server: Regularly collects the latest inappropriate content lists from the Internet and updates the database. This list includes inappropriate keywords, website URLs, and content metadata, allowing it to respond to the latest threat information.

[0040] Generative AI Training

[0041] Server: Trains generative AI models using large amounts of text and other media data. This process uses natural language processing (NLP) techniques and deep learning algorithms. The trained AI continues to learn based on new data, resulting in more accurate detection capabilities.

[0042] 2. Terminal side embodiment

[0043] Download the filtering database

[0044] Device: Periodically downloads the filtering database from the server and stores it in local storage. This database provides the basis for assessing the appropriateness of content.

[0045] Real-time content evaluation

[0046] On the device: When a user attempts to access a website or app, the generative AI consults a locally stored filtering database and evaluates the content's appropriateness in real time. Content deemed inappropriate is immediately blocked and a warning message is displayed to the user.

[0047] Educational content recommendations

[0048] On-device: Generative AI will recommend appropriate educational content offline and online based on the user's usage history and interests. These recommendations will be customized to the user's learning style and interests and displayed on the smartphone screen.

[0049] 3. User-side implementation

[0050] Remote Parental Monitoring

[0051] Users (Parents): Through a dedicated mobile application or web portal, they can monitor their children's smartphone usage in real time, including data such as apps used, websites visited, and screen time.

[0052] Providing appropriate feedback

[0053] User (Parent): Parents can use the collected data to provide feedback on their children's smartphone usage, with the assistance of generative AI, which can then send appropriate guidance and advice via text message.

[0054] Specific examples

[0055] Website Filtering

[0056] User: Your child tries to visit a new website.

[0057] On the device: Website URLs are matched against a local filtering database and generative AI analyzes the content.

[0058] Generative AI: Determines whether a website contains inappropriate content.

[0059] Device: Blocks website access and displays "This site is inappropriate."

[0060] Educational content recommendations

[0061] Users: Children are interested in online learning.

[0062] Device: Generative AI recommends appropriate educational content based on a child's interests and past usage history.

[0063] Device: Sends a notification to the parent app saying "New educational content: Math game recommended."

[0064] Remote Monitoring

[0065] User (Parent): Check your child's usage on the mobile app.

[0066] Device: View apps used, websites visited, and screen time history.

[0067] User (parent): Sends appropriate advice to the child based on suggestions from the generative AI.

[0068] Through these steps, the system not only provides features to help children stay safe in the digital environment, but also gives parents the tools to effectively monitor and provide feedback remotely, making their children's online experience safer and more educational.

[0069] The processing flow will be explained below.

[0070] Step 1:

[0071] Server: Collects inappropriate content and updates the database. The server collects inappropriate keywords, URLs, and metadata from the Internet and updates the filtering database. This collection is done using web scraping technology and APIs.

[0072] Step 2:

[0073] Server: Trains the generative AI model. The server uses large amounts of collected text and media data to train the generative AI. Natural language processing (NLP) techniques and deep learning algorithms (e.g., Transformer-based models) are used for training.

[0074] Step 3:

[0075] Device: Downloads the filtering database. The device periodically sends a request to the server, downloads the latest filtering database, and stores it in local storage.

[0076] Step 4:

[0077] Device: Monitors user website access and compares the URLs they attempt to access against a local filtering database. Generative AI analyzes the content of the webpage in real time and evaluates its appropriateness. If it is deemed inappropriate, it proceeds to the next step.

[0078] Step 5:

[0079] Device: Blocks inappropriate content. Real-time analysis is performed, and if content is deemed inappropriate, a warning message such as "This site is inappropriate" is displayed to the user and access is blocked.

[0080] Step 6:

[0081] Device: Recommends educational content. By analyzing the user's usage history, interests, and past search history, generative AI selects appropriate educational content and notifies the device.

[0082] Step 7:

[0083] Device: Display recommended educational content on the user's (child's) screen. Promote access to educational content through notifications, dedicated pages, etc.

[0084] Step 8:

[0085] User (Parent): Monitor your child's smartphone usage in real time using a dedicated mobile app or web portal, retrieve data from the server, and view activity reports.

[0086] Step 9:

[0087] User (Parent): Provides appropriate feedback on children's smartphone usage. Supported by generative AI, real-time guidance and comments are sent as text messages. When the parent enters the feedback and presses the send button, the message appears on the child's device.

[0088] Example 1

[0089] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0090] The internet is filled with a wealth of harmful and inappropriate content for children, making it difficult for parents to maintain a safe online environment for their children. Providing an environment where children can access appropriate educational content based on their interests is also a challenge. Furthermore, there is a lack of systems that allow parents to remotely monitor their children's online activity and provide appropriate feedback. This invention aims to comprehensively solve these problems and maintain a healthy digital environment for children.

[0091] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0092] In this invention, the server includes a means for collecting a list of inappropriate content from the Internet and updating the database, a means for training a generative artificial intelligence model using large amounts of data, and a means for providing a filtering database to the terminal. This enables the latest inappropriate content to be quickly detected and blocked. Furthermore, the filtering database downloaded from the server is stored in local storage, and the appropriateness of content the user attempts to access is evaluated and blocked in real time, thereby maintaining a safe digital environment for children. Furthermore, the system analyzes the user's usage history and can recommend appropriate educational content based on that analysis, thereby enhancing children's motivation to learn. Parents can remotely monitor their children's smartphone usage through a dedicated application and provide appropriate feedback, providing an educational and safe digital environment.

[0093] "Inappropriate Content" refers to content that is harmful to children and contains violence, pornography, discrimination, or other ethically objectionable information.

[0094] "Generative artificial intelligence" is a form of artificial intelligence that can learn from large amounts of data and make predictions and classifications based on new data, and specifically refers to models that involve data generation.

[0095] A "database" refers to a collection of information that is systematically organized and managed so that it can be stored, searched, updated, and deleted.

[0096] A "filtering database" refers to a database that collects keywords, URLs, metadata, etc. that contain inappropriate content and provides basic data for evaluating the appropriateness of the content accessed by users.

[0097] "Real-time evaluation" refers to the process of evaluating and analyzing an action or event almost simultaneously when that action or event occurs.

[0098] "Educational content" refers to information, teaching materials, applications, etc. intended for learning and knowledge acquisition.

[0099] "Usage history" refers to a record of the operations performed by the user and the content accessed.

[0100] "Remote monitoring" refers to monitoring the status of devices or systems from a physically distant location using communication technology.

[0101] "Feedback" refers to providing evaluation, guidance, and advice to others regarding usage and behavior.

[0102] A "server" refers to a computer system that provides services and data to other computers (clients) on a network.

[0103] "Terminal" refers to a computer or mobile device that is directly operated by a user.

[0104] First, this system operates with three main components: a server, a terminal, and a user. Specific embodiments of each component will be described in detail below.

[0105] Server-side implementation

[0106] Collecting and updating inappropriate content lists

[0107] The server first collects a list of inappropriate content from the Internet and updates the database. This is done using web crawling technology. Specifically, it uses Python libraries such as Scrapy and BeautifulSoup to collect inappropriate keywords, URLs, and metadata from various sites. This allows the server to always have the latest threat information.

[0108] Generative AI Training

[0109] The server then uses the large amount of data to train a generative artificial intelligence (AI) model, using tools such as Hugging Face's Transformers library, TensorFlow, and PyTorch. The generative AI model learns from the collected data and is able to detect inappropriate content with high accuracy even when new data is used.

[0110] Terminal side embodiment

[0111] Download the filtering database

[0112] The device periodically downloads the filtering database from the server and stores it in local storage. This operation is performed via an API call, and the downloaded database is saved in a local database such as SQLite so that it can be used the next time the device is accessed.

[0113] Real-time content evaluation

[0114] When a user attempts to access a website or app, the device references a locally stored filtering database and uses generative AI to evaluate the appropriateness of the content in real time. For example, using AI models using Scikit-learn and Keras, the device immediately determines whether the content is inappropriate. If it is determined to be inappropriate, the content is immediately blocked and a warning message is displayed to the user.

[0115] Educational content recommendations

[0116] The device uses generative AI to recommend appropriate educational content based on the user's usage history and interests. Recommended content is not only displayed on the device in real time, but also notified to the parent's mobile app. For example, a message may appear saying, "New educational content: Math games recommended."

[0117] User-Side Embodiment

[0118] Remote Parental Monitoring

[0119] Users (parents) can monitor their children's smartphone usage in real time through a dedicated mobile application or web portal. This monitoring includes detailed data such as apps used, websites visited, screen time, etc. The mobile application syncs and displays the data using cloud services such as Firebase and AWS.

[0120] Providing appropriate feedback

[0121] Based on this data, parents can provide appropriate feedback on their children's smartphone usage. With the help of generative AI, parents can send gentle guidance and advice to their children, such as "Try to focus more on your studies," via text message.

[0122] Examples of concrete examples and prompts

[0123] Website Filtering

[0124] User: Your child tries to visit a new website.

[0125] On the device: Website URLs are matched against a local filtering database and generative AI analyzes the content.

[0126] Generative AI: Determines whether a website contains inappropriate content.

[0127] Device: Blocks website access and displays "This site is inappropriate."

[0128] Example prompt: "Please rate whether a new website you visit is inappropriate."

[0129] Educational content recommendations

[0130] Users: Children are interested in online learning.

[0131] Device: Generative AI recommends appropriate educational content based on a child's interests and past usage history.

[0132] Device: Sends a notification to the parent app saying "New educational content: Math learning game recommended."

[0133] Example prompt: "Recommend appropriate educational content based on your child's interests."

[0134] Remote Monitoring

[0135] User (Parent): Check your child's usage on the mobile app.

[0136] Device: View apps used, websites visited, and screen time history.

[0137] User (parent): Sends appropriate advice to the child based on suggestions from the generative AI.

[0138] Example prompt: "Please provide appropriate feedback regarding your child's smartphone usage."

[0139] Through these steps, the system can keep children's digital environments safe and educationally rewarding, while allowing parents to effectively monitor and provide feedback remotely.

[0140] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0141] Step 1:

[0142] Collecting and updating inappropriate content lists

[0143] The server collects inappropriate content lists from across the Internet, including inappropriate keywords, website URLs, and content metadata.

[0144] Input: Multiple sources on the internet

[0145] Data processing: Collect data using web crawling techniques (such as Python's Scrapy or BeautifulSoup).

[0146] Output: Save and update the inappropriate content list in the database.

[0147] Specific behavior:

[0148] 1. The server accesses the list of web pages to be crawled.

[0149] 2. Extract inappropriate keywords, URLs, and metadata from each page.

[0150] 3. This data will be centrally managed and added / updated to the inappropriate content list database.

[0151] Step 2:

[0152] Generative AI Training

[0153] The server trains generative AI models using large amounts of text and other media data.

[0154] Input: Large collected dataset

[0155] Data processing: Training AI models using libraries such as Hugging Face Transformers, TensorFlow, and PyTorch

[0156] Output: A trained generative AI model

[0157] Specific behavior:

[0158] 1. The server supplies the collected data to the AI ​​model.

[0159] 2. AI models use deep learning algorithms to analyze and learn from data.

[0160] 3. The trained model is saved on the server and updated as needed.

[0161] Step 3:

[0162] Download the filtering database

[0163] The terminal periodically downloads the filtering database from the server and stores it in local storage.

[0164] Input: Server filtering database

[0165] Data processing: Download by API call and save to local database (e.g. SQLite)

[0166] Output: Locally stored filtering database

[0167] Specific behavior:

[0168] 1. The device sends a request to the server API.

[0169] 2. The server responds with the latest filtering database.

[0170] 3. The device saves the downloaded database in local storage.

[0171] Step 4:

[0172] Real-time content evaluation

[0173] The device uses generative AI to evaluate the appropriateness of content in real time when a user attempts to access a website or app.

[0174] Input: URL or text data of the content you want to access

[0175] Data calculation: Matching with local filtering database and analysis by AI evaluation model

[0176] Output: Evaluation result: Good or bad

[0177] Specific behavior:

[0178] 1. The device receives the URL and text data of the content the user wants to access.

[0179] 2. Match with a local database and use a generative AI model to analyze the content.

[0180] 3. If the content is deemed inappropriate, access will be blocked and a warning message will be displayed.

[0181] Step 5:

[0182] Educational content recommendations

[0183] The device recommends appropriate educational content based on the user's usage history and interests.

[0184] Input: Usage history, interest data

[0185] Data processing: Content recommendation using usage history analysis and generative AI

[0186] Output: Recommended educational content

[0187] Specific behavior:

[0188] 1. The device analyzes the user's usage history and interest data.

[0189] 2. Generative AI selects the most appropriate educational content based on this data.

[0190] 3. The recommendation results are displayed on the user's device and also notified to the parent's app.

[0191] Step 6:

[0192] Remote Parental Monitoring

[0193] Parents can monitor their children's smartphone usage in real time through a dedicated mobile app.

[0194] Input: Child usage data (apps used, websites visited, screen time)

[0195] Data Processing: Collection and Display of Usage Data

[0196] Output: Display data in the parent mobile application

[0197] Specific behavior:

[0198] 1. Collect usage data from your child's device.

[0199] 2. Synchronize the collected data with a cloud service (Firebase or AWS).

[0200] 3. Display data in real time on parent's mobile app.

[0201] Step 7:

[0202] Providing appropriate feedback

[0203] Parents provide appropriate feedback based on the collected data, and text messages are sent based on generative AI suggestions.

[0204] Input: Collected usage data and generative AI feedback suggestions

[0205] Data processing: generating feedback suggestions

[0206] Output: Feedback as a text message

[0207] Specific behavior:

[0208] 1. Parents can check their children's usage within the mobile app.

[0209] 2. Generative AI automatically generates appropriate feedback and advice.

[0210] 3. The parent will then send the suggested content as a text message to their child.

[0211] (Application example 1)

[0212] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0213] In today's digital society, children have more opportunities to access a variety of media content. However, some media content is inappropriate for children, making it difficult for parents to monitor all content. Furthermore, there is a lack of systems that efficiently provide educational media content to support children's healthy development. There is a need to develop a system that can solve these problems and help children grow up in a safe and beneficial digital environment.

[0214] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0215] In this invention, the server includes a means for periodically collecting the latest inappropriate media content list from the Internet and updating the database, a means for training a generative artificial intelligence model using a large amount of text data and other media data, and a means for evaluating the appropriateness of media content accessed by users in real time. This allows parents to remotely monitor and provide feedback while maintaining a healthy digital environment for their children. The generative artificial intelligence model can also recommend educational media content based on children's interests.

[0216] "Generative AI" is an AI that learns from large amounts of data and generates information in a way that can respond to new data.

[0217] "Inappropriate media content" means media content that contains violent, sexual, discriminatory, or otherwise harmful to children.

[0218] A "media content list" identifies inappropriate media content and manages the list as a database.

[0219] The "means for assessing appropriateness" is a system that has the function of analyzing and assessing whether media content is inappropriate.

[0220] "Educational media content" is media content that is beneficial to children's learning and development and has high educational value.

[0221] "User usage status" refers to information such as what content the user has viewed, how long the user has viewed it, and what interests the user has.

[0222] "Remote monitoring capabilities" are features that allow parents or supervisors to monitor their children's digital environments remotely and in real time.

[0223] The "means of providing feedback" is a function that allows parents to send appropriate guidance and advice to their children with the assistance of generative artificial intelligence.

[0224] A "device for real-time analysis and blocking" is a device that has the capability to analyze media content on the fly and immediately block it if necessary.

[0225] A "generative artificial intelligence model" is a learning model that allows generative artificial intelligence to suggest optimal media content based on data.

[0226] This invention is a system for detecting and blocking inappropriate media content using generative artificial intelligence to maintain a healthy digital environment for children. Specific embodiments of this system are described below.

[0227] 1. Server-side implementation

[0228] The server has the means to perform the following functions:

[0229] Collecting and updating inappropriate media content lists

[0230] Server: Regularly collects the latest inappropriate media content list from the Internet and updates the database. This list includes inappropriate keywords, website URLs, and content metadata, allowing it to respond to the latest threat information.

[0231] Generative AI Training

[0232] Server: Trains generative artificial intelligence models using large amounts of text and other media data. This process uses natural language processing (NLP) techniques and deep learning algorithms. The trained AI continues to learn based on new data, resulting in more accurate detection capabilities.

[0233] 2. Terminal side embodiment

[0234] The terminal has the means to perform the following functions:

[0235] Download the filtering database

[0236] Device: Periodically downloads the filtering database from the server and stores it in local storage. This database provides the foundation data used to evaluate the appropriateness of media content.

[0237] Real-time content evaluation

[0238] On the device: When a user attempts to access media content, the generative AI will refer to a locally stored filtering database and assess the appropriateness of the content in real time. Content deemed inappropriate will be immediately blocked and a warning message will be displayed to the user.

[0239] Educational content recommendations

[0240] On-device: Generative AI will recommend appropriate educational content based on the user's usage history and interests. These recommendations will be customized to the user's learning style and interests and displayed on the smartphone screen.

[0241] 3. User-side implementation

[0242] The user (parent) has the means to perform the following functions:

[0243] Remote Parental Monitoring

[0244] Users (Parents): Through a dedicated mobile application or web portal, they can monitor their children's device usage in real time, including data such as apps used, websites visited, and screen time.

[0245] Providing appropriate feedback

[0246] User (Parent): Parents can use the collected data to provide feedback on their children's device usage, with the assistance of generative AI, which can then send appropriate guidance and advice via text message.

[0247] Specific examples

[0248] Real-time evaluation of media content

[0249] User: A child tries to access a new video.

[0250] On device: Video frames are analyzed in real time, and generative AI detects inappropriate content.

[0251] Generative AI: Determines whether a video contains inappropriate content.

[0252] Device: Immediately stop playing the video and display "This content is inappropriate."

[0253] Educational content recommendations

[0254] Users: Children are interested in online learning.

[0255] Device: Generative AI recommends appropriate educational content based on a child's interests and past usage history.

[0256] Device: Sends a notification to the parent app saying "New educational content: Math game recommended."

[0257] Remote Monitoring

[0258] User (Parent): Check your child's usage on the mobile app.

[0259] Device: View apps used, websites visited, and screen time history.

[0260] User (parent): Based on the generative AI's suggestions, send advice to the child such as "This anime is violent, so please watch other educational content."

[0261] Prompt Sentence Examples

[0262] Below are some example prompts for recommending appropriate educational content to generative AI:

[0263] "Please recommend appropriate educational content based on my child's interests. Past viewing has been dominated by action anime, so please suggest more peaceful educational content."

[0264] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0265] Step 1:

[0266] Collecting and updating inappropriate media content lists

[0267] Server: The server periodically collects data about objectionable media content from multiple sources across the internet, which may involve web scraping techniques. Data collected may include objectionable keywords, website URLs, and metadata.

[0268] Input: Inappropriate content information on the internet.

[0269] Output: An updated list of inappropriate media content.

[0270] What it does: The crawler visits specific sites on the internet, extracts data, stores this data in a database on the server, and updates the inappropriate media content list.

[0271] Step 2:

[0272] Generative AI Training

[0273] Server: The server trains generative artificial intelligence models using large amounts of text and other media data, a process that uses natural language processing (NLP) techniques and deep learning algorithms.

[0274] Input: Large amounts of text and media data.

[0275] Output: A highly accurate generative artificial intelligence model.

[0276] What it does: Train AI with a dataset, validate the results, and update the model. For example, run deep learning algorithms using Python and TensorFlow.

[0277] Step 3:

[0278] Download the filtering database

[0279] Device: The device periodically downloads the filtering database from the server and stores it in local storage.

[0280] Input: Filtering database on the server.

[0281] Output: Filtering database stored in the device's local storage.

[0282] Specific operation: The device sends an API request to retrieve the latest database from the server, and then saves the database in the device's storage, making it available for offline reference.

[0283] Step 4:

[0284] Real-time content evaluation

[0285] Device: When a user attempts to access media content, the device consults a locally stored filtering database, and generative AI assesses the content's appropriateness in real time.

[0286] Input: Media content data and local filtering database.

[0287] Output: Content appropriateness assessment result (appropriate / inappropriate).

[0288] What it does: The device captures the media content being played frame by frame and passes it to a generative AI for analysis. If the content is deemed inappropriate, it blocks it and displays a warning message to the user.

[0289] Step 5:

[0290] Educational content recommendations

[0291] Device: Generative AI recommends appropriate educational content based on the user's usage history and interests.

[0292] Input: User usage history and interest data.

[0293] Output: Recommended educational content.

[0294] Specific operation: The device sends prompts to the generative AI to generate educational content based on the user's interests. For example, a prompt such as, "Please recommend appropriate educational content based on my child's interests. Most of his past viewings have been action anime, so please suggest more peaceful educational content."

[0295] Step 6:

[0296] Remote Parental Monitoring

[0297] Users (Parents): Monitor their children's device usage in real time through a dedicated mobile application or web portal.

[0298] Input: Child's device usage data.

[0299] Output: Monitoring results displayed on the parent device.

[0300] What it does: The app sends usage data from the child's device to a server. The application on the parent's device receives the data and visualizes it in the form of a dashboard.

[0301] Step 7:

[0302] Providing appropriate feedback

[0303] User (Parent): Parents use the collected data to provide feedback on their children's device usage, with the assistance of generative AI to send appropriate guidance and advice via text message.

[0304] Input: Parent feedback and generative AI advice.

[0305] Output: A text message sent to the child's device.

[0306] How it works: Parents use the app to send prompts to the generative AI, which then generates appropriate messages, such as "This cartoon is violent, so please watch other educational content" and send it to their child's device.

[0307] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0308] This invention is a system that uses generative artificial intelligence (AI) and an emotion engine to detect and block inappropriate content and maintain a healthy digital environment for children. Furthermore, this system analyzes the emotional state of the user (child), recommends appropriate educational content based on the user's emotion, and provides feedback support to parents.

[0309] 1. Server-side implementation

[0310] Collecting and updating inappropriate content lists

[0311] Server: The server periodically collects the latest inappropriate content list from the internet and updates the database. This list includes inappropriate keywords, website URLs, and metadata, and is collected using web scraping technology and APIs.

[0312] Generative AI Training

[0313] Server: The server trains the generative AI using large amounts of text and media data, using natural language processing (NLP) techniques and deep learning algorithms (e.g., Transformer-based models). The trained AI continues to learn based on new data, achieving highly accurate detection capabilities.

[0314] 2. Terminal side embodiment

[0315] Download the filtering database

[0316] Device: The device periodically sends a request to the server to download the latest filtering database and save it in local storage.

[0317] Real-time filtering

[0318] On your device: Every time you access a website or app, we use a local filtering database and generative AI to analyze the content, detect and block inappropriate content.

[0319] Implementing the Emotion Engine

[0320] Device: The device uses sensors such as a camera and microphone to capture the user's (child's) facial expressions and voice, and sends the data to the emotion engine, which analyzes the data and determines the user's emotional state in real time.

[0321] Educational content recommendations

[0322] Device: Based on the user's usage history, interests, and emotional state, generative AI selects the most appropriate educational content and notifies the device. For example, if the user is feeling stressed, it will recommend relaxation content, and if the user is concentrating on a learning topic they are interested in, it will recommend learning reinforcement content.

[0323] 3. User-side implementation

[0324] Remote Parental Monitoring

[0325] User (Parent): Parents can monitor their children's smartphone usage and emotional state in real time through a dedicated mobile app or web portal. Data collected includes apps used, websites visited, screen time, and emotional state.

[0326] Providing appropriate feedback

[0327] User (Parent): Parents can provide appropriate feedback with the support of generative AI based on their child's smartphone usage and emotional state. When a parent enters feedback and presses the send button, the feedback is displayed on the child's device in real time. For example, if the child is stressed, they can send an encouraging message, or if the child is concentrating on their studies, they can send a message of encouragement.

[0328] Specific examples

[0329] Website Filtering

[0330] User (child): A child tries to access a new website.

[0331] On the device: Website URLs are matched against a local filtering database and generative AI analyzes the content.

[0332] Generative AI: Determines whether a website contains inappropriate content.

[0333] Device: Blocks website access and displays "This site is inappropriate."

[0334] Educational content recommendations

[0335] User (child): Children are interested in online learning.

[0336] Device: Generative AI recommends appropriate educational content based on a child's interests, past usage history, and emotional state.

[0337] Device: Sends a notification to the parent app saying "New educational content: Math game recommended."

[0338] Real-time emotion monitoring and feedback

[0339] User (child): The emotion engine detects stress levels while children are learning remotely.

[0340] Device: A notification will be sent to the parent's device saying "Your child is stressed."

[0341] User (parent): Create an encouraging message remotely and press the send button, and the message "Good luck!" will appear in real time on the child's device.

[0342] In this way, the system maintains a healthy digital environment for children, while enhancing educational value and providing parents with tools for effective remote monitoring and feedback.

[0343] The processing flow will be explained below.

[0344] Step 1:

[0345] Server: Collects inappropriate content and updates the database. The server collects the latest inappropriate content list from the Internet using scraping or APIs, and periodically updates the filtering database. This database includes inappropriate keywords, URLs, and metadata.

[0346] Step 2:

[0347] Server: Trains the generative AI model. The server trains the generative AI using large amounts of text and media data. Using natural language processing (NLP) techniques and deep learning algorithms (e.g., Transformer-based models), the AI ​​continues to learn based on new data.

[0348] Step 3:

[0349] Device: Downloads the filtering database. The device periodically sends a request to the server to download the latest filtering database and saves it in local storage. This database is used to evaluate the appropriateness of content.

[0350] Step 4:

[0351] On the device: Every time a user attempts to access a website or app, the content is evaluated for appropriateness. The URL is checked against a local filtering database, and generative AI analyzes the content in real time. If it is deemed inappropriate, it is processed in the next step.

[0352] Step 5:

[0353] Device: Blocks inappropriate content. If real-time analysis determines that content is inappropriate, a warning message will be displayed to the user stating "This site is inappropriate." Access will be blocked.

[0354] Step 6:

[0355] Device: Implementation of the emotion engine. The device's camera and microphone are used to capture the user's (child's) facial expressions and voice, and the data is sent to the emotion engine. The emotion engine analyzes the data and determines the user's emotional state in real time.

[0356] Step 7:

[0357] Device: Recommends educational content. Based on the user's usage history, interests, and emotional state, generative AI selects the most appropriate educational content and notifies the device. For example, if the user is feeling stressed, it will recommend relaxation content.

[0358] Step 8:

[0359] Device: Display recommended educational content on the user's (child's) screen. Promote access to educational content via push notifications and dedicated pages.

[0360] Step 9:

[0361] User (Parent): Monitor their child's smartphone usage and emotional state in real time using a dedicated mobile app or web portal. Data collected includes apps used, websites visited, screen time, and emotional state.

[0362] Step 10:

[0363] User (Parent): Provides feedback on the child's smartphone usage and emotional state. With the support of generative AI, the parent enters the feedback and presses the send button, and the feedback is displayed in real time on the child's device. For example, if stress is detected, an encouraging message is sent, and if the child is concentrating on their studies, a supportive message is sent.

[0364] Example 2

[0365] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0366] In today's digital environment, there is a lot of inappropriate content available online, which can have a negative impact on children. It is also difficult to understand in real time what content children are interested in and their emotional state, making it difficult to provide appropriate educational content. Furthermore, there is a lack of means for parents to remotely monitor their children's digital behavior and provide appropriate feedback. Therefore, there is a need for a system that integrates the detection and blocking of inappropriate content, analysis of children's emotions, recommendation of educational content, and remote monitoring and feedback provision functions.

[0367] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for detecting and blocking inappropriate content using generative artificial intelligence, means for updating an inappropriate content list based on collected data, means for evaluating the appropriateness of content a user attempts to access based on the inappropriate content list, means for analyzing a user's emotional state using a camera and a microphone, means for recommending educational content based on the user's emotional state and usage status, means for collecting and remotely monitoring the user's usage status and emotional state, and means for parents to provide feedback. This not only enables children to enjoy learning and entertainment in a safe and healthy digital environment, but also enables parents to effectively monitor their children's digital behavior and provide appropriate feedback.

[0368] "Generative AI" is an AI technology that learns using large amounts of data and generates and analyzes content based on new data.

[0369] "Inappropriate Content" means any information or media that may be harmful to children or other users, such as violent, pornographic, or libelous material.

[0370] "Updating" refers to adding new information to existing data or lists and keeping them up to date.

[0371] A "filtering database" is a database that accumulates information about inappropriate content and uses that data to evaluate the appropriateness of the content.

[0372] "Evaluating the appropriateness of content" means determining whether the content a user is attempting to access is inappropriate.

[0373] "Camera and microphone" refers to a photographing and recording device for capturing the user's facial expressions and voice.

[0374] An "emotional state" refers to a user's mental and emotional state, and may include, for example, stress, joy, excitement, and the like.

[0375] "Analyzing" refers to extracting and understanding specific information from collected data.

[0376] "Educational content" is information or media intended to educate and increase the user's knowledge.

[0377] "Usage" refers to information about how users use devices and applications.

[0378] "Enabling remote monitoring" means being able to check the device usage status and the user's emotional state from a remote location.

[0379] "Providing feedback" means conveying advice or comments based on the user's actions or situation.

[0380] An "inappropriate content list" is a collection of information about inappropriate content that exists on the Internet.

[0381] This invention is a system that uses generative artificial intelligence (AI) and an emotion engine to detect and block inappropriate content and maintain a healthy digital environment for children. Furthermore, this system analyzes the emotional state of the user (child), recommends appropriate educational content based on the user's emotion, and provides feedback support to parents. Specific embodiments are described below.

[0382] Server-side implementation

[0383] Collecting and updating inappropriate content lists

[0384] The server periodically collects the latest inappropriate content list from the Internet and updates the database. This list includes inappropriate keywords, website URLs, and metadata. The server retrieves the data using a web scraping library such as Python's BeautifulSoup or Scrapy and stores it in an SQL database (e.g., MySQL or PostgreSQL). This allows the server to always maintain the latest inappropriate content information.

[0385] Generative AI Training

[0386] The server trains the generative AI using large amounts of text and media data. Deep learning frameworks used include TensorFlow and PyTorch, and Transformer-based models (e.g., BERT and GPT-3) are used. Natural language processing (NLP) techniques and deep learning algorithms are used for training, and the AI ​​continues to learn based on new data.

[0387] Terminal side embodiment

[0388] Download the filtering database

[0389] The device periodically sends a request to the server to download the latest filtering database and save it in local storage. The device periodically communicates with the server using the HTTP protocol to keep the database up to date.

[0390] Real-time filtering

[0391] When a user attempts to access a website or app, the device uses a local filtering database and generative AI to analyze the content, detect inappropriate content, and block it. Specifically, an AI model running on the device evaluates the content in real time and blocks access if it is deemed inappropriate. For example, when a user (child) attempts to access a new website, the device compares the URL with the local filtering database and the generative AI analyzes the content. If it is deemed inappropriate, the device blocks access to the website and displays the message, "This site is inappropriate."

[0392] Implementing the Emotion Engine

[0393] The device uses sensors such as a camera and microphone to capture the user's facial expressions and voice and send them to the emotion engine. Using libraries such as FaceAPI and AudioAnalyzer, this data is analyzed in real time to determine the user's emotional state. For example, if the emotion engine detects a child is stressed while learning remotely, the parent's device will receive a notification saying, "Your child is feeling stressed."

[0394] Educational content recommendations

[0395] Based on the user's usage history, interests, and emotional state, the generative AI selects the most suitable educational content and notifies the device. For example, if a user (child) shows interest in online learning, the generative AI will recommend appropriate educational content based on their interests, past usage history, and emotional state. The selected content is notified to the device and a message appears saying, "New educational content: Math game recommended."

[0396] User-side implementation (parent)

[0397] Remote Monitoring

[0398] Parents can monitor their children's smartphone usage and emotional state in real time through a dedicated mobile app or web portal. Collected data is displayed using Firebase and cloud services. This feature allows parents to understand their children's current situation and take appropriate action if necessary.

[0399] Providing feedback

[0400] Parents can provide appropriate feedback with the support of generative AI based on their child's smartphone usage and emotional state. When a parent enters the feedback and presses the send button, the feedback is displayed on the child's device in real time. For example, if the child is stressed, they can send an encouraging message, or if they are concentrating on their studies, they can send a message of support.

[0401] Example prompt sentence:

[0402] Prompt: "Please recommend some new educational content. My child is interested in math and has been feeling stressed lately. I'd like some content that's relaxing yet educational."

[0403] The system will enhance educational value while maintaining a healthy digital environment for children, allowing parents to effectively monitor and provide feedback remotely.

[0404] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0405] Step 1:

[0406] Server: Collects inappropriate content lists from the Internet and updates the database.

[0407] Input: Web page data obtained from the Internet

[0408] Processing: The server uses Python's BeautifulSoup or Scrapy library to perform web scraping and extract inappropriate keywords, URLs, and metadata from web page data. The extracted data is stored in an SQL database (e.g., MySQL or PostgreSQL).

[0409] Output: Updated inappropriate content list

[0410] Specific Actions: The server runs a scheduled job (e.g., once daily) that crawls designated websites and stores inappropriate content information in a database.

[0411] Step 2:

[0412] Server: Trains the generative AI model.

[0413] Input: Text data, media data

[0414] Processing: Use TensorFlow or PyTorch to train Transformer-based models (e.g., BERT or GPT-3). Use natural language processing (NLP) techniques and deep learning algorithms to learn from large amounts of data.

[0415] Output: A trained generative AI model

[0416] What it does: The server loads the training data and trains the deep learning model. After the training period is over, the optimized model is saved and deployed as an inappropriate content detection algorithm.

[0417] Step 3:

[0418] On your device: Download the filtering database.

[0419] Input: Latest filtering database on the server

[0420] Processing: The device communicates with the server using an HTTP request and downloads the latest filtering database.

[0421] Output: Filtering database stored in local storage

[0422] Specific operation: The device executes a periodic job (for example, once a day) and sends an HTTP GET request to the server. The device receives the latest database from the server and stores it in local storage.

[0423] Step 4:

[0424] Device: Filter content in real time.

[0425] Input: URL and content data of the website or app the user is trying to access

[0426] Processing: The device analyzes the content using a local filtering database and generative AI, blocking access if it is deemed inappropriate.

[0427] Output: Filtered content (allowed or blocked)

[0428] How it works: When a user opens a browser and accesses a new website, the device sends the URL to a local filtering database and generative AI for analysis. If it is deemed inappropriate, the device blocks access and displays the message "This site is inappropriate."

[0429] Step 5:

[0430] Device: Uses camera and microphone to analyze the user's emotional state.

[0431] Input: User's facial expression data (still image or video), voice data

[0432] Processing: The device sends this data to the emotion engine, which analyzes it using FaceAPI and AudioAnalyzer to determine the user's emotional state (stress, joy, excitement, etc.).

[0433] Output: Information about the user's emotional state

[0434] How it works: The device's camera captures the user's face and the microphone records their voice. The emotion engine analyzes this data in real time to determine their emotional state. For example, if the user is feeling stressed, that information is stored on the device as an analysis result.

[0435] Step 6:

[0436] Device: Recommend educational content.

[0437] Input: User usage history, interest data, emotional state data

[0438] Processing: The generative AI selects the most appropriate educational content based on the above input data.

[0439] Output: Recommended educational content

[0440] Specific operation: The device uses generative AI to select the most suitable educational content based on the user's usage history, interests, and emotional state. The selected content is notified to the device, and a message appears saying, "New educational content: Math game recommended."

[0441] Step 7:

[0442] Parents (users): Remotely monitor their children's smartphone usage and emotional state.

[0443] Input: Child usage data, emotional state data

[0444] Processing: Parents can view this data in real time via a dedicated mobile app or web portal.

[0445] Output: Displayed usage and emotional state

[0446] What it does: When parents open the mobile app, they can see their child's current usage and emotional state in real time, allowing them to take necessary actions.

[0447] Step 8:

[0448] Parents (users): Provide feedback such as encouragement and advice.

[0449] Input: Child usage and emotional state data, parent-entered feedback

[0450] What happens: Parents create feedback in the app or web portal, press the submit button, and the feedback is sent in real time to their child's device.

[0451] Output: Feedback message displayed on the child's device

[0452] How it works: Parents can create an encouraging message such as "Good luck!" using the app and press the "Send" button, and the message will instantly appear on their child's device.

[0453] Through each of these steps, the system maintains a healthy digital environment for children, enhances educational value, and allows parents to effectively monitor and provide feedback remotely.

[0454] (Application example 2)

[0455] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0456] In recent years, the use of the Internet and digital devices has become more widespread, increasing the risk that children will be exposed to inappropriate media content. There is also a need for a method to recommend educationally valuable media to children based on their emotional state while maintaining a healthy digital environment. It is also difficult for parents to remotely monitor their children's usage and emotional state in real time and provide appropriate feedback. Therefore, the present invention aims to solve these issues by utilizing generative artificial intelligence and an emotion engine, and to provide a system that provides educational value while maintaining a healthy digital environment for children.

[0457] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for detecting and blocking inappropriate media using generative artificial intelligence, means for updating an inappropriate media list based on collected data, means for evaluating the appropriateness of media a user attempts to access based on the inappropriate media list, means for recommending educational media, means for collecting and remotely monitoring user usage, means for parents to provide feedback, and means for analyzing a user's emotional state and recommending appropriate media according to that emotional state. This reduces the risk of children being exposed to inappropriate media and makes it possible to provide media with educational value according to the user's emotional state. It also makes it easy for parents to remotely monitor their children's usage and emotional state in real time and provide appropriate feedback.

[0458] "Generative AI" is a technology that uses natural language processing and deep learning algorithms to automatically generate content based on new data, and then analyze and evaluate it.

[0459] "Inappropriate media" is digital content, such as text, images, audio, or video, that is deemed harmful or inappropriate for children.

[0460] An "inappropriate media list" is a list containing inappropriate keywords, URLs, and metadata collected from across the Internet.

[0461] "Educational media" refers to digital content such as text, images, audio, and video that supports children's education and learning.

[0462] "Emotional state" is data that indicates the psychological state of the user, detected using sensors such as a camera or microphone.

[0463] "Remote monitoring" is a feature that allows parents to monitor their children's digital device usage in real time from a distance using dedicated devices or applications.

[0464] "Feedback" is the act of parents sending messages of encouragement or warning based on their children's digital device usage and emotional state.

[0465] "Recommendation" is the act of generative artificial intelligence selecting and presenting appropriate media content based on the user's interests and emotional state.

[0466] MODE FOR CARRYING OUT THE INVENTION

[0467] This invention combines generative artificial intelligence (AI) and an emotion engine to detect and block inappropriate media in virtual stores, recommend products based on emotional states, and provide remote monitoring and feedback. The system ensures a safe digital environment for children and recommends appropriate products based on their emotional states.

[0468] System Configuration

[0469] server

[0470] Collecting and updating inappropriate media lists

[0471] Generative AI Training

[0472] Managing the filtering database

[0473] Terminal

[0474] Real-time detection and blocking of inappropriate media

[0475] Analysis of user emotional states using an emotion engine

[0476] Educational and relevant product recommendations

[0477] Remote monitoring data collection

[0478] User

[0479] Parents can monitor their children's usage and emotional state in real time through a dedicated application.

[0480] Parents provide appropriate feedback

[0481] Server-side implementation

[0482] The server performs the following functions:

[0483] Collecting and updating inappropriate media lists

[0484] The server periodically collects the latest inappropriate media list from the Internet and updates the database, which includes inappropriate keywords, website URLs, and metadata, using web scraping technology and APIs.

[0485] Generative AI Training

[0486] The server trains the generative AI using large amounts of text and media data, using natural language processing (NLP) techniques and deep learning algorithms (e.g., Transformer-based models). The trained AI continues to learn based on new data, achieving highly accurate detection capabilities.

[0487] Terminal side embodiment

[0488] The terminal performs the following functions:

[0489] Download the filtering database

[0490] The device periodically sends a request to the server to download the latest filtering database and store it in local storage.

[0491] Real-time filtering

[0492] Every time a user (child) tries to access a product in a virtual store, the content is analyzed using a local filtering database and generative AI to detect and block inappropriate content.

[0493] Implementing the Emotion Engine

[0494] The device uses sensors such as a camera and microphone to capture the user's (child's) facial expressions and voice, and sends the data to the emotion engine, which analyzes the data and determines the user's emotional state in real time.

[0495] Educational content recommendations

[0496] Based on the user's usage history, interests, and emotional state, generative AI selects the most suitable educational content and products and notifies the device. For example, if the user is feeling stressed, it will recommend relaxation products, and if the user is concentrating on a learning content that interests them, it will recommend learning reinforcement content.

[0497] User-Side Embodiment

[0498] Remote Parental Monitoring

[0499] Parents can monitor their children's smartphone usage and emotional state in real time through a dedicated mobile app or web portal. Data collected includes apps used, websites visited, screen time, and emotional state.

[0500] Providing appropriate feedback

[0501] Parents can provide appropriate feedback with the support of generative AI based on their child's smartphone usage and emotional state. When a parent enters the feedback and presses the send button, the feedback is displayed on the child's device in real time. For example, if the child is stressed, they can send an encouraging message, or if they are concentrating on their studies, they can send a message of support.

[0502] Specific examples

[0503] Website Filtering

[0504] A user (child) attempts to access a new website. The inappropriate media detected is processed as follows:

[0505] The URL is checked against a local filtering database

[0506] Generative AI analyzes content, and if it determines that a website is inappropriate media, it blocks it from being displayed and notifies users that "this site is inappropriate."

[0507] Educational content recommendations

[0508] Users (children) become interested in online learning

[0509] Generative AI suggests appropriate educational content and products based on user interests and past usage history

[0510] A notification is sent to the parent app saying, "New educational content: Math games recommended."

[0511] Real-time emotion monitoring and feedback

[0512] The emotion engine detects that the user (child) is feeling stressed during remote learning.

[0513] Parents receive a notification on their device that their child is feeling stressed.

[0514] Parents can remotely create encouraging messages and press the send button, which will then appear on their child's device with the message "Good luck!"

[0515] Prompt Sentence Examples

[0516] Prompt the generative AI model:

[0517] What products do you recommend for users when they are feeling stressed?

[0518] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0519] Step 1:

[0520] The server collects and periodically updates an inappropriate media list from the Internet. This list includes inappropriate keywords, URLs, and metadata. The server uses web scraping technology and public APIs to collect the latest data from the Internet and adds it to the filtering database, thereby keeping the inappropriate media list up to date.

[0521] Input: Data on the Internet

[0522] Output: Updated inappropriate media list

[0523] Step 2:

[0524] The server uses large amounts of text and media data to train the generative AI, using natural language processing (NLP) techniques and deep learning algorithms. For example, a Transformer-based model is used to train the AI ​​to detect inappropriate media with high accuracy. The trained AI model then continues to learn and improve based on new data.

[0525] Input: Large amounts of text and media data

[0526] Output: A trained generative AI model

[0527] Step 3:

[0528] The device periodically sends a request to the server to download the latest filtering database and save it in local storage, so that the device can always use the latest inappropriate media list.

[0529] Input: A request to the server

[0530] Output: Locally stored filtering database

[0531] Step 4:

[0532] When a user (child) tries to access a product in the virtual store, the device analyzes the content using a local filtering database and generative AI. The generative AI evaluates the product's text, images, and multimedia data to determine whether it contains inappropriate elements. If the generative AI determines that the product is inappropriate, it blocks the product and notifies the user, "This product is inappropriate."

[0533] Input: The product data the user is trying to access

[0534] Output: Inappropriate media blocking notification

[0535] Step 5:

[0536] The device uses sensors such as a camera and microphone to capture the user's (child's) facial expressions and voice, and sends the data to the emotion engine, which analyzes the sensor data and determines the user's emotional state in real time.

[0537] Input: Sensor data from camera and microphone

[0538] Output: Parsed user's emotional state

[0539] Step 6:

[0540] The device uses generative AI to recommend appropriate educational content and products based on the user's usage history, interests, and emotional state. For example, if the user is feeling stressed, it will recommend relaxation products, and if the user is concentrating on a learning topic they are interested in, it will recommend learning reinforcement content.

[0541] Input: User usage history, interests, emotional state

[0542] Output: Recommended educational content and products

[0543] Step 7:

[0544] The device collects the user's usage and emotional state and transmits it in real time to a dedicated parental application, through which parents can monitor their child's current usage and emotional state.

[0545] Input: User usage status, emotional state data

[0546] Output: Real-time data displayed in a dedicated parent application

[0547] Step 8:

[0548] The user (parent) creates a feedback message in a dedicated application based on the child's usage and emotional state, and presses the send button. The feedback is then displayed in real time on the child's device. For example, if the child is stressed, an encouraging message can be sent.

[0549] Input: Feedback message entered by the parent

[0550] Output: Messages displayed on the child's device in real time

[0551] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0552] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0553] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0554] [Second embodiment]

[0555] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0556] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0557] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0558] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0559] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0560] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0561] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0562] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0563] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific 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.

[0564] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0565] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0566] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. 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."

[0567] This invention is a system that uses generative artificial intelligence (AI) to detect and block inappropriate content and maintain a healthy digital environment for children. This system works in cooperation with different entities: a server, a device, and a user.

[0568] 1. Server-side implementation

[0569] Collecting and updating inappropriate content lists

[0570] Server: Regularly collects the latest inappropriate content lists from the Internet and updates the database. This list includes inappropriate keywords, website URLs, and content metadata, allowing it to respond to the latest threat information.

[0571] Generative AI Training

[0572] Server: Trains generative AI models using large amounts of text and other media data. This process uses natural language processing (NLP) techniques and deep learning algorithms. The trained AI continues to learn based on new data, resulting in more accurate detection capabilities.

[0573] 2. Terminal side embodiment

[0574] Download the filtering database

[0575] Device: Periodically downloads the filtering database from the server and stores it in local storage. This database provides the basis for assessing the appropriateness of content.

[0576] Real-time content evaluation

[0577] On the device: When a user attempts to access a website or app, the generative AI consults a locally stored filtering database and evaluates the content's appropriateness in real time. Content deemed inappropriate is immediately blocked and a warning message is displayed to the user.

[0578] Educational content recommendations

[0579] On-device: Generative AI will recommend appropriate educational content offline and online based on the user's usage history and interests. These recommendations will be customized to the user's learning style and interests and displayed on the smartphone screen.

[0580] 3. User-side implementation

[0581] Remote Parental Monitoring

[0582] Users (Parents): Through a dedicated mobile application or web portal, they can monitor their children's smartphone usage in real time, including data such as apps used, websites visited, and screen time.

[0583] Providing appropriate feedback

[0584] User (Parent): Parents can use the collected data to provide feedback on their children's smartphone usage, with the assistance of generative AI, which can then send appropriate guidance and advice via text message.

[0585] Specific examples

[0586] Website Filtering

[0587] User: Your child tries to visit a new website.

[0588] On the device: Website URLs are matched against a local filtering database and generative AI analyzes the content.

[0589] Generative AI: Determines whether a website contains inappropriate content.

[0590] Device: Blocks website access and displays "This site is inappropriate."

[0591] Educational content recommendations

[0592] Users: Children are interested in online learning.

[0593] Device: Generative AI recommends appropriate educational content based on a child's interests and past usage history.

[0594] Device: Sends a notification to the parent app saying "New educational content: Math game recommended."

[0595] Remote Monitoring

[0596] User (Parent): Check your child's usage on the mobile app.

[0597] Device: View apps used, websites visited, and screen time history.

[0598] User (parent): Sends appropriate advice to the child based on suggestions from the generative AI.

[0599] Through these steps, the system not only provides features to help children stay safe in the digital environment, but also gives parents the tools to effectively monitor and provide feedback remotely, making their children's online experience safer and more educational.

[0600] The processing flow will be explained below.

[0601] Step 1:

[0602] Server: Collects inappropriate content and updates the database. The server collects inappropriate keywords, URLs, and metadata from the Internet and updates the filtering database. This collection is done using web scraping technology and APIs.

[0603] Step 2:

[0604] Server: Trains the generative AI model. The server uses large amounts of collected text and media data to train the generative AI. Natural language processing (NLP) techniques and deep learning algorithms (e.g., Transformer-based models) are used for training.

[0605] Step 3:

[0606] Device: Downloads the filtering database. The device periodically sends a request to the server, downloads the latest filtering database, and stores it in local storage.

[0607] Step 4:

[0608] Device: Monitors user website access and compares the URLs they attempt to access against a local filtering database. Generative AI analyzes the content of the webpage in real time and evaluates its appropriateness. If it is deemed inappropriate, it proceeds to the next step.

[0609] Step 5:

[0610] Device: Blocks inappropriate content. Real-time analysis is performed, and if content is deemed inappropriate, a warning message such as "This site is inappropriate" is displayed to the user and access is blocked.

[0611] Step 6:

[0612] Device: Recommends educational content. By analyzing the user's usage history, interests, and past search history, generative AI selects appropriate educational content and notifies the device.

[0613] Step 7:

[0614] Device: Display recommended educational content on the user's (child's) screen. Promote access to educational content through notifications, dedicated pages, etc.

[0615] Step 8:

[0616] User (Parent): Monitor your child's smartphone usage in real time using a dedicated mobile app or web portal, retrieve data from the server, and view activity reports.

[0617] Step 9:

[0618] User (Parent): Provides appropriate feedback on children's smartphone usage. Supported by generative AI, real-time guidance and comments are sent as text messages. When the parent enters the feedback and presses the send button, the message appears on the child's device.

[0619] Example 1

[0620] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0621] The internet is filled with a wealth of harmful and inappropriate content for children, making it difficult for parents to maintain a safe online environment for their children. Providing an environment where children can access appropriate educational content based on their interests is also a challenge. Furthermore, there is a lack of systems that allow parents to remotely monitor their children's online activity and provide appropriate feedback. This invention aims to comprehensively solve these problems and maintain a healthy digital environment for children.

[0622] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0623] In this invention, the server includes a means for collecting a list of inappropriate content from the Internet and updating the database, a means for training a generative artificial intelligence model using large amounts of data, and a means for providing a filtering database to the terminal. This enables the latest inappropriate content to be quickly detected and blocked. Furthermore, the filtering database downloaded from the server is stored in local storage, and the appropriateness of content the user attempts to access is evaluated and blocked in real time, thereby maintaining a safe digital environment for children. Furthermore, the system analyzes the user's usage history and can recommend appropriate educational content based on that analysis, thereby enhancing children's motivation to learn. Parents can remotely monitor their children's smartphone usage through a dedicated application and provide appropriate feedback, providing an educational and safe digital environment.

[0624] "Inappropriate Content" refers to content that is harmful to children and contains violence, pornography, discrimination, or other ethically objectionable information.

[0625] "Generative artificial intelligence" is a form of artificial intelligence that can learn from large amounts of data and make predictions and classifications based on new data, and specifically refers to models that involve data generation.

[0626] A "database" refers to a collection of information that is systematically organized and managed so that it can be stored, searched, updated, and deleted.

[0627] A "filtering database" refers to a database that collects keywords, URLs, metadata, etc. that contain inappropriate content and provides basic data for evaluating the appropriateness of the content accessed by users.

[0628] "Real-time evaluation" refers to the process of evaluating and analyzing an action or event almost simultaneously when that action or event occurs.

[0629] "Educational content" refers to information, teaching materials, applications, etc. intended for learning and knowledge acquisition.

[0630] "Usage history" refers to a record of the operations performed by the user and the content accessed.

[0631] "Remote monitoring" refers to monitoring the status of devices or systems from a physically distant location using communication technology.

[0632] "Feedback" refers to providing evaluation, guidance, and advice to others regarding usage and behavior.

[0633] A "server" refers to a computer system that provides services and data to other computers (clients) on a network.

[0634] "Terminal" refers to a computer or mobile device that is directly operated by a user.

[0635] First, this system operates with three main components: a server, a terminal, and a user. Specific embodiments of each component will be described in detail below.

[0636] Server-side implementation

[0637] Collecting and updating inappropriate content lists

[0638] The server first collects a list of inappropriate content from the Internet and updates the database. This is done using web crawling technology. Specifically, it uses Python libraries such as Scrapy and BeautifulSoup to collect inappropriate keywords, URLs, and metadata from various sites. This allows the server to always have the latest threat information.

[0639] Generative AI Training

[0640] The server then uses the large amount of data to train a generative artificial intelligence (AI) model, using tools such as Hugging Face's Transformers library, TensorFlow, and PyTorch. The generative AI model learns from the collected data and is able to detect inappropriate content with high accuracy even when new data is used.

[0641] Terminal side embodiment

[0642] Download the filtering database

[0643] The device periodically downloads the filtering database from the server and stores it in local storage. This operation is performed via an API call, and the downloaded database is saved in a local database such as SQLite so that it can be used the next time the device is accessed.

[0644] Real-time content evaluation

[0645] When a user attempts to access a website or app, the device references a locally stored filtering database and uses generative AI to evaluate the appropriateness of the content in real time. For example, using AI models using Scikit-learn and Keras, the device immediately determines whether the content is inappropriate. If it is determined to be inappropriate, the content is immediately blocked and a warning message is displayed to the user.

[0646] Educational content recommendations

[0647] The device uses generative AI to recommend appropriate educational content based on the user's usage history and interests. Recommended content is not only displayed on the device in real time, but also notified to the parent's mobile app. For example, a message may appear saying, "New educational content: Math games recommended."

[0648] User-Side Embodiment

[0649] Remote Parental Monitoring

[0650] Users (parents) can monitor their children's smartphone usage in real time through a dedicated mobile application or web portal. This monitoring includes detailed data such as apps used, websites visited, screen time, etc. The mobile application syncs and displays the data using cloud services such as Firebase and AWS.

[0651] Providing appropriate feedback

[0652] Based on this data, parents can provide appropriate feedback on their children's smartphone usage. With the help of generative AI, parents can send gentle guidance and advice to their children, such as "Try to focus more on your studies," via text message.

[0653] Examples of concrete examples and prompts

[0654] Website Filtering

[0655] User: Your child tries to visit a new website.

[0656] On the device: Website URLs are matched against a local filtering database and generative AI analyzes the content.

[0657] Generative AI: Determines whether a website contains inappropriate content.

[0658] Device: Blocks website access and displays "This site is inappropriate."

[0659] Example prompt: "Please rate whether a new website you visit is inappropriate."

[0660] Educational content recommendations

[0661] Users: Children are interested in online learning.

[0662] Device: Generative AI recommends appropriate educational content based on a child's interests and past usage history.

[0663] Device: Sends a notification to the parent app saying "New educational content: Math learning game recommended."

[0664] Example prompt: "Recommend appropriate educational content based on your child's interests."

[0665] Remote Monitoring

[0666] User (Parent): Check your child's usage on the mobile app.

[0667] Device: View apps used, websites visited, and screen time history.

[0668] User (parent): Sends appropriate advice to the child based on suggestions from the generative AI.

[0669] Example prompt: "Please provide appropriate feedback regarding your child's smartphone usage."

[0670] Through these steps, the system can keep children's digital environments safe and educationally rewarding, while allowing parents to effectively monitor and provide feedback remotely.

[0671] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0672] Step 1:

[0673] Collecting and updating inappropriate content lists

[0674] The server collects inappropriate content lists from across the Internet, including inappropriate keywords, website URLs, and content metadata.

[0675] Input: Multiple sources on the internet

[0676] Data processing: Collect data using web crawling techniques (such as Python's Scrapy or BeautifulSoup).

[0677] Output: Save and update the inappropriate content list in the database.

[0678] Specific behavior:

[0679] 1. The server accesses the list of web pages to be crawled.

[0680] 2. Extract inappropriate keywords, URLs, and metadata from each page.

[0681] 3. This data will be centrally managed and added / updated to the inappropriate content list database.

[0682] Step 2:

[0683] Generative AI Training

[0684] The server trains generative AI models using large amounts of text and other media data.

[0685] Input: Large collected dataset

[0686] Data processing: Training AI models using libraries such as Hugging Face Transformers, TensorFlow, and PyTorch

[0687] Output: A trained generative AI model

[0688] Specific behavior:

[0689] 1. The server supplies the collected data to the AI ​​model.

[0690] 2. AI models use deep learning algorithms to analyze and learn from data.

[0691] 3. The trained model is saved on the server and updated as needed.

[0692] Step 3:

[0693] Download the filtering database

[0694] The terminal periodically downloads the filtering database from the server and stores it in local storage.

[0695] Input: Server filtering database

[0696] Data processing: Download by API call and save to local database (e.g. SQLite)

[0697] Output: Locally stored filtering database

[0698] Specific behavior:

[0699] 1. The device sends a request to the server API.

[0700] 2. The server responds with the latest filtering database.

[0701] 3. The device saves the downloaded database in local storage.

[0702] Step 4:

[0703] Real-time content evaluation

[0704] The device uses generative AI to evaluate the appropriateness of content in real time when a user attempts to access a website or app.

[0705] Input: URL or text data of the content you want to access

[0706] Data calculation: Matching with local filtering database and analysis by AI evaluation model

[0707] Output: Evaluation result: Good or bad

[0708] Specific behavior:

[0709] 1. The device receives the URL and text data of the content the user wants to access.

[0710] 2. Match with a local database and use a generative AI model to analyze the content.

[0711] 3. If the content is deemed inappropriate, access will be blocked and a warning message will be displayed.

[0712] Step 5:

[0713] Educational content recommendations

[0714] The device recommends appropriate educational content based on the user's usage history and interests.

[0715] Input: Usage history, interest data

[0716] Data processing: Content recommendation using usage history analysis and generative AI

[0717] Output: Recommended educational content

[0718] Specific behavior:

[0719] 1. The device analyzes the user's usage history and interest data.

[0720] 2. Generative AI selects the most appropriate educational content based on this data.

[0721] 3. The recommendation results are displayed on the user's device and also notified to the parent's app.

[0722] Step 6:

[0723] Remote Parental Monitoring

[0724] Parents can monitor their children's smartphone usage in real time through a dedicated mobile app.

[0725] Input: Child usage data (apps used, websites visited, screen time)

[0726] Data Processing: Collection and Display of Usage Data

[0727] Output: Display data in the parent mobile application

[0728] Specific behavior:

[0729] 1. Collect usage data from your child's device.

[0730] 2. Synchronize the collected data with a cloud service (Firebase or AWS).

[0731] 3. Display data in real time on parent's mobile app.

[0732] Step 7:

[0733] Providing appropriate feedback

[0734] Parents provide appropriate feedback based on the collected data, and text messages are sent based on generative AI suggestions.

[0735] Input: Collected usage data and generative AI feedback suggestions

[0736] Data processing: generating feedback suggestions

[0737] Output: Feedback as a text message

[0738] Specific behavior:

[0739] 1. Parents can check their children's usage within the mobile app.

[0740] 2. Generative AI automatically generates appropriate feedback and advice.

[0741] 3. The parent will then send the suggested content as a text message to their child.

[0742] (Application example 1)

[0743] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0744] In today's digital society, children have more opportunities to access a variety of media content. However, some media content is inappropriate for children, making it difficult for parents to monitor all content. Furthermore, there is a lack of systems that efficiently provide educational media content to support children's healthy development. There is a need to develop a system that can solve these problems and help children grow up in a safe and beneficial digital environment.

[0745] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0746] In this invention, the server includes a means for periodically collecting the latest inappropriate media content list from the Internet and updating the database, a means for training a generative artificial intelligence model using a large amount of text data and other media data, and a means for evaluating the appropriateness of media content accessed by users in real time. This allows parents to remotely monitor and provide feedback while maintaining a healthy digital environment for their children. The generative artificial intelligence model can also recommend educational media content based on children's interests.

[0747] "Generative AI" is an AI that learns from large amounts of data and generates information in a way that can respond to new data.

[0748] "Inappropriate media content" means media content that contains violent, sexual, discriminatory, or otherwise harmful to children.

[0749] A "media content list" identifies inappropriate media content and manages the list as a database.

[0750] The "means for assessing appropriateness" is a system that has the function of analyzing and assessing whether media content is inappropriate.

[0751] "Educational media content" is media content that is beneficial to children's learning and development and has high educational value.

[0752] "User usage status" refers to information such as what content the user has viewed, how long the user has viewed it, and what interests the user has.

[0753] "Remote monitoring capabilities" are features that allow parents or supervisors to monitor their children's digital environments remotely and in real time.

[0754] The "means of providing feedback" is a function that allows parents to send appropriate guidance and advice to their children with the assistance of generative artificial intelligence.

[0755] A "device for real-time analysis and blocking" is a device that has the capability to analyze media content on the fly and immediately block it if necessary.

[0756] A "generative artificial intelligence model" is a learning model that allows generative artificial intelligence to suggest optimal media content based on data.

[0757] This invention is a system for detecting and blocking inappropriate media content using generative artificial intelligence to maintain a healthy digital environment for children. Specific embodiments of this system are described below.

[0758] 1. Server-side implementation

[0759] The server has the means to perform the following functions:

[0760] Collecting and updating inappropriate media content lists

[0761] Server: Regularly collects the latest inappropriate media content list from the Internet and updates the database. This list includes inappropriate keywords, website URLs, and content metadata, allowing it to respond to the latest threat information.

[0762] Generative AI Training

[0763] Server: Trains generative artificial intelligence models using large amounts of text and other media data. This process uses natural language processing (NLP) techniques and deep learning algorithms. The trained AI continues to learn based on new data, resulting in more accurate detection capabilities.

[0764] 2. Terminal side embodiment

[0765] The terminal has the means to perform the following functions:

[0766] Download the filtering database

[0767] Device: Periodically downloads the filtering database from the server and stores it in local storage. This database provides the foundation data used to evaluate the appropriateness of media content.

[0768] Real-time content evaluation

[0769] On the device: When a user attempts to access media content, the generative AI will refer to a locally stored filtering database and assess the appropriateness of the content in real time. Content deemed inappropriate will be immediately blocked and a warning message will be displayed to the user.

[0770] Educational content recommendations

[0771] On-device: Generative AI will recommend appropriate educational content based on the user's usage history and interests. These recommendations will be customized to the user's learning style and interests and displayed on the smartphone screen.

[0772] 3. User-side implementation

[0773] The user (parent) has the means to perform the following functions:

[0774] Remote Parental Monitoring

[0775] Users (Parents): Through a dedicated mobile application or web portal, they can monitor their children's device usage in real time, including data such as apps used, websites visited, and screen time.

[0776] Providing appropriate feedback

[0777] User (Parent): Parents can use the collected data to provide feedback on their children's device usage, with the assistance of generative AI, which can then send appropriate guidance and advice via text message.

[0778] Specific examples

[0779] Real-time evaluation of media content

[0780] User: A child tries to access a new video.

[0781] On device: Video frames are analyzed in real time, and generative AI detects inappropriate content.

[0782] Generative AI: Determines whether a video contains inappropriate content.

[0783] Device: Immediately stop playing the video and display "This content is inappropriate."

[0784] Educational content recommendations

[0785] Users: Children are interested in online learning.

[0786] Device: Generative AI recommends appropriate educational content based on a child's interests and past usage history.

[0787] Device: Sends a notification to the parent app saying "New educational content: Math game recommended."

[0788] Remote Monitoring

[0789] User (Parent): Check your child's usage on the mobile app.

[0790] Device: View apps used, websites visited, and screen time history.

[0791] User (parent): Based on the generative AI's suggestions, send advice to the child such as "This anime is violent, so please watch other educational content."

[0792] Prompt Sentence Examples

[0793] Below are some example prompts for recommending appropriate educational content to generative AI:

[0794] "Please recommend appropriate educational content based on my child's interests. Past viewing has been dominated by action anime, so please suggest more peaceful educational content."

[0795] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0796] Step 1:

[0797] Collecting and updating inappropriate media content lists

[0798] Server: The server periodically collects data about objectionable media content from multiple sources across the internet, which may involve web scraping techniques. Data collected may include objectionable keywords, website URLs, and metadata.

[0799] Input: Inappropriate content information on the internet.

[0800] Output: An updated list of inappropriate media content.

[0801] What it does: The crawler visits specific sites on the internet, extracts data, stores this data in a database on the server, and updates the inappropriate media content list.

[0802] Step 2:

[0803] Generative AI Training

[0804] Server: The server trains generative artificial intelligence models using large amounts of text and other media data, a process that uses natural language processing (NLP) techniques and deep learning algorithms.

[0805] Input: Large amounts of text and media data.

[0806] Output: A highly accurate generative artificial intelligence model.

[0807] What it does: Train AI with a dataset, validate the results, and update the model. For example, run deep learning algorithms using Python and TensorFlow.

[0808] Step 3:

[0809] Download the filtering database

[0810] Device: The device periodically downloads the filtering database from the server and stores it in local storage.

[0811] Input: Filtering database on the server.

[0812] Output: Filtering database stored in the device's local storage.

[0813] Specific operation: The device sends an API request to retrieve the latest database from the server, and then saves the database in the device's storage, making it available for offline reference.

[0814] Step 4:

[0815] Real-time content evaluation

[0816] Device: When a user attempts to access media content, the device consults a locally stored filtering database, and generative AI assesses the content's appropriateness in real time.

[0817] Input: Media content data and local filtering database.

[0818] Output: Content appropriateness assessment result (appropriate / inappropriate).

[0819] What it does: The device captures the media content being played frame by frame and passes it to a generative AI for analysis. If the content is deemed inappropriate, it blocks it and displays a warning message to the user.

[0820] Step 5:

[0821] Educational content recommendations

[0822] Device: Generative AI recommends appropriate educational content based on the user's usage history and interests.

[0823] Input: User usage history and interest data.

[0824] Output: Recommended educational content.

[0825] Specific operation: The device sends prompts to the generative AI to generate educational content based on the user's interests. For example, a prompt such as, "Please recommend appropriate educational content based on my child's interests. Most of his past viewings have been action anime, so please suggest more peaceful educational content."

[0826] Step 6:

[0827] Remote Parental Monitoring

[0828] Users (Parents): Monitor their children's device usage in real time through a dedicated mobile application or web portal.

[0829] Input: Child's device usage data.

[0830] Output: Monitoring results displayed on the parent device.

[0831] What it does: The app sends usage data from the child's device to a server. The application on the parent's device receives the data and visualizes it in the form of a dashboard.

[0832] Step 7:

[0833] Providing appropriate feedback

[0834] User (Parent): Parents use the collected data to provide feedback on their children's device usage, with the assistance of generative AI to send appropriate guidance and advice via text message.

[0835] Input: Parent feedback and generative AI advice.

[0836] Output: A text message sent to the child's device.

[0837] How it works: Parents use the app to send prompts to the generative AI, which then generates appropriate messages, such as "This cartoon is violent, so please watch other educational content" and send it to their child's device.

[0838] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0839] This invention is a system that uses generative artificial intelligence (AI) and an emotion engine to detect and block inappropriate content and maintain a healthy digital environment for children. Furthermore, this system analyzes the emotional state of the user (child), recommends appropriate educational content based on the user's emotion, and provides feedback support to parents.

[0840] 1. Server-side implementation

[0841] Collecting and updating inappropriate content lists

[0842] Server: The server periodically collects the latest inappropriate content list from the internet and updates the database. This list includes inappropriate keywords, website URLs, and metadata, and is collected using web scraping technology and APIs.

[0843] Generative AI Training

[0844] Server: The server trains the generative AI using large amounts of text and media data, using natural language processing (NLP) techniques and deep learning algorithms (e.g., Transformer-based models). The trained AI continues to learn based on new data, achieving highly accurate detection capabilities.

[0845] 2. Terminal side embodiment

[0846] Download the filtering database

[0847] Device: The device periodically sends a request to the server to download the latest filtering database and save it in local storage.

[0848] Real-time filtering

[0849] On your device: Every time you access a website or app, we use a local filtering database and generative AI to analyze the content, detect and block inappropriate content.

[0850] Implementing the Emotion Engine

[0851] Device: The device uses sensors such as a camera and microphone to capture the user's (child's) facial expressions and voice, and sends the data to the emotion engine, which analyzes the data and determines the user's emotional state in real time.

[0852] Educational content recommendations

[0853] Device: Based on the user's usage history, interests, and emotional state, generative AI selects the most appropriate educational content and notifies the device. For example, if the user is feeling stressed, it will recommend relaxation content, and if the user is concentrating on a learning topic they are interested in, it will recommend learning reinforcement content.

[0854] 3. User-side implementation

[0855] Remote Parental Monitoring

[0856] User (Parent): Parents can monitor their children's smartphone usage and emotional state in real time through a dedicated mobile app or web portal. Data collected includes apps used, websites visited, screen time, and emotional state.

[0857] Providing appropriate feedback

[0858] User (Parent): Parents can provide appropriate feedback with the support of generative AI based on their child's smartphone usage and emotional state. When a parent enters feedback and presses the send button, the feedback is displayed on the child's device in real time. For example, if the child is stressed, they can send an encouraging message, or if the child is concentrating on their studies, they can send a message of encouragement.

[0859] Specific examples

[0860] Website Filtering

[0861] User (child): A child tries to access a new website.

[0862] On the device: Website URLs are matched against a local filtering database and generative AI analyzes the content.

[0863] Generative AI: Determines whether a website contains inappropriate content.

[0864] Device: Blocks website access and displays "This site is inappropriate."

[0865] Educational content recommendations

[0866] User (child): Children are interested in online learning.

[0867] Device: Generative AI recommends appropriate educational content based on a child's interests, past usage history, and emotional state.

[0868] Device: Sends a notification to the parent app saying "New educational content: Math game recommended."

[0869] Real-time emotion monitoring and feedback

[0870] User (child): The emotion engine detects stress levels while children are learning remotely.

[0871] Device: A notification will be sent to the parent's device saying "Your child is stressed."

[0872] User (parent): Create an encouraging message remotely and press the send button, and the message "Good luck!" will appear in real time on the child's device.

[0873] In this way, the system maintains a healthy digital environment for children, while enhancing educational value and providing parents with tools for effective remote monitoring and feedback.

[0874] The processing flow will be explained below.

[0875] Step 1:

[0876] Server: Collects inappropriate content and updates the database. The server collects the latest inappropriate content list from the Internet using scraping or APIs, and periodically updates the filtering database. This database includes inappropriate keywords, URLs, and metadata.

[0877] Step 2:

[0878] Server: Trains the generative AI model. The server trains the generative AI using large amounts of text and media data. Using natural language processing (NLP) techniques and deep learning algorithms (e.g., Transformer-based models), the AI ​​continues to learn based on new data.

[0879] Step 3:

[0880] Device: Downloads the filtering database. The device periodically sends a request to the server to download the latest filtering database and saves it in local storage. This database is used to evaluate the appropriateness of content.

[0881] Step 4:

[0882] On the device: Every time a user attempts to access a website or app, the content is evaluated for appropriateness. The URL is checked against a local filtering database, and generative AI analyzes the content in real time. If it is deemed inappropriate, it is processed in the next step.

[0883] Step 5:

[0884] Device: Blocks inappropriate content. If real-time analysis determines that content is inappropriate, a warning message will be displayed to the user stating "This site is inappropriate." Access will be blocked.

[0885] Step 6:

[0886] Device: Implementation of the emotion engine. The device's camera and microphone are used to capture the user's (child's) facial expressions and voice, and the data is sent to the emotion engine. The emotion engine analyzes the data and determines the user's emotional state in real time.

[0887] Step 7:

[0888] Device: Recommends educational content. Based on the user's usage history, interests, and emotional state, generative AI selects the most appropriate educational content and notifies the device. For example, if the user is feeling stressed, it will recommend relaxation content.

[0889] Step 8:

[0890] Device: Display recommended educational content on the user's (child's) screen. Promote access to educational content via push notifications and dedicated pages.

[0891] Step 9:

[0892] User (Parent): Monitor their child's smartphone usage and emotional state in real time using a dedicated mobile app or web portal. Data collected includes apps used, websites visited, screen time, and emotional state.

[0893] Step 10:

[0894] User (Parent): Provides feedback on the child's smartphone usage and emotional state. With the support of generative AI, the parent enters the feedback and presses the send button, and the feedback is displayed in real time on the child's device. For example, if stress is detected, an encouraging message is sent, and if the child is concentrating on their studies, a supportive message is sent.

[0895] Example 2

[0896] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0897] In today's digital environment, there is a lot of inappropriate content available online, which can have a negative impact on children. It is also difficult to understand in real time what content children are interested in and their emotional state, making it difficult to provide appropriate educational content. Furthermore, there is a lack of means for parents to remotely monitor their children's digital behavior and provide appropriate feedback. Therefore, there is a need for a system that integrates the detection and blocking of inappropriate content, analysis of children's emotions, recommendation of educational content, and remote monitoring and feedback provision functions.

[0898] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for detecting and blocking inappropriate content using generative artificial intelligence, means for updating an inappropriate content list based on collected data, means for evaluating the appropriateness of content a user attempts to access based on the inappropriate content list, means for analyzing a user's emotional state using a camera and a microphone, means for recommending educational content based on the user's emotional state and usage status, means for collecting and remotely monitoring the user's usage status and emotional state, and means for parents to provide feedback. This not only enables children to enjoy learning and entertainment in a safe and healthy digital environment, but also enables parents to effectively monitor their children's digital behavior and provide appropriate feedback.

[0899] "Generative AI" is an AI technology that learns using large amounts of data and generates and analyzes content based on new data.

[0900] "Inappropriate Content" means any information or media that may be harmful to children or other users, such as violent, pornographic, or libelous material.

[0901] "Updating" refers to adding new information to existing data or lists and keeping them up to date.

[0902] A "filtering database" is a database that accumulates information about inappropriate content and uses that data to evaluate the appropriateness of the content.

[0903] "Evaluating the appropriateness of content" means determining whether the content a user is attempting to access is inappropriate.

[0904] "Camera and microphone" refers to a photographing and recording device for capturing the user's facial expressions and voice.

[0905] An "emotional state" refers to a user's mental and emotional state, and may include, for example, stress, joy, excitement, and the like.

[0906] "Analyzing" refers to extracting and understanding specific information from collected data.

[0907] "Educational content" is information or media intended to educate and increase the user's knowledge.

[0908] "Usage" refers to information about how users use devices and applications.

[0909] "Enabling remote monitoring" means being able to check the device usage status and the user's emotional state from a remote location.

[0910] "Providing feedback" means conveying advice or comments based on the user's actions or situation.

[0911] An "inappropriate content list" is a collection of information about inappropriate content that exists on the Internet.

[0912] This invention is a system that uses generative artificial intelligence (AI) and an emotion engine to detect and block inappropriate content and maintain a healthy digital environment for children. Furthermore, this system analyzes the emotional state of the user (child), recommends appropriate educational content based on the user's emotion, and provides feedback support to parents. Specific embodiments are described below.

[0913] Server-side implementation

[0914] Collecting and updating inappropriate content lists

[0915] The server periodically collects the latest inappropriate content list from the Internet and updates the database. This list includes inappropriate keywords, website URLs, and metadata. The server retrieves the data using a web scraping library such as Python's BeautifulSoup or Scrapy and stores it in an SQL database (e.g., MySQL or PostgreSQL). This allows the server to always maintain the latest inappropriate content information.

[0916] Generative AI Training

[0917] The server trains the generative AI using large amounts of text and media data. Deep learning frameworks used include TensorFlow and PyTorch, and Transformer-based models (e.g., BERT and GPT-3) are used. Natural language processing (NLP) techniques and deep learning algorithms are used for training, and the AI ​​continues to learn based on new data.

[0918] Terminal side embodiment

[0919] Download the filtering database

[0920] The device periodically sends a request to the server to download the latest filtering database and save it in local storage. The device periodically communicates with the server using the HTTP protocol to keep the database up to date.

[0921] Real-time filtering

[0922] When a user attempts to access a website or app, the device uses a local filtering database and generative AI to analyze the content, detect inappropriate content, and block it. Specifically, an AI model running on the device evaluates the content in real time and blocks access if it is deemed inappropriate. For example, when a user (child) attempts to access a new website, the device compares the URL with the local filtering database and the generative AI analyzes the content. If it is deemed inappropriate, the device blocks access to the website and displays the message, "This site is inappropriate."

[0923] Implementing the Emotion Engine

[0924] The device uses sensors such as a camera and microphone to capture the user's facial expressions and voice and send them to the emotion engine. Using libraries such as FaceAPI and AudioAnalyzer, this data is analyzed in real time to determine the user's emotional state. For example, if the emotion engine detects a child is stressed while learning remotely, the parent's device will receive a notification saying, "Your child is feeling stressed."

[0925] Educational content recommendations

[0926] Based on the user's usage history, interests, and emotional state, the generative AI selects the most suitable educational content and notifies the device. For example, if a user (child) shows interest in online learning, the generative AI will recommend appropriate educational content based on their interests, past usage history, and emotional state. The selected content is notified to the device and a message appears saying, "New educational content: Math game recommended."

[0927] User-side implementation (parent)

[0928] Remote Monitoring

[0929] Parents can monitor their children's smartphone usage and emotional state in real time through a dedicated mobile app or web portal. Collected data is displayed using Firebase and cloud services. This feature allows parents to understand their children's current situation and take appropriate action if necessary.

[0930] Providing feedback

[0931] Parents can provide appropriate feedback with the support of generative AI based on their child's smartphone usage and emotional state. When a parent enters the feedback and presses the send button, the feedback is displayed on the child's device in real time. For example, if the child is stressed, they can send an encouraging message, or if they are concentrating on their studies, they can send a message of support.

[0932] Example prompt sentence:

[0933] Prompt: "Please recommend some new educational content. My child is interested in math and has been feeling stressed lately. I'd like some content that's relaxing yet educational."

[0934] The system will enhance educational value while maintaining a healthy digital environment for children, allowing parents to effectively monitor and provide feedback remotely.

[0935] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0936] Step 1:

[0937] Server: Collects inappropriate content lists from the Internet and updates the database.

[0938] Input: Web page data obtained from the Internet

[0939] Processing: The server uses Python's BeautifulSoup or Scrapy library to perform web scraping and extract inappropriate keywords, URLs, and metadata from web page data. The extracted data is stored in an SQL database (e.g., MySQL or PostgreSQL).

[0940] Output: Updated inappropriate content list

[0941] Specific Actions: The server runs a scheduled job (e.g., once daily) that crawls designated websites and stores inappropriate content information in a database.

[0942] Step 2:

[0943] Server: Trains the generative AI model.

[0944] Input: Text data, media data

[0945] Processing: Use TensorFlow or PyTorch to train Transformer-based models (e.g., BERT or GPT-3). Use natural language processing (NLP) techniques and deep learning algorithms to learn from large amounts of data.

[0946] Output: A trained generative AI model

[0947] What it does: The server loads the training data and trains the deep learning model. After the training period is over, the optimized model is saved and deployed as an inappropriate content detection algorithm.

[0948] Step 3:

[0949] On your device: Download the filtering database.

[0950] Input: Latest filtering database on the server

[0951] Processing: The device communicates with the server using an HTTP request and downloads the latest filtering database.

[0952] Output: Filtering database stored in local storage

[0953] Specific operation: The device executes a periodic job (for example, once a day) and sends an HTTP GET request to the server. The device receives the latest database from the server and stores it in local storage.

[0954] Step 4:

[0955] Device: Filter content in real time.

[0956] Input: URL and content data of the website or app the user is trying to access

[0957] Processing: The device analyzes the content using a local filtering database and generative AI, blocking access if it is deemed inappropriate.

[0958] Output: Filtered content (allowed or blocked)

[0959] How it works: When a user opens a browser and accesses a new website, the device sends the URL to a local filtering database and generative AI for analysis. If it is deemed inappropriate, the device blocks access and displays the message "This site is inappropriate."

[0960] Step 5:

[0961] Device: Uses camera and microphone to analyze the user's emotional state.

[0962] Input: User's facial expression data (still image or video), voice data

[0963] Processing: The device sends this data to the emotion engine, which analyzes it using FaceAPI and AudioAnalyzer to determine the user's emotional state (stress, joy, excitement, etc.).

[0964] Output: Information about the user's emotional state

[0965] How it works: The device's camera captures the user's face and the microphone records their voice. The emotion engine analyzes this data in real time to determine their emotional state. For example, if the user is feeling stressed, that information is stored on the device as an analysis result.

[0966] Step 6:

[0967] Device: Recommend educational content.

[0968] Input: User usage history, interest data, emotional state data

[0969] Processing: The generative AI selects the most appropriate educational content based on the above input data.

[0970] Output: Recommended educational content

[0971] Specific operation: The device uses generative AI to select the most suitable educational content based on the user's usage history, interests, and emotional state. The selected content is notified to the device, and a message appears saying, "New educational content: Math game recommended."

[0972] Step 7:

[0973] Parents (users): Remotely monitor their children's smartphone usage and emotional state.

[0974] Input: Child usage data, emotional state data

[0975] Processing: Parents can view this data in real time via a dedicated mobile app or web portal.

[0976] Output: Displayed usage and emotional state

[0977] What it does: When parents open the mobile app, they can see their child's current usage and emotional state in real time, allowing them to take necessary actions.

[0978] Step 8:

[0979] Parents (users): Provide feedback such as encouragement and advice.

[0980] Input: Child usage and emotional state data, parent-entered feedback

[0981] What happens: Parents create feedback in the app or web portal, press the submit button, and the feedback is sent in real time to their child's device.

[0982] Output: Feedback message displayed on the child's device

[0983] How it works: Parents can create an encouraging message such as "Good luck!" using the app and press the "Send" button, and the message will instantly appear on their child's device.

[0984] Through each of these steps, the system maintains a healthy digital environment for children, enhances educational value, and allows parents to effectively monitor and provide feedback remotely.

[0985] (Application example 2)

[0986] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0987] In recent years, the use of the Internet and digital devices has become more widespread, increasing the risk that children will be exposed to inappropriate media content. There is also a need for a method to recommend educationally valuable media to children based on their emotional state while maintaining a healthy digital environment. It is also difficult for parents to remotely monitor their children's usage and emotional state in real time and provide appropriate feedback. Therefore, the present invention aims to solve these issues by utilizing generative artificial intelligence and an emotion engine, and to provide a system that provides educational value while maintaining a healthy digital environment for children.

[0988] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for detecting and blocking inappropriate media using generative artificial intelligence, means for updating an inappropriate media list based on collected data, means for evaluating the appropriateness of media a user attempts to access based on the inappropriate media list, means for recommending educational media, means for collecting and remotely monitoring user usage, means for parents to provide feedback, and means for analyzing a user's emotional state and recommending appropriate media according to that emotional state. This reduces the risk of children being exposed to inappropriate media and makes it possible to provide media with educational value according to the user's emotional state. It also makes it easy for parents to remotely monitor their children's usage and emotional state in real time and provide appropriate feedback.

[0989] "Generative AI" is a technology that uses natural language processing and deep learning algorithms to automatically generate content based on new data, and then analyze and evaluate it.

[0990] "Inappropriate media" is digital content, such as text, images, audio, or video, that is deemed harmful or inappropriate for children.

[0991] An "inappropriate media list" is a list containing inappropriate keywords, URLs, and metadata collected from across the Internet.

[0992] "Educational media" refers to digital content such as text, images, audio, and video that supports children's education and learning.

[0993] "Emotional state" is data that indicates the psychological state of the user, detected using sensors such as a camera or microphone.

[0994] "Remote monitoring" is a feature that allows parents to monitor their children's digital device usage in real time from a distance using dedicated devices or applications.

[0995] "Feedback" is the act of parents sending messages of encouragement or warning based on their children's digital device usage and emotional state.

[0996] "Recommendation" is the act of generative artificial intelligence selecting and presenting appropriate media content based on the user's interests and emotional state.

[0997] MODE FOR CARRYING OUT THE INVENTION

[0998] This invention combines generative artificial intelligence (AI) and an emotion engine to detect and block inappropriate media in virtual stores, recommend products based on emotional states, and provide remote monitoring and feedback. The system ensures a safe digital environment for children and recommends appropriate products based on their emotional states.

[0999] System Configuration

[1000] server

[1001] Collecting and updating inappropriate media lists

[1002] Generative AI Training

[1003] Managing the filtering database

[1004] Terminal

[1005] Real-time detection and blocking of inappropriate media

[1006] Analysis of user emotional states using an emotion engine

[1007] Educational and relevant product recommendations

[1008] Remote monitoring data collection

[1009] User

[1010] Parents can monitor their children's usage and emotional state in real time through a dedicated application.

[1011] Parents provide appropriate feedback

[1012] Server-side implementation

[1013] The server performs the following functions:

[1014] Collecting and updating inappropriate media lists

[1015] The server periodically collects the latest inappropriate media list from the Internet and updates the database, which includes inappropriate keywords, website URLs, and metadata, using web scraping technology and APIs.

[1016] Generative AI Training

[1017] The server trains the generative AI using large amounts of text and media data, using natural language processing (NLP) techniques and deep learning algorithms (e.g., Transformer-based models). The trained AI continues to learn based on new data, achieving highly accurate detection capabilities.

[1018] Terminal side embodiment

[1019] The terminal performs the following functions:

[1020] Download the filtering database

[1021] The device periodically sends a request to the server to download the latest filtering database and store it in local storage.

[1022] Real-time filtering

[1023] Every time a user (child) tries to access a product in a virtual store, the content is analyzed using a local filtering database and generative AI to detect and block inappropriate content.

[1024] Implementing the Emotion Engine

[1025] The device uses sensors such as a camera and microphone to capture the user's (child's) facial expressions and voice, and sends the data to the emotion engine, which analyzes the data and determines the user's emotional state in real time.

[1026] Educational content recommendations

[1027] Based on the user's usage history, interests, and emotional state, generative AI selects the most suitable educational content and products and notifies the device. For example, if the user is feeling stressed, it will recommend relaxation products, and if the user is concentrating on a learning content that interests them, it will recommend learning reinforcement content.

[1028] User-Side Embodiment

[1029] Remote Parental Monitoring

[1030] Parents can monitor their children's smartphone usage and emotional state in real time through a dedicated mobile app or web portal. Data collected includes apps used, websites visited, screen time, and emotional state.

[1031] Providing appropriate feedback

[1032] Parents can provide appropriate feedback with the support of generative AI based on their child's smartphone usage and emotional state. When a parent enters the feedback and presses the send button, the feedback is displayed on the child's device in real time. For example, if the child is stressed, they can send an encouraging message, or if they are concentrating on their studies, they can send a message of support.

[1033] Specific examples

[1034] Website Filtering

[1035] A user (child) attempts to access a new website. The inappropriate media detected is processed as follows:

[1036] The URL is checked against a local filtering database

[1037] Generative AI analyzes content, and if it determines that a website is inappropriate media, it blocks it from being displayed and notifies users that "this site is inappropriate."

[1038] Educational content recommendations

[1039] Users (children) become interested in online learning

[1040] Generative AI suggests appropriate educational content and products based on user interests and past usage history

[1041] A notification is sent to the parent app saying, "New educational content: Math games recommended."

[1042] Real-time emotion monitoring and feedback

[1043] The emotion engine detects that the user (child) is feeling stressed during remote learning.

[1044] Parents receive a notification on their device that their child is feeling stressed.

[1045] Parents can remotely create encouraging messages and press the send button, which will then appear on their child's device with the message "Good luck!"

[1046] Prompt Sentence Examples

[1047] Prompt the generative AI model:

[1048] What products do you recommend for users when they are feeling stressed?

[1049] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1050] Step 1:

[1051] The server collects and periodically updates an inappropriate media list from the Internet. This list includes inappropriate keywords, URLs, and metadata. The server uses web scraping technology and public APIs to collect the latest data from the Internet and adds it to the filtering database, thereby keeping the inappropriate media list up to date.

[1052] Input: Data on the Internet

[1053] Output: Updated inappropriate media list

[1054] Step 2:

[1055] The server uses large amounts of text and media data to train the generative AI, using natural language processing (NLP) techniques and deep learning algorithms. For example, a Transformer-based model is used to train the AI ​​to detect inappropriate media with high accuracy. The trained AI model then continues to learn and improve based on new data.

[1056] Input: Large amounts of text and media data

[1057] Output: A trained generative AI model

[1058] Step 3:

[1059] The device periodically sends a request to the server to download the latest filtering database and save it in local storage, so that the device can always use the latest inappropriate media list.

[1060] Input: A request to the server

[1061] Output: Locally stored filtering database

[1062] Step 4:

[1063] When a user (child) tries to access a product in the virtual store, the device analyzes the content using a local filtering database and generative AI. The generative AI evaluates the product's text, images, and multimedia data to determine whether it contains inappropriate elements. If the generative AI determines that the product is inappropriate, it blocks the product and notifies the user, "This product is inappropriate."

[1064] Input: The product data the user is trying to access

[1065] Output: Inappropriate media blocking notification

[1066] Step 5:

[1067] The device uses sensors such as a camera and microphone to capture the user's (child's) facial expressions and voice, and sends the data to the emotion engine, which analyzes the sensor data and determines the user's emotional state in real time.

[1068] Input: Sensor data from camera and microphone

[1069] Output: Parsed user's emotional state

[1070] Step 6:

[1071] The device uses generative AI to recommend appropriate educational content and products based on the user's usage history, interests, and emotional state. For example, if the user is feeling stressed, it will recommend relaxation products, and if the user is concentrating on a learning topic they are interested in, it will recommend learning reinforcement content.

[1072] Input: User usage history, interests, emotional state

[1073] Output: Recommended educational content and products

[1074] Step 7:

[1075] The device collects the user's usage and emotional state and transmits it in real time to a dedicated parental application, through which parents can monitor their child's current usage and emotional state.

[1076] Input: User usage status, emotional state data

[1077] Output: Real-time data displayed in a dedicated parent application

[1078] Step 8:

[1079] The user (parent) creates a feedback message in a dedicated application based on the child's usage and emotional state, and presses the send button. The feedback is then displayed in real time on the child's device. For example, if the child is stressed, an encouraging message can be sent.

[1080] Input: Feedback message entered by the parent

[1081] Output: Messages displayed on the child's device in real time

[1082] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1083] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1084] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1085] [Third embodiment]

[1086] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1087] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1088] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1089] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1090] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1091] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1092] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1093] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1094] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific 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.

[1095] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1096] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1097] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. 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."

[1098] This invention is a system that uses generative artificial intelligence (AI) to detect and block inappropriate content and maintain a healthy digital environment for children. This system works in cooperation with different entities: a server, a device, and a user.

[1099] 1. Server-side implementation

[1100] Collecting and updating inappropriate content lists

[1101] Server: Regularly collects the latest inappropriate content lists from the Internet and updates the database. This list includes inappropriate keywords, website URLs, and content metadata, allowing it to respond to the latest threat information.

[1102] Generative AI Training

[1103] Server: Trains generative AI models using large amounts of text and other media data. This process uses natural language processing (NLP) techniques and deep learning algorithms. The trained AI continues to learn based on new data, resulting in more accurate detection capabilities.

[1104] 2. Terminal side embodiment

[1105] Download the filtering database

[1106] Device: Periodically downloads the filtering database from the server and stores it in local storage. This database provides the basis for assessing the appropriateness of content.

[1107] Real-time content evaluation

[1108] On the device: When a user attempts to access a website or app, the generative AI consults a locally stored filtering database and evaluates the content's appropriateness in real time. Content deemed inappropriate is immediately blocked and a warning message is displayed to the user.

[1109] Educational content recommendations

[1110] On-device: Generative AI will recommend appropriate educational content offline and online based on the user's usage history and interests. These recommendations will be customized to the user's learning style and interests and displayed on the smartphone screen.

[1111] 3. User-side implementation

[1112] Remote Parental Monitoring

[1113] Users (Parents): Through a dedicated mobile application or web portal, they can monitor their children's smartphone usage in real time, including data such as apps used, websites visited, and screen time.

[1114] Providing appropriate feedback

[1115] User (Parent): Parents can use the collected data to provide feedback on their children's smartphone usage, with the assistance of generative AI, which can then send appropriate guidance and advice via text message.

[1116] Specific examples

[1117] Website Filtering

[1118] User: Your child tries to visit a new website.

[1119] On the device: Website URLs are matched against a local filtering database and generative AI analyzes the content.

[1120] Generative AI: Determines whether a website contains inappropriate content.

[1121] Device: Blocks website access and displays "This site is inappropriate."

[1122] Educational content recommendations

[1123] Users: Children are interested in online learning.

[1124] Device: Generative AI recommends appropriate educational content based on a child's interests and past usage history.

[1125] Device: Sends a notification to the parent app saying "New educational content: Math game recommended."

[1126] Remote Monitoring

[1127] User (Parent): Check your child's usage on the mobile app.

[1128] Device: View apps used, websites visited, and screen time history.

[1129] User (parent): Sends appropriate advice to the child based on suggestions from the generative AI.

[1130] Through these steps, the system not only provides features to help children stay safe in the digital environment, but also gives parents the tools to effectively monitor and provide feedback remotely, making their children's online experience safer and more educational.

[1131] The processing flow will be explained below.

[1132] Step 1:

[1133] Server: Collects inappropriate content and updates the database. The server collects inappropriate keywords, URLs, and metadata from the Internet and updates the filtering database. This collection is done using web scraping technology and APIs.

[1134] Step 2:

[1135] Server: Trains the generative AI model. The server uses large amounts of collected text and media data to train the generative AI. Natural language processing (NLP) techniques and deep learning algorithms (e.g., Transformer-based models) are used for training.

[1136] Step 3:

[1137] Device: Downloads the filtering database. The device periodically sends a request to the server, downloads the latest filtering database, and stores it in local storage.

[1138] Step 4:

[1139] Device: Monitors user website access and compares the URLs they attempt to access against a local filtering database. Generative AI analyzes the content of the webpage in real time and evaluates its appropriateness. If it is deemed inappropriate, it proceeds to the next step.

[1140] Step 5:

[1141] Device: Blocks inappropriate content. Real-time analysis is performed, and if content is deemed inappropriate, a warning message such as "This site is inappropriate" is displayed to the user and access is blocked.

[1142] Step 6:

[1143] Device: Recommends educational content. By analyzing the user's usage history, interests, and past search history, generative AI selects appropriate educational content and notifies the device.

[1144] Step 7:

[1145] Device: Display recommended educational content on the user's (child's) screen. Promote access to educational content through notifications, dedicated pages, etc.

[1146] Step 8:

[1147] User (Parent): Monitor your child's smartphone usage in real time using a dedicated mobile app or web portal, retrieve data from the server, and view activity reports.

[1148] Step 9:

[1149] User (Parent): Provides appropriate feedback on children's smartphone usage. Supported by generative AI, real-time guidance and comments are sent as text messages. When the parent enters the feedback and presses the send button, the message appears on the child's device.

[1150] Example 1

[1151] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1152] The internet is filled with a wealth of harmful and inappropriate content for children, making it difficult for parents to maintain a safe online environment for their children. Providing an environment where children can access appropriate educational content based on their interests is also a challenge. Furthermore, there is a lack of systems that allow parents to remotely monitor their children's online activity and provide appropriate feedback. This invention aims to comprehensively solve these problems and maintain a healthy digital environment for children.

[1153] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1154] In this invention, the server includes a means for collecting a list of inappropriate content from the Internet and updating the database, a means for training a generative artificial intelligence model using large amounts of data, and a means for providing a filtering database to the terminal. This enables the latest inappropriate content to be quickly detected and blocked. Furthermore, the filtering database downloaded from the server is stored in local storage, and the appropriateness of content the user attempts to access is evaluated and blocked in real time, thereby maintaining a safe digital environment for children. Furthermore, the system analyzes the user's usage history and can recommend appropriate educational content based on that analysis, thereby enhancing children's motivation to learn. Parents can remotely monitor their children's smartphone usage through a dedicated application and provide appropriate feedback, providing an educational and safe digital environment.

[1155] "Inappropriate Content" refers to content that is harmful to children and contains violence, pornography, discrimination, or other ethically objectionable information.

[1156] "Generative artificial intelligence" is a form of artificial intelligence that can learn from large amounts of data and make predictions and classifications based on new data, and specifically refers to models that involve data generation.

[1157] A "database" refers to a collection of information that is systematically organized and managed so that it can be stored, searched, updated, and deleted.

[1158] A "filtering database" refers to a database that collects keywords, URLs, metadata, etc. that contain inappropriate content and provides basic data for evaluating the appropriateness of the content accessed by users.

[1159] "Real-time evaluation" refers to the process of evaluating and analyzing an action or event almost simultaneously when that action or event occurs.

[1160] "Educational content" refers to information, teaching materials, applications, etc. intended for learning and knowledge acquisition.

[1161] "Usage history" refers to a record of the operations performed by the user and the content accessed.

[1162] "Remote monitoring" refers to monitoring the status of devices or systems from a physically distant location using communication technology.

[1163] "Feedback" refers to providing evaluation, guidance, and advice to others regarding usage and behavior.

[1164] A "server" refers to a computer system that provides services and data to other computers (clients) on a network.

[1165] "Terminal" refers to a computer or mobile device that is directly operated by a user.

[1166] First, this system operates with three main components: a server, a terminal, and a user. Specific embodiments of each component will be described in detail below.

[1167] Server-side implementation

[1168] Collecting and updating inappropriate content lists

[1169] The server first collects a list of inappropriate content from the Internet and updates the database. This is done using web crawling technology. Specifically, it uses Python libraries such as Scrapy and BeautifulSoup to collect inappropriate keywords, URLs, and metadata from various sites. This allows the server to always have the latest threat information.

[1170] Generative AI Training

[1171] The server then uses the large amount of data to train a generative artificial intelligence (AI) model, using tools such as Hugging Face's Transformers library, TensorFlow, and PyTorch. The generative AI model learns from the collected data and is able to detect inappropriate content with high accuracy even when new data is used.

[1172] Terminal side embodiment

[1173] Download the filtering database

[1174] The device periodically downloads the filtering database from the server and stores it in local storage. This operation is performed via an API call, and the downloaded database is saved in a local database such as SQLite so that it can be used the next time the device is accessed.

[1175] Real-time content evaluation

[1176] When a user attempts to access a website or app, the device references a locally stored filtering database and uses generative AI to evaluate the appropriateness of the content in real time. For example, using AI models using Scikit-learn and Keras, the device immediately determines whether the content is inappropriate. If it is determined to be inappropriate, the content is immediately blocked and a warning message is displayed to the user.

[1177] Educational content recommendations

[1178] The device uses generative AI to recommend appropriate educational content based on the user's usage history and interests. Recommended content is not only displayed on the device in real time, but also notified to the parent's mobile app. For example, a message may appear saying, "New educational content: Math games recommended."

[1179] User-Side Embodiment

[1180] Remote Parental Monitoring

[1181] Users (parents) can monitor their children's smartphone usage in real time through a dedicated mobile application or web portal. This monitoring includes detailed data such as apps used, websites visited, screen time, etc. The mobile application syncs and displays the data using cloud services such as Firebase and AWS.

[1182] Providing appropriate feedback

[1183] Based on this data, parents can provide appropriate feedback on their children's smartphone usage. With the help of generative AI, parents can send gentle guidance and advice to their children, such as "Try to focus more on your studies," via text message.

[1184] Examples of concrete examples and prompts

[1185] Website Filtering

[1186] User: Your child tries to visit a new website.

[1187] On the device: Website URLs are matched against a local filtering database and generative AI analyzes the content.

[1188] Generative AI: Determines whether a website contains inappropriate content.

[1189] Device: Blocks website access and displays "This site is inappropriate."

[1190] Example prompt: "Please rate whether a new website you visit is inappropriate."

[1191] Educational content recommendations

[1192] Users: Children are interested in online learning.

[1193] Device: Generative AI recommends appropriate educational content based on a child's interests and past usage history.

[1194] Device: Sends a notification to the parent app saying "New educational content: Math learning game recommended."

[1195] Example prompt: "Recommend appropriate educational content based on your child's interests."

[1196] Remote Monitoring

[1197] User (Parent): Check your child's usage on the mobile app.

[1198] Device: View apps used, websites visited, and screen time history.

[1199] User (parent): Sends appropriate advice to the child based on suggestions from the generative AI.

[1200] Example prompt: "Please provide appropriate feedback regarding your child's smartphone usage."

[1201] Through these steps, the system can keep children's digital environments safe and educationally rewarding, while allowing parents to effectively monitor and provide feedback remotely.

[1202] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1203] Step 1:

[1204] Collecting and updating inappropriate content lists

[1205] The server collects inappropriate content lists from across the Internet, including inappropriate keywords, website URLs, and content metadata.

[1206] Input: Multiple sources on the internet

[1207] Data processing: Collect data using web crawling techniques (such as Python's Scrapy or BeautifulSoup).

[1208] Output: Save and update the inappropriate content list in the database.

[1209] Specific behavior:

[1210] 1. The server accesses the list of web pages to be crawled.

[1211] 2. Extract inappropriate keywords, URLs, and metadata from each page.

[1212] 3. This data will be centrally managed and added / updated to the inappropriate content list database.

[1213] Step 2:

[1214] Generative AI Training

[1215] The server trains generative AI models using large amounts of text and other media data.

[1216] Input: Large collected dataset

[1217] Data processing: Training AI models using libraries such as Hugging Face Transformers, TensorFlow, and PyTorch

[1218] Output: A trained generative AI model

[1219] Specific behavior:

[1220] 1. The server supplies the collected data to the AI ​​model.

[1221] 2. AI models use deep learning algorithms to analyze and learn from data.

[1222] 3. The trained model is saved on the server and updated as needed.

[1223] Step 3:

[1224] Download the filtering database

[1225] The terminal periodically downloads the filtering database from the server and stores it in local storage.

[1226] Input: Server filtering database

[1227] Data processing: Download by API call and save to local database (e.g. SQLite)

[1228] Output: Locally stored filtering database

[1229] Specific behavior:

[1230] 1. The device sends a request to the server API.

[1231] 2. The server responds with the latest filtering database.

[1232] 3. The device saves the downloaded database in local storage.

[1233] Step 4:

[1234] Real-time content evaluation

[1235] The device uses generative AI to evaluate the appropriateness of content in real time when a user attempts to access a website or app.

[1236] Input: URL or text data of the content you want to access

[1237] Data calculation: Matching with local filtering database and analysis by AI evaluation model

[1238] Output: Evaluation result: Good or bad

[1239] Specific behavior:

[1240] 1. The device receives the URL and text data of the content the user wants to access.

[1241] 2. Match with a local database and use a generative AI model to analyze the content.

[1242] 3. If the content is deemed inappropriate, access will be blocked and a warning message will be displayed.

[1243] Step 5:

[1244] Educational content recommendations

[1245] The device recommends appropriate educational content based on the user's usage history and interests.

[1246] Input: Usage history, interest data

[1247] Data processing: Content recommendation using usage history analysis and generative AI

[1248] Output: Recommended educational content

[1249] Specific behavior:

[1250] 1. The device analyzes the user's usage history and interest data.

[1251] 2. Generative AI selects the most appropriate educational content based on this data.

[1252] 3. The recommendation results are displayed on the user's device and also notified to the parent's app.

[1253] Step 6:

[1254] Remote Parental Monitoring

[1255] Parents can monitor their children's smartphone usage in real time through a dedicated mobile app.

[1256] Input: Child usage data (apps used, websites visited, screen time)

[1257] Data Processing: Collection and Display of Usage Data

[1258] Output: Display data in the parent mobile application

[1259] Specific behavior:

[1260] 1. Collect usage data from your child's device.

[1261] 2. Synchronize the collected data with a cloud service (Firebase or AWS).

[1262] 3. Display data in real time on parent's mobile app.

[1263] Step 7:

[1264] Providing appropriate feedback

[1265] Parents provide appropriate feedback based on the collected data, and text messages are sent based on generative AI suggestions.

[1266] Input: Collected usage data and generative AI feedback suggestions

[1267] Data processing: generating feedback suggestions

[1268] Output: Feedback as a text message

[1269] Specific behavior:

[1270] 1. Parents can check their children's usage within the mobile app.

[1271] 2. Generative AI automatically generates appropriate feedback and advice.

[1272] 3. The parent will then send the suggested content as a text message to their child.

[1273] (Application example 1)

[1274] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1275] In today's digital society, children have more opportunities to access a variety of media content. However, some media content is inappropriate for children, making it difficult for parents to monitor all content. Furthermore, there is a lack of systems that efficiently provide educational media content to support children's healthy development. There is a need to develop a system that can solve these problems and help children grow up in a safe and beneficial digital environment.

[1276] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1277] In this invention, the server includes a means for periodically collecting the latest inappropriate media content list from the Internet and updating the database, a means for training a generative artificial intelligence model using a large amount of text data and other media data, and a means for evaluating the appropriateness of media content accessed by users in real time. This allows parents to remotely monitor and provide feedback while maintaining a healthy digital environment for their children. The generative artificial intelligence model can also recommend educational media content based on children's interests.

[1278] "Generative AI" is an AI that learns from large amounts of data and generates information in a way that can respond to new data.

[1279] "Inappropriate media content" means media content that contains violent, sexual, discriminatory, or otherwise harmful to children.

[1280] A "media content list" identifies inappropriate media content and manages the list as a database.

[1281] The "means for assessing appropriateness" is a system that has the function of analyzing and assessing whether media content is inappropriate.

[1282] "Educational media content" is media content that is beneficial to children's learning and development and has high educational value.

[1283] "User usage status" refers to information such as what content the user has viewed, how long the user has viewed it, and what interests the user has.

[1284] "Remote monitoring capabilities" are features that allow parents or supervisors to monitor their children's digital environments remotely and in real time.

[1285] The "means of providing feedback" is a function that allows parents to send appropriate guidance and advice to their children with the assistance of generative artificial intelligence.

[1286] A "device for real-time analysis and blocking" is a device that has the capability to analyze media content on the fly and immediately block it if necessary.

[1287] A "generative artificial intelligence model" is a learning model that allows generative artificial intelligence to suggest optimal media content based on data.

[1288] This invention is a system for detecting and blocking inappropriate media content using generative artificial intelligence to maintain a healthy digital environment for children. Specific embodiments of this system are described below.

[1289] 1. Server-side implementation

[1290] The server has the means to perform the following functions:

[1291] Collecting and updating inappropriate media content lists

[1292] Server: Regularly collects the latest inappropriate media content list from the Internet and updates the database. This list includes inappropriate keywords, website URLs, and content metadata, allowing it to respond to the latest threat information.

[1293] Generative AI Training

[1294] Server: Trains generative artificial intelligence models using large amounts of text and other media data. This process uses natural language processing (NLP) techniques and deep learning algorithms. The trained AI continues to learn based on new data, resulting in more accurate detection capabilities.

[1295] 2. Terminal side embodiment

[1296] The terminal has the means to perform the following functions:

[1297] Download the filtering database

[1298] Device: Periodically downloads the filtering database from the server and stores it in local storage. This database provides the foundation data used to evaluate the appropriateness of media content.

[1299] Real-time content evaluation

[1300] On the device: When a user attempts to access media content, the generative AI will refer to a locally stored filtering database and assess the appropriateness of the content in real time. Content deemed inappropriate will be immediately blocked and a warning message will be displayed to the user.

[1301] Educational content recommendations

[1302] On-device: Generative AI will recommend appropriate educational content based on the user's usage history and interests. These recommendations will be customized to the user's learning style and interests and displayed on the smartphone screen.

[1303] 3. User-side implementation

[1304] The user (parent) has the means to perform the following functions:

[1305] Remote Parental Monitoring

[1306] Users (Parents): Through a dedicated mobile application or web portal, they can monitor their children's device usage in real time, including data such as apps used, websites visited, and screen time.

[1307] Providing appropriate feedback

[1308] User (Parent): Parents can use the collected data to provide feedback on their children's device usage, with the assistance of generative AI, which can then send appropriate guidance and advice via text message.

[1309] Specific examples

[1310] Real-time evaluation of media content

[1311] User: A child tries to access a new video.

[1312] On device: Video frames are analyzed in real time, and generative AI detects inappropriate content.

[1313] Generative AI: Determines whether a video contains inappropriate content.

[1314] Device: Immediately stop playing the video and display "This content is inappropriate."

[1315] Educational content recommendations

[1316] Users: Children are interested in online learning.

[1317] Device: Generative AI recommends appropriate educational content based on a child's interests and past usage history.

[1318] Device: Sends a notification to the parent app saying "New educational content: Math game recommended."

[1319] Remote Monitoring

[1320] User (Parent): Check your child's usage on the mobile app.

[1321] Device: View apps used, websites visited, and screen time history.

[1322] User (parent): Based on the generative AI's suggestions, send advice to the child such as "This anime is violent, so please watch other educational content."

[1323] Prompt Sentence Examples

[1324] Below are some example prompts for recommending appropriate educational content to generative AI:

[1325] "Please recommend appropriate educational content based on my child's interests. Past viewing has been dominated by action anime, so please suggest more peaceful educational content."

[1326] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1327] Step 1:

[1328] Collecting and updating inappropriate media content lists

[1329] Server: The server periodically collects data about objectionable media content from multiple sources across the internet, which may involve web scraping techniques. Data collected may include objectionable keywords, website URLs, and metadata.

[1330] Input: Inappropriate content information on the internet.

[1331] Output: An updated list of inappropriate media content.

[1332] What it does: The crawler visits specific sites on the internet, extracts data, stores this data in a database on the server, and updates the inappropriate media content list.

[1333] Step 2:

[1334] Generative AI Training

[1335] Server: The server trains generative artificial intelligence models using large amounts of text and other media data, a process that uses natural language processing (NLP) techniques and deep learning algorithms.

[1336] Input: Large amounts of text and media data.

[1337] Output: A highly accurate generative artificial intelligence model.

[1338] What it does: Train AI with a dataset, validate the results, and update the model. For example, run deep learning algorithms using Python and TensorFlow.

[1339] Step 3:

[1340] Download the filtering database

[1341] Device: The device periodically downloads the filtering database from the server and stores it in local storage.

[1342] Input: Filtering database on the server.

[1343] Output: Filtering database stored in the device's local storage.

[1344] Specific operation: The device sends an API request to retrieve the latest database from the server, and then saves the database in the device's storage, making it available for offline reference.

[1345] Step 4:

[1346] Real-time content evaluation

[1347] Device: When a user attempts to access media content, the device consults a locally stored filtering database, and generative AI assesses the content's appropriateness in real time.

[1348] Input: Media content data and local filtering database.

[1349] Output: Content appropriateness assessment result (appropriate / inappropriate).

[1350] What it does: The device captures the media content being played frame by frame and passes it to a generative AI for analysis. If the content is deemed inappropriate, it blocks it and displays a warning message to the user.

[1351] Step 5:

[1352] Educational content recommendations

[1353] Device: Generative AI recommends appropriate educational content based on the user's usage history and interests.

[1354] Input: User usage history and interest data.

[1355] Output: Recommended educational content.

[1356] Specific operation: The device sends prompts to the generative AI to generate educational content based on the user's interests. For example, a prompt such as, "Please recommend appropriate educational content based on my child's interests. Most of his past viewings have been action anime, so please suggest more peaceful educational content."

[1357] Step 6:

[1358] Remote Parental Monitoring

[1359] Users (Parents): Monitor their children's device usage in real time through a dedicated mobile application or web portal.

[1360] Input: Child's device usage data.

[1361] Output: Monitoring results displayed on the parent device.

[1362] What it does: The app sends usage data from the child's device to a server. The application on the parent's device receives the data and visualizes it in the form of a dashboard.

[1363] Step 7:

[1364] Providing appropriate feedback

[1365] User (Parent): Parents use the collected data to provide feedback on their children's device usage, with the assistance of generative AI to send appropriate guidance and advice via text message.

[1366] Input: Parent feedback and generative AI advice.

[1367] Output: A text message sent to the child's device.

[1368] How it works: Parents use the app to send prompts to the generative AI, which then generates appropriate messages, such as "This cartoon is violent, so please watch other educational content" and send it to their child's device.

[1369] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1370] This invention is a system that uses generative artificial intelligence (AI) and an emotion engine to detect and block inappropriate content and maintain a healthy digital environment for children. Furthermore, this system analyzes the emotional state of the user (child), recommends appropriate educational content based on the user's emotion, and provides feedback support to parents.

[1371] 1. Server-side implementation

[1372] Collecting and updating inappropriate content lists

[1373] Server: The server periodically collects the latest inappropriate content list from the internet and updates the database. This list includes inappropriate keywords, website URLs, and metadata, and is collected using web scraping technology and APIs.

[1374] Generative AI Training

[1375] Server: The server trains the generative AI using large amounts of text and media data, using natural language processing (NLP) techniques and deep learning algorithms (e.g., Transformer-based models). The trained AI continues to learn based on new data, achieving highly accurate detection capabilities.

[1376] 2. Terminal side embodiment

[1377] Download the filtering database

[1378] Device: The device periodically sends a request to the server to download the latest filtering database and save it in local storage.

[1379] Real-time filtering

[1380] On your device: Every time you access a website or app, we use a local filtering database and generative AI to analyze the content, detect and block inappropriate content.

[1381] Implementing the Emotion Engine

[1382] Device: The device uses sensors such as a camera and microphone to capture the user's (child's) facial expressions and voice, and sends the data to the emotion engine, which analyzes the data and determines the user's emotional state in real time.

[1383] Educational content recommendations

[1384] Device: Based on the user's usage history, interests, and emotional state, generative AI selects the most appropriate educational content and notifies the device. For example, if the user is feeling stressed, it will recommend relaxation content, and if the user is concentrating on a learning topic they are interested in, it will recommend learning reinforcement content.

[1385] 3. User-side implementation

[1386] Remote Parental Monitoring

[1387] User (Parent): Parents can monitor their children's smartphone usage and emotional state in real time through a dedicated mobile app or web portal. Data collected includes apps used, websites visited, screen time, and emotional state.

[1388] Providing appropriate feedback

[1389] User (Parent): Parents can provide appropriate feedback with the support of generative AI based on their child's smartphone usage and emotional state. When a parent enters feedback and presses the send button, the feedback is displayed on the child's device in real time. For example, if the child is stressed, they can send an encouraging message, or if the child is concentrating on their studies, they can send a message of encouragement.

[1390] Specific examples

[1391] Website Filtering

[1392] User (child): A child tries to access a new website.

[1393] On the device: Website URLs are matched against a local filtering database and generative AI analyzes the content.

[1394] Generative AI: Determines whether a website contains inappropriate content.

[1395] Device: Blocks website access and displays "This site is inappropriate."

[1396] Educational content recommendations

[1397] User (child): Children are interested in online learning.

[1398] Device: Generative AI recommends appropriate educational content based on a child's interests, past usage history, and emotional state.

[1399] Device: Sends a notification to the parent app saying "New educational content: Math game recommended."

[1400] Real-time emotion monitoring and feedback

[1401] User (child): The emotion engine detects stress levels while children are learning remotely.

[1402] Device: A notification will be sent to the parent's device saying "Your child is stressed."

[1403] User (parent): Create an encouraging message remotely and press the send button, and the message "Good luck!" will appear in real time on the child's device.

[1404] In this way, the system maintains a healthy digital environment for children, while enhancing educational value and providing parents with tools for effective remote monitoring and feedback.

[1405] The processing flow will be explained below.

[1406] Step 1:

[1407] Server: Collects inappropriate content and updates the database. The server collects the latest inappropriate content list from the Internet using scraping or APIs, and periodically updates the filtering database. This database includes inappropriate keywords, URLs, and metadata.

[1408] Step 2:

[1409] Server: Trains the generative AI model. The server trains the generative AI using large amounts of text and media data. Using natural language processing (NLP) techniques and deep learning algorithms (e.g., Transformer-based models), the AI ​​continues to learn based on new data.

[1410] Step 3:

[1411] Device: Downloads the filtering database. The device periodically sends a request to the server to download the latest filtering database and saves it in local storage. This database is used to evaluate the appropriateness of content.

[1412] Step 4:

[1413] On the device: Every time a user attempts to access a website or app, the content is evaluated for appropriateness. The URL is checked against a local filtering database, and generative AI analyzes the content in real time. If it is deemed inappropriate, it is processed in the next step.

[1414] Step 5:

[1415] Device: Blocks inappropriate content. If real-time analysis determines that content is inappropriate, a warning message will be displayed to the user stating "This site is inappropriate." Access will be blocked.

[1416] Step 6:

[1417] Device: Implementation of the emotion engine. The device's camera and microphone are used to capture the user's (child's) facial expressions and voice, and the data is sent to the emotion engine. The emotion engine analyzes the data and determines the user's emotional state in real time.

[1418] Step 7:

[1419] Device: Recommends educational content. Based on the user's usage history, interests, and emotional state, generative AI selects the most appropriate educational content and notifies the device. For example, if the user is feeling stressed, it will recommend relaxation content.

[1420] Step 8:

[1421] Device: Display recommended educational content on the user's (child's) screen. Promote access to educational content via push notifications and dedicated pages.

[1422] Step 9:

[1423] User (Parent): Monitor their child's smartphone usage and emotional state in real time using a dedicated mobile app or web portal. Data collected includes apps used, websites visited, screen time, and emotional state.

[1424] Step 10:

[1425] User (Parent): Provides feedback on the child's smartphone usage and emotional state. With the support of generative AI, the parent enters the feedback and presses the send button, and the feedback is displayed in real time on the child's device. For example, if stress is detected, an encouraging message is sent, and if the child is concentrating on their studies, a supportive message is sent.

[1426] Example 2

[1427] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1428] In today's digital environment, there is a lot of inappropriate content available online, which can have a negative impact on children. It is also difficult to understand in real time what content children are interested in and their emotional state, making it difficult to provide appropriate educational content. Furthermore, there is a lack of means for parents to remotely monitor their children's digital behavior and provide appropriate feedback. Therefore, there is a need for a system that integrates the detection and blocking of inappropriate content, analysis of children's emotions, recommendation of educational content, and remote monitoring and feedback provision functions.

[1429] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for detecting and blocking inappropriate content using generative artificial intelligence, means for updating an inappropriate content list based on collected data, means for evaluating the appropriateness of content a user attempts to access based on the inappropriate content list, means for analyzing a user's emotional state using a camera and a microphone, means for recommending educational content based on the user's emotional state and usage status, means for collecting and remotely monitoring the user's usage status and emotional state, and means for parents to provide feedback. This not only enables children to enjoy learning and entertainment in a safe and healthy digital environment, but also enables parents to effectively monitor their children's digital behavior and provide appropriate feedback.

[1430] "Generative AI" is an AI technology that learns using large amounts of data and generates and analyzes content based on new data.

[1431] "Inappropriate Content" means any information or media that may be harmful to children or other users, such as violent, pornographic, or libelous material.

[1432] "Updating" refers to adding new information to existing data or lists and keeping them up to date.

[1433] A "filtering database" is a database that accumulates information about inappropriate content and uses that data to evaluate the appropriateness of the content.

[1434] "Evaluating the appropriateness of content" means determining whether the content a user is attempting to access is inappropriate.

[1435] "Camera and microphone" refers to a photographing and recording device for capturing the user's facial expressions and voice.

[1436] An "emotional state" refers to a user's mental and emotional state, and may include, for example, stress, joy, excitement, and the like.

[1437] "Analyzing" refers to extracting and understanding specific information from collected data.

[1438] "Educational content" is information or media intended to educate and increase the user's knowledge.

[1439] "Usage" refers to information about how users use devices and applications.

[1440] "Enabling remote monitoring" means being able to check the device usage status and the user's emotional state from a remote location.

[1441] "Providing feedback" means conveying advice or comments based on the user's actions or situation.

[1442] An "inappropriate content list" is a collection of information about inappropriate content that exists on the Internet.

[1443] This invention is a system that uses generative artificial intelligence (AI) and an emotion engine to detect and block inappropriate content and maintain a healthy digital environment for children. Furthermore, this system analyzes the emotional state of the user (child), recommends appropriate educational content based on the user's emotion, and provides feedback support to parents. Specific embodiments are described below.

[1444] Server-side implementation

[1445] Collecting and updating inappropriate content lists

[1446] The server periodically collects the latest inappropriate content list from the Internet and updates the database. This list includes inappropriate keywords, website URLs, and metadata. The server retrieves the data using a web scraping library such as Python's BeautifulSoup or Scrapy and stores it in an SQL database (e.g., MySQL or PostgreSQL). This allows the server to always maintain the latest inappropriate content information.

[1447] Generative AI Training

[1448] The server trains the generative AI using large amounts of text and media data. Deep learning frameworks used include TensorFlow and PyTorch, and Transformer-based models (e.g., BERT and GPT-3) are used. Natural language processing (NLP) techniques and deep learning algorithms are used for training, and the AI ​​continues to learn based on new data.

[1449] Terminal side embodiment

[1450] Download the filtering database

[1451] The device periodically sends a request to the server to download the latest filtering database and save it in local storage. The device periodically communicates with the server using the HTTP protocol to keep the database up to date.

[1452] Real-time filtering

[1453] When a user attempts to access a website or app, the device uses a local filtering database and generative AI to analyze the content, detect inappropriate content, and block it. Specifically, an AI model running on the device evaluates the content in real time and blocks access if it is deemed inappropriate. For example, when a user (child) attempts to access a new website, the device compares the URL with the local filtering database and the generative AI analyzes the content. If it is deemed inappropriate, the device blocks access to the website and displays the message, "This site is inappropriate."

[1454] Implementing the Emotion Engine

[1455] The device uses sensors such as a camera and microphone to capture the user's facial expressions and voice and send them to the emotion engine. Using libraries such as FaceAPI and AudioAnalyzer, this data is analyzed in real time to determine the user's emotional state. For example, if the emotion engine detects a child is stressed while learning remotely, the parent's device will receive a notification saying, "Your child is feeling stressed."

[1456] Educational content recommendations

[1457] Based on the user's usage history, interests, and emotional state, the generative AI selects the most suitable educational content and notifies the device. For example, if a user (child) shows interest in online learning, the generative AI will recommend appropriate educational content based on their interests, past usage history, and emotional state. The selected content is notified to the device and a message appears saying, "New educational content: Math game recommended."

[1458] User-side implementation (parent)

[1459] Remote Monitoring

[1460] Parents can monitor their children's smartphone usage and emotional state in real time through a dedicated mobile app or web portal. Collected data is displayed using Firebase and cloud services. This feature allows parents to understand their children's current situation and take appropriate action if necessary.

[1461] Providing feedback

[1462] Parents can provide appropriate feedback with the support of generative AI based on their child's smartphone usage and emotional state. When a parent enters the feedback and presses the send button, the feedback is displayed on the child's device in real time. For example, if the child is stressed, they can send an encouraging message, or if they are concentrating on their studies, they can send a message of support.

[1463] Example prompt sentence:

[1464] Prompt: "Please recommend some new educational content. My child is interested in math and has been feeling stressed lately. I'd like some content that's relaxing yet educational."

[1465] The system will enhance educational value while maintaining a healthy digital environment for children, allowing parents to effectively monitor and provide feedback remotely.

[1466] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1467] Step 1:

[1468] Server: Collects inappropriate content lists from the Internet and updates the database.

[1469] Input: Web page data obtained from the Internet

[1470] Processing: The server uses Python's BeautifulSoup or Scrapy library to perform web scraping and extract inappropriate keywords, URLs, and metadata from web page data. The extracted data is stored in an SQL database (e.g., MySQL or PostgreSQL).

[1471] Output: Updated inappropriate content list

[1472] Specific Actions: The server runs a scheduled job (e.g., once daily) that crawls designated websites and stores inappropriate content information in a database.

[1473] Step 2:

[1474] Server: Trains the generative AI model.

[1475] Input: Text data, media data

[1476] Processing: Use TensorFlow or PyTorch to train Transformer-based models (e.g., BERT or GPT-3). Use natural language processing (NLP) techniques and deep learning algorithms to learn from large amounts of data.

[1477] Output: A trained generative AI model

[1478] What it does: The server loads the training data and trains the deep learning model. After the training period is over, the optimized model is saved and deployed as an inappropriate content detection algorithm.

[1479] Step 3:

[1480] On your device: Download the filtering database.

[1481] Input: Latest filtering database on the server

[1482] Processing: The device communicates with the server using an HTTP request and downloads the latest filtering database.

[1483] Output: Filtering database stored in local storage

[1484] Specific operation: The device executes a periodic job (for example, once a day) and sends an HTTP GET request to the server. The device receives the latest database from the server and stores it in local storage.

[1485] Step 4:

[1486] Device: Filter content in real time.

[1487] Input: URL and content data of the website or app the user is trying to access

[1488] Processing: The device analyzes the content using a local filtering database and generative AI, blocking access if it is deemed inappropriate.

[1489] Output: Filtered content (allowed or blocked)

[1490] How it works: When a user opens a browser and accesses a new website, the device sends the URL to a local filtering database and generative AI for analysis. If it is deemed inappropriate, the device blocks access and displays the message "This site is inappropriate."

[1491] Step 5:

[1492] Device: Uses camera and microphone to analyze the user's emotional state.

[1493] Input: User's facial expression data (still image or video), voice data

[1494] Processing: The device sends this data to the emotion engine, which analyzes it using FaceAPI and AudioAnalyzer to determine the user's emotional state (stress, joy, excitement, etc.).

[1495] Output: Information about the user's emotional state

[1496] How it works: The device's camera captures the user's face and the microphone records their voice. The emotion engine analyzes this data in real time to determine their emotional state. For example, if the user is feeling stressed, that information is stored on the device as an analysis result.

[1497] Step 6:

[1498] Device: Recommend educational content.

[1499] Input: User usage history, interest data, emotional state data

[1500] Processing: The generative AI selects the most appropriate educational content based on the above input data.

[1501] Output: Recommended educational content

[1502] Specific operation: The device uses generative AI to select the most suitable educational content based on the user's usage history, interests, and emotional state. The selected content is notified to the device, and a message appears saying, "New educational content: Math game recommended."

[1503] Step 7:

[1504] Parents (users): Remotely monitor their children's smartphone usage and emotional state.

[1505] Input: Child usage data, emotional state data

[1506] Processing: Parents can view this data in real time via a dedicated mobile app or web portal.

[1507] Output: Displayed usage and emotional state

[1508] What it does: When parents open the mobile app, they can see their child's current usage and emotional state in real time, allowing them to take necessary actions.

[1509] Step 8:

[1510] Parents (users): Provide feedback such as encouragement and advice.

[1511] Input: Child usage and emotional state data, parent-entered feedback

[1512] What happens: Parents create feedback in the app or web portal, press the submit button, and the feedback is sent in real time to their child's device.

[1513] Output: Feedback message displayed on the child's device

[1514] How it works: Parents can create an encouraging message such as "Good luck!" using the app and press the "Send" button, and the message will instantly appear on their child's device.

[1515] Through each of these steps, the system maintains a healthy digital environment for children, enhances educational value, and allows parents to effectively monitor and provide feedback remotely.

[1516] (Application example 2)

[1517] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1518] In recent years, the use of the Internet and digital devices has become more widespread, increasing the risk that children will be exposed to inappropriate media content. There is also a need for a method to recommend educationally valuable media to children based on their emotional state while maintaining a healthy digital environment. It is also difficult for parents to remotely monitor their children's usage and emotional state in real time and provide appropriate feedback. Therefore, the present invention aims to solve these issues by utilizing generative artificial intelligence and an emotion engine, and to provide a system that provides educational value while maintaining a healthy digital environment for children.

[1519] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for detecting and blocking inappropriate media using generative artificial intelligence, means for updating an inappropriate media list based on collected data, means for evaluating the appropriateness of media a user attempts to access based on the inappropriate media list, means for recommending educational media, means for collecting and remotely monitoring user usage, means for parents to provide feedback, and means for analyzing a user's emotional state and recommending appropriate media according to that emotional state. This reduces the risk of children being exposed to inappropriate media and makes it possible to provide media with educational value according to the user's emotional state. It also makes it easy for parents to remotely monitor their children's usage and emotional state in real time and provide appropriate feedback.

[1520] "Generative AI" is a technology that uses natural language processing and deep learning algorithms to automatically generate content based on new data, and then analyze and evaluate it.

[1521] "Inappropriate media" is digital content, such as text, images, audio, or video, that is deemed harmful or inappropriate for children.

[1522] An "inappropriate media list" is a list containing inappropriate keywords, URLs, and metadata collected from across the Internet.

[1523] "Educational media" refers to digital content such as text, images, audio, and video that supports children's education and learning.

[1524] "Emotional state" is data that indicates the psychological state of the user, detected using sensors such as a camera or microphone.

[1525] "Remote monitoring" is a feature that allows parents to monitor their children's digital device usage in real time from a distance using dedicated devices or applications.

[1526] "Feedback" is the act of parents sending messages of encouragement or warning based on their children's digital device usage and emotional state.

[1527] "Recommendation" is the act of generative artificial intelligence selecting and presenting appropriate media content based on the user's interests and emotional state.

[1528] MODE FOR CARRYING OUT THE INVENTION

[1529] This invention combines generative artificial intelligence (AI) and an emotion engine to detect and block inappropriate media in virtual stores, recommend products based on emotional states, and provide remote monitoring and feedback. The system ensures a safe digital environment for children and recommends appropriate products based on their emotional states.

[1530] System Configuration

[1531] server

[1532] Collecting and updating inappropriate media lists

[1533] Generative AI Training

[1534] Managing the filtering database

[1535] Terminal

[1536] Real-time detection and blocking of inappropriate media

[1537] Analysis of user emotional states using an emotion engine

[1538] Educational and relevant product recommendations

[1539] Remote monitoring data collection

[1540] User

[1541] Parents can monitor their children's usage and emotional state in real time through a dedicated application.

[1542] Parents provide appropriate feedback

[1543] Server-side implementation

[1544] The server performs the following functions:

[1545] Collecting and updating inappropriate media lists

[1546] The server periodically collects the latest inappropriate media list from the Internet and updates the database, which includes inappropriate keywords, website URLs, and metadata, using web scraping technology and APIs.

[1547] Generative AI Training

[1548] The server trains the generative AI using large amounts of text and media data, using natural language processing (NLP) techniques and deep learning algorithms (e.g., Transformer-based models). The trained AI continues to learn based on new data, achieving highly accurate detection capabilities.

[1549] Terminal side embodiment

[1550] The terminal performs the following functions:

[1551] Download the filtering database

[1552] The device periodically sends a request to the server to download the latest filtering database and store it in local storage.

[1553] Real-time filtering

[1554] Every time a user (child) tries to access a product in a virtual store, the content is analyzed using a local filtering database and generative AI to detect and block inappropriate content.

[1555] Implementing the Emotion Engine

[1556] The device uses sensors such as a camera and microphone to capture the user's (child's) facial expressions and voice, and sends the data to the emotion engine, which analyzes the data and determines the user's emotional state in real time.

[1557] Educational content recommendations

[1558] Based on the user's usage history, interests, and emotional state, generative AI selects the most suitable educational content and products and notifies the device. For example, if the user is feeling stressed, it will recommend relaxation products, and if the user is concentrating on a learning content that interests them, it will recommend learning reinforcement content.

[1559] User-Side Embodiment

[1560] Remote Parental Monitoring

[1561] Parents can monitor their children's smartphone usage and emotional state in real time through a dedicated mobile app or web portal. Data collected includes apps used, websites visited, screen time, and emotional state.

[1562] Providing appropriate feedback

[1563] Parents can provide appropriate feedback with the support of generative AI based on their child's smartphone usage and emotional state. When a parent enters the feedback and presses the send button, the feedback is displayed on the child's device in real time. For example, if the child is stressed, they can send an encouraging message, or if they are concentrating on their studies, they can send a message of support.

[1564] Specific examples

[1565] Website Filtering

[1566] A user (child) attempts to access a new website. The inappropriate media detected is processed as follows:

[1567] The URL is checked against a local filtering database

[1568] Generative AI analyzes content, and if it determines that a website is inappropriate media, it blocks it from being displayed and notifies users that "this site is inappropriate."

[1569] Educational content recommendations

[1570] Users (children) become interested in online learning

[1571] Generative AI suggests appropriate educational content and products based on user interests and past usage history

[1572] A notification is sent to the parent app saying, "New educational content: Math games recommended."

[1573] Real-time emotion monitoring and feedback

[1574] The emotion engine detects that the user (child) is feeling stressed during remote learning.

[1575] Parents receive a notification on their device that their child is feeling stressed.

[1576] Parents can remotely create encouraging messages and press the send button, which will then appear on their child's device with the message "Good luck!"

[1577] Prompt Sentence Examples

[1578] Prompt the generative AI model:

[1579] What products do you recommend for users when they are feeling stressed?

[1580] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1581] Step 1:

[1582] The server collects and periodically updates an inappropriate media list from the Internet. This list includes inappropriate keywords, URLs, and metadata. The server uses web scraping technology and public APIs to collect the latest data from the Internet and adds it to the filtering database, thereby keeping the inappropriate media list up to date.

[1583] Input: Data on the Internet

[1584] Output: Updated inappropriate media list

[1585] Step 2:

[1586] The server uses large amounts of text and media data to train the generative AI, using natural language processing (NLP) techniques and deep learning algorithms. For example, a Transformer-based model is used to train the AI ​​to detect inappropriate media with high accuracy. The trained AI model then continues to learn and improve based on new data.

[1587] Input: Large amounts of text and media data

[1588] Output: A trained generative AI model

[1589] Step 3:

[1590] The device periodically sends a request to the server to download the latest filtering database and save it in local storage, so that the device can always use the latest inappropriate media list.

[1591] Input: A request to the server

[1592] Output: Locally stored filtering database

[1593] Step 4:

[1594] When a user (child) tries to access a product in the virtual store, the device analyzes the content using a local filtering database and generative AI. The generative AI evaluates the product's text, images, and multimedia data to determine whether it contains inappropriate elements. If the generative AI determines that the product is inappropriate, it blocks the product and notifies the user, "This product is inappropriate."

[1595] Input: The product data the user is trying to access

[1596] Output: Inappropriate media blocking notification

[1597] Step 5:

[1598] The device uses sensors such as a camera and microphone to capture the user's (child's) facial expressions and voice, and sends the data to the emotion engine, which analyzes the sensor data and determines the user's emotional state in real time.

[1599] Input: Sensor data from camera and microphone

[1600] Output: Parsed user's emotional state

[1601] Step 6:

[1602] The device uses generative AI to recommend appropriate educational content and products based on the user's usage history, interests, and emotional state. For example, if the user is feeling stressed, it will recommend relaxation products, and if the user is concentrating on a learning topic they are interested in, it will recommend learning reinforcement content.

[1603] Input: User usage history, interests, emotional state

[1604] Output: Recommended educational content and products

[1605] Step 7:

[1606] The device collects the user's usage and emotional state and transmits it in real time to a dedicated parental application, through which parents can monitor their child's current usage and emotional state.

[1607] Input: User usage status, emotional state data

[1608] Output: Real-time data displayed in a dedicated parent application

[1609] Step 8:

[1610] The user (parent) creates a feedback message in a dedicated application based on the child's usage and emotional state, and presses the send button. The feedback is then displayed in real time on the child's device. For example, if the child is stressed, an encouraging message can be sent.

[1611] Input: Feedback message entered by the parent

[1612] Output: Messages displayed on the child's device in real time

[1613] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1614] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1615] 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 the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1616] [Fourth embodiment]

[1617] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1618] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1619] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1620] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1621] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1622] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1623] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1624] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1625] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1626] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific 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.

[1627] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1628] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. 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 process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1629] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1630] This invention is a system that uses generative artificial intelligence (AI) to detect and block inappropriate content and maintain a healthy digital environment for children. This system works in cooperation with different entities: a server, a device, and a user.

[1631] 1. Server-side implementation

[1632] Collecting and updating inappropriate content lists

[1633] Server: Regularly collects the latest inappropriate content lists from the Internet and updates the database. This list includes inappropriate keywords, website URLs, and content metadata, allowing it to respond to the latest threat information.

[1634] Generative AI Training

[1635] Server: Trains generative AI models using large amounts of text and other media data. This process uses natural language processing (NLP) techniques and deep learning algorithms. The trained AI continues to learn based on new data, resulting in more accurate detection capabilities.

[1636] 2. Terminal side embodiment

[1637] Download the filtering database

[1638] Device: Periodically downloads the filtering database from the server and stores it in local storage. This database provides the basis for assessing the appropriateness of content.

[1639] Real-time content evaluation

[1640] On the device: When a user attempts to access a website or app, the generative AI consults a locally stored filtering database and evaluates the content's appropriateness in real time. Content deemed inappropriate is immediately blocked and a warning message is displayed to the user.

[1641] Educational content recommendations

[1642] On-device: Generative AI will recommend appropriate educational content offline and online based on the user's usage history and interests. These recommendations will be customized to the user's learning style and interests and displayed on the smartphone screen.

[1643] 3. User-side implementation

[1644] Remote Parental Monitoring

[1645] Users (Parents): Through a dedicated mobile application or web portal, they can monitor their children's smartphone usage in real time, including data such as apps used, websites visited, and screen time.

[1646] Providing appropriate feedback

[1647] User (Parent): Parents can use the collected data to provide feedback on their children's smartphone usage, with the assistance of generative AI, which can then send appropriate guidance and advice via text message.

[1648] Specific examples

[1649] Website Filtering

[1650] User: Your child tries to visit a new website.

[1651] On the device: Website URLs are matched against a local filtering database and generative AI analyzes the content.

[1652] Generative AI: Determines whether a website contains inappropriate content.

[1653] Device: Blocks website access and displays "This site is inappropriate."

[1654] Educational content recommendations

[1655] Users: Children are interested in online learning.

[1656] Device: Generative AI recommends appropriate educational content based on a child's interests and past usage history.

[1657] Device: Sends a notification to the parent app saying "New educational content: Math game recommended."

[1658] Remote Monitoring

[1659] User (Parent): Check your child's usage on the mobile app.

[1660] Device: View apps used, websites visited, and screen time history.

[1661] User (parent): Sends appropriate advice to the child based on suggestions from the generative AI.

[1662] Through these steps, the system not only provides features to help children stay safe in the digital environment, but also gives parents the tools to effectively monitor and provide feedback remotely, making their children's online experience safer and more educational.

[1663] The processing flow will be explained below.

[1664] Step 1:

[1665] Server: Collects inappropriate content and updates the database. The server collects inappropriate keywords, URLs, and metadata from the Internet and updates the filtering database. This collection is done using web scraping technology and APIs.

[1666] Step 2:

[1667] Server: Trains the generative AI model. The server uses large amounts of collected text and media data to train the generative AI. Natural language processing (NLP) techniques and deep learning algorithms (e.g., Transformer-based models) are used for training.

[1668] Step 3:

[1669] Device: Downloads the filtering database. The device periodically sends a request to the server, downloads the latest filtering database, and stores it in local storage.

[1670] Step 4:

[1671] Device: Monitors user website access and compares the URLs they attempt to access against a local filtering database. Generative AI analyzes the content of the webpage in real time and evaluates its appropriateness. If it is deemed inappropriate, it proceeds to the next step.

[1672] Step 5:

[1673] Device: Blocks inappropriate content. Real-time analysis is performed, and if content is deemed inappropriate, a warning message such as "This site is inappropriate" is displayed to the user and access is blocked.

[1674] Step 6:

[1675] Device: Recommends educational content. By analyzing the user's usage history, interests, and past search history, generative AI selects appropriate educational content and notifies the device.

[1676] Step 7:

[1677] Device: Display recommended educational content on the user's (child's) screen. Promote access to educational content through notifications, dedicated pages, etc.

[1678] Step 8:

[1679] User (Parent): Monitor your child's smartphone usage in real time using a dedicated mobile app or web portal, retrieve data from the server, and view activity reports.

[1680] Step 9:

[1681] User (Parent): Provides appropriate feedback on children's smartphone usage. Supported by generative AI, real-time guidance and comments are sent as text messages. When the parent enters the feedback and presses the send button, the message appears on the child's device.

[1682] Example 1

[1683] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1684] The internet is filled with a wealth of harmful and inappropriate content for children, making it difficult for parents to maintain a safe online environment for their children. Providing an environment where children can access appropriate educational content based on their interests is also a challenge. Furthermore, there is a lack of systems that allow parents to remotely monitor their children's online activity and provide appropriate feedback. This invention aims to comprehensively solve these problems and maintain a healthy digital environment for children.

[1685] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1686] In this invention, the server includes a means for collecting a list of inappropriate content from the Internet and updating the database, a means for training a generative artificial intelligence model using large amounts of data, and a means for providing a filtering database to the terminal. This enables the latest inappropriate content to be quickly detected and blocked. Furthermore, the filtering database downloaded from the server is stored in local storage, and the appropriateness of content the user attempts to access is evaluated and blocked in real time, thereby maintaining a safe digital environment for children. Furthermore, the system analyzes the user's usage history and can recommend appropriate educational content based on that analysis, thereby enhancing children's motivation to learn. Parents can remotely monitor their children's smartphone usage through a dedicated application and provide appropriate feedback, providing an educational and safe digital environment.

[1687] "Inappropriate Content" refers to content that is harmful to children and contains violence, pornography, discrimination, or other ethically objectionable information.

[1688] "Generative artificial intelligence" is a form of artificial intelligence that can learn from large amounts of data and make predictions and classifications based on new data, and specifically refers to models that involve data generation.

[1689] A "database" refers to a collection of information that is systematically organized and managed so that it can be stored, searched, updated, and deleted.

[1690] A "filtering database" refers to a database that collects keywords, URLs, metadata, etc. that contain inappropriate content and provides basic data for evaluating the appropriateness of the content accessed by users.

[1691] "Real-time evaluation" refers to the process of evaluating and analyzing an action or event almost simultaneously when that action or event occurs.

[1692] "Educational content" refers to information, teaching materials, applications, etc. intended for learning and knowledge acquisition.

[1693] "Usage history" refers to a record of the operations performed by the user and the content accessed.

[1694] "Remote monitoring" refers to monitoring the status of devices or systems from a physically distant location using communication technology.

[1695] "Feedback" refers to providing evaluation, guidance, and advice to others regarding usage and behavior.

[1696] A "server" refers to a computer system that provides services and data to other computers (clients) on a network.

[1697] "Terminal" refers to a computer or mobile device that is directly operated by a user.

[1698] First, this system operates with three main components: a server, a terminal, and a user. Specific embodiments of each component will be described in detail below.

[1699] Server-side implementation

[1700] Collecting and updating inappropriate content lists

[1701] The server first collects a list of inappropriate content from the Internet and updates the database. This is done using web crawling technology. Specifically, it uses Python libraries such as Scrapy and BeautifulSoup to collect inappropriate keywords, URLs, and metadata from various sites. This allows the server to always have the latest threat information.

[1702] Generative AI Training

[1703] The server then uses the large amount of data to train a generative artificial intelligence (AI) model, using tools such as Hugging Face's Transformers library, TensorFlow, and PyTorch. The generative AI model learns from the collected data and is able to detect inappropriate content with high accuracy even when new data is used.

[1704] Terminal side embodiment

[1705] Download the filtering database

[1706] The device periodically downloads the filtering database from the server and stores it in local storage. This operation is performed via an API call, and the downloaded database is saved in a local database such as SQLite so that it can be used the next time the device is accessed.

[1707] Real-time content evaluation

[1708] When a user attempts to access a website or app, the device references a locally stored filtering database and uses generative AI to evaluate the appropriateness of the content in real time. For example, using AI models using Scikit-learn and Keras, the device immediately determines whether the content is inappropriate. If it is determined to be inappropriate, the content is immediately blocked and a warning message is displayed to the user.

[1709] Educational content recommendations

[1710] The device uses generative AI to recommend appropriate educational content based on the user's usage history and interests. Recommended content is not only displayed on the device in real time, but also notified to the parent's mobile app. For example, a message may appear saying, "New educational content: Math games recommended."

[1711] User-Side Embodiment

[1712] Remote Parental Monitoring

[1713] Users (parents) can monitor their children's smartphone usage in real time through a dedicated mobile application or web portal. This monitoring includes detailed data such as apps used, websites visited, screen time, etc. The mobile application syncs and displays the data using cloud services such as Firebase and AWS.

[1714] Providing appropriate feedback

[1715] Based on this data, parents can provide appropriate feedback on their children's smartphone usage. With the help of generative AI, parents can send gentle guidance and advice to their children, such as "Try to focus more on your studies," via text message.

[1716] Examples of concrete examples and prompts

[1717] Website Filtering

[1718] User: Your child tries to visit a new website.

[1719] On the device: Website URLs are matched against a local filtering database and generative AI analyzes the content.

[1720] Generative AI: Determines whether a website contains inappropriate content.

[1721] Device: Blocks website access and displays "This site is inappropriate."

[1722] Example prompt: "Please rate whether a new website you visit is inappropriate."

[1723] Educational content recommendations

[1724] Users: Children are interested in online learning.

[1725] Device: Generative AI recommends appropriate educational content based on a child's interests and past usage history.

[1726] Device: Sends a notification to the parent app saying "New educational content: Math learning game recommended."

[1727] Example prompt: "Recommend appropriate educational content based on your child's interests."

[1728] Remote Monitoring

[1729] User (Parent): Check your child's usage on the mobile app.

[1730] Device: View apps used, websites visited, and screen time history.

[1731] User (parent): Sends appropriate advice to the child based on suggestions from the generative AI.

[1732] Example prompt: "Please provide appropriate feedback regarding your child's smartphone usage."

[1733] Through these steps, the system can keep children's digital environments safe and educationally rewarding, while allowing parents to effectively monitor and provide feedback remotely.

[1734] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1735] Step 1:

[1736] Collecting and updating inappropriate content lists

[1737] The server collects inappropriate content lists from across the Internet, including inappropriate keywords, website URLs, and content metadata.

[1738] Input: Multiple sources on the internet

[1739] Data processing: Collect data using web crawling techniques (such as Python's Scrapy or BeautifulSoup).

[1740] Output: Save and update the inappropriate content list in the database.

[1741] Specific behavior:

[1742] 1. The server accesses the list of web pages to be crawled.

[1743] 2. Extract inappropriate keywords, URLs, and metadata from each page.

[1744] 3. This data will be centrally managed and added / updated to the inappropriate content list database.

[1745] Step 2:

[1746] Generative AI Training

[1747] The server trains generative AI models using large amounts of text and other media data.

[1748] Input: Large collected dataset

[1749] Data processing: Training AI models using libraries such as Hugging Face Transformers, TensorFlow, and PyTorch

[1750] Output: A trained generative AI model

[1751] Specific behavior:

[1752] 1. The server supplies the collected data to the AI ​​model.

[1753] 2. AI models use deep learning algorithms to analyze and learn from data.

[1754] 3. The trained model is saved on the server and updated as needed.

[1755] Step 3:

[1756] Download the filtering database

[1757] The terminal periodically downloads the filtering database from the server and stores it in local storage.

[1758] Input: Server filtering database

[1759] Data processing: Download by API call and save to local database (e.g. SQLite)

[1760] Output: Locally stored filtering database

[1761] Specific behavior:

[1762] 1. The device sends a request to the server API.

[1763] 2. The server responds with the latest filtering database.

[1764] 3. The device saves the downloaded database in local storage.

[1765] Step 4:

[1766] Real-time content evaluation

[1767] The device uses generative AI to evaluate the appropriateness of content in real time when a user attempts to access a website or app.

[1768] Input: URL or text data of the content you want to access

[1769] Data calculation: Matching with local filtering database and analysis by AI evaluation model

[1770] Output: Evaluation result: Good or bad

[1771] Specific behavior:

[1772] 1. The device receives the URL and text data of the content the user wants to access.

[1773] 2. Match with a local database and use a generative AI model to analyze the content.

[1774] 3. If the content is deemed inappropriate, access will be blocked and a warning message will be displayed.

[1775] Step 5:

[1776] Educational content recommendations

[1777] The device recommends appropriate educational content based on the user's usage history and interests.

[1778] Input: Usage history, interest data

[1779] Data processing: Content recommendation using usage history analysis and generative AI

[1780] Output: Recommended educational content

[1781] Specific behavior:

[1782] 1. The device analyzes the user's usage history and interest data.

[1783] 2. Generative AI selects the most appropriate educational content based on this data.

[1784] 3. The recommendation results are displayed on the user's device and also notified to the parent's app.

[1785] Step 6:

[1786] Remote Parental Monitoring

[1787] Parents can monitor their children's smartphone usage in real time through a dedicated mobile app.

[1788] Input: Child usage data (apps used, websites visited, screen time)

[1789] Data Processing: Collection and Display of Usage Data

[1790] Output: Display data in the parent mobile application

[1791] Specific behavior:

[1792] 1. Collect usage data from your child's device.

[1793] 2. Synchronize the collected data with a cloud service (Firebase or AWS).

[1794] 3. Display data in real time on parent's mobile app.

[1795] Step 7:

[1796] Providing appropriate feedback

[1797] Parents provide appropriate feedback based on the collected data, and text messages are sent based on generative AI suggestions.

[1798] Input: Collected usage data and generative AI feedback suggestions

[1799] Data processing: generating feedback suggestions

[1800] Output: Feedback as a text message

[1801] Specific behavior:

[1802] 1. Parents can check their children's usage within the mobile app.

[1803] 2. Generative AI automatically generates appropriate feedback and advice.

[1804] 3. The parent will then send the suggested content as a text message to their child.

[1805] (Application example 1)

[1806] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1807] In today's digital society, children have more opportunities to access a variety of media content. However, some media content is inappropriate for children, making it difficult for parents to monitor all content. Furthermore, there is a lack of systems that efficiently provide educational media content to support children's healthy development. There is a need to develop a system that can solve these problems and help children grow up in a safe and beneficial digital environment.

[1808] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1809] In this invention, the server includes a means for periodically collecting the latest inappropriate media content list from the Internet and updating the database, a means for training a generative artificial intelligence model using a large amount of text data and other media data, and a means for evaluating the appropriateness of media content accessed by users in real time. This allows parents to remotely monitor and provide feedback while maintaining a healthy digital environment for their children. The generative artificial intelligence model can also recommend educational media content based on children's interests.

[1810] "Generative AI" is an AI that learns from large amounts of data and generates information in a way that can respond to new data.

[1811] "Inappropriate media content" means media content that contains violent, sexual, discriminatory, or otherwise harmful to children.

[1812] A "media content list" identifies inappropriate media content and manages the list as a database.

[1813] The "means for assessing appropriateness" is a system that has the function of analyzing and assessing whether media content is inappropriate.

[1814] "Educational media content" is media content that is beneficial to children's learning and development and has high educational value.

[1815] "User usage status" refers to information such as what content the user has viewed, how long the user has viewed it, and what interests the user has.

[1816] "Remote monitoring capabilities" are features that allow parents or supervisors to monitor their children's digital environments remotely and in real time.

[1817] The "means of providing feedback" is a function that allows parents to send appropriate guidance and advice to their children with the assistance of generative artificial intelligence.

[1818] A "device for real-time analysis and blocking" is a device that has the capability to analyze media content on the fly and immediately block it if necessary.

[1819] A "generative artificial intelligence model" is a learning model that allows generative artificial intelligence to suggest optimal media content based on data.

[1820] This invention is a system for detecting and blocking inappropriate media content using generative artificial intelligence to maintain a healthy digital environment for children. Specific embodiments of this system are described below.

[1821] 1. Server-side implementation

[1822] The server has the means to perform the following functions:

[1823] Collecting and updating inappropriate media content lists

[1824] Server: Regularly collects the latest inappropriate media content list from the Internet and updates the database. This list includes inappropriate keywords, website URLs, and content metadata, allowing it to respond to the latest threat information.

[1825] Generative AI Training

[1826] Server: Trains generative artificial intelligence models using large amounts of text and other media data. This process uses natural language processing (NLP) techniques and deep learning algorithms. The trained AI continues to learn based on new data, resulting in more accurate detection capabilities.

[1827] 2. Terminal side embodiment

[1828] The terminal has the means to perform the following functions:

[1829] Download the filtering database

[1830] Device: Periodically downloads the filtering database from the server and stores it in local storage. This database provides the foundation data used to evaluate the appropriateness of media content.

[1831] Real-time content evaluation

[1832] On the device: When a user attempts to access media content, the generative AI will refer to a locally stored filtering database and assess the appropriateness of the content in real time. Content deemed inappropriate will be immediately blocked and a warning message will be displayed to the user.

[1833] Educational content recommendations

[1834] On-device: Generative AI will recommend appropriate educational content based on the user's usage history and interests. These recommendations will be customized to the user's learning style and interests and displayed on the smartphone screen.

[1835] 3. User-side implementation

[1836] The user (parent) has the means to perform the following functions:

[1837] Remote Parental Monitoring

[1838] Users (Parents): Through a dedicated mobile application or web portal, they can monitor their children's device usage in real time, including data such as apps used, websites visited, and screen time.

[1839] Providing appropriate feedback

[1840] User (Parent): Parents can use the collected data to provide feedback on their children's device usage, with the assistance of generative AI, which can then send appropriate guidance and advice via text message.

[1841] Specific examples

[1842] Real-time evaluation of media content

[1843] User: A child tries to access a new video.

[1844] On device: Video frames are analyzed in real time, and generative AI detects inappropriate content.

[1845] Generative AI: Determines whether a video contains inappropriate content.

[1846] Device: Immediately stop playing the video and display "This content is inappropriate."

[1847] Educational content recommendations

[1848] Users: Children are interested in online learning.

[1849] Device: Generative AI recommends appropriate educational content based on a child's interests and past usage history.

[1850] Device: Sends a notification to the parent app saying "New educational content: Math game recommended."

[1851] Remote Monitoring

[1852] User (Parent): Check your child's usage on the mobile app.

[1853] Device: View apps used, websites visited, and screen time history.

[1854] User (parent): Based on the generative AI's suggestions, send advice to the child such as "This anime is violent, so please watch other educational content."

[1855] Prompt Sentence Examples

[1856] Below are some example prompts for recommending appropriate educational content to generative AI:

[1857] "Please recommend appropriate educational content based on my child's interests. Past viewing has been dominated by action anime, so please suggest more peaceful educational content."

[1858] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1859] Step 1:

[1860] Collecting and updating inappropriate media content lists

[1861] Server: The server periodically collects data about objectionable media content from multiple sources across the internet, which may involve web scraping techniques. Data collected may include objectionable keywords, website URLs, and metadata.

[1862] Input: Inappropriate content information on the internet.

[1863] Output: An updated list of inappropriate media content.

[1864] What it does: The crawler visits specific sites on the internet, extracts data, stores this data in a database on the server, and updates the inappropriate media content list.

[1865] Step 2:

[1866] Generative AI Training

[1867] Server: The server trains generative artificial intelligence models using large amounts of text and other media data, a process that uses natural language processing (NLP) techniques and deep learning algorithms.

[1868] Input: Large amounts of text and media data.

[1869] Output: A highly accurate generative artificial intelligence model.

[1870] What it does: Train AI with a dataset, validate the results, and update the model. For example, run deep learning algorithms using Python and TensorFlow.

[1871] Step 3:

[1872] Download the filtering database

[1873] Device: The device periodically downloads the filtering database from the server and stores it in local storage.

[1874] Input: Filtering database on the server.

[1875] Output: Filtering database stored in the device's local storage.

[1876] Specific operation: The device sends an API request to retrieve the latest database from the server, and then saves the database in the device's storage, making it available for offline reference.

[1877] Step 4:

[1878] Real-time content evaluation

[1879] Device: When a user attempts to access media content, the device consults a locally stored filtering database, and generative AI assesses the content's appropriateness in real time.

[1880] Input: Media content data and local filtering database.

[1881] Output: Content appropriateness assessment result (appropriate / inappropriate).

[1882] What it does: The device captures the media content being played frame by frame and passes it to a generative AI for analysis. If the content is deemed inappropriate, it blocks it and displays a warning message to the user.

[1883] Step 5:

[1884] Educational content recommendations

[1885] Device: Generative AI recommends appropriate educational content based on the user's usage history and interests.

[1886] Input: User usage history and interest data.

[1887] Output: Recommended educational content.

[1888] Specific operation: The device sends prompts to the generative AI to generate educational content based on the user's interests. For example, a prompt such as, "Please recommend appropriate educational content based on my child's interests. Most of his past viewings have been action anime, so please suggest more peaceful educational content."

[1889] Step 6:

[1890] Remote Parental Monitoring

[1891] Users (Parents): Monitor their children's device usage in real time through a dedicated mobile application or web portal.

[1892] Input: Child's device usage data.

[1893] Output: Monitoring results displayed on the parent device.

[1894] What it does: The app sends usage data from the child's device to a server. The application on the parent's device receives the data and visualizes it in the form of a dashboard.

[1895] Step 7:

[1896] Providing appropriate feedback

[1897] User (Parent): Parents use the collected data to provide feedback on their children's device usage, with the assistance of generative AI to send appropriate guidance and advice via text message.

[1898] Input: Parent feedback and generative AI advice.

[1899] Output: A text message sent to the child's device.

[1900] How it works: Parents use the app to send prompts to the generative AI, which then generates appropriate messages, such as "This cartoon is violent, so please watch other educational content" and send it to their child's device.

[1901] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1902] This invention is a system that uses generative artificial intelligence (AI) and an emotion engine to detect and block inappropriate content and maintain a healthy digital environment for children. Furthermore, this system analyzes the emotional state of the user (child), recommends appropriate educational content based on the user's emotion, and provides feedback support to parents.

[1903] 1. Server-side implementation

[1904] Collecting and updating inappropriate content lists

[1905] Server: The server periodically collects the latest inappropriate content list from the internet and updates the database. This list includes inappropriate keywords, website URLs, and metadata, and is collected using web scraping technology and APIs.

[1906] Generative AI Training

[1907] Server: The server trains the generative AI using large amounts of text and media data, using natural language processing (NLP) techniques and deep learning algorithms (e.g., Transformer-based models). The trained AI continues to learn based on new data, achieving highly accurate detection capabilities.

[1908] 2. Terminal side embodiment

[1909] Download the filtering database

[1910] Device: The device periodically sends a request to the server to download the latest filtering database and save it in local storage.

[1911] Real-time filtering

[1912] On your device: Every time you access a website or app, we use a local filtering database and generative AI to analyze the content, detect and block inappropriate content.

[1913] Implementing the Emotion Engine

[1914] Device: The device uses sensors such as a camera and microphone to capture the user's (child's) facial expressions and voice, and sends the data to the emotion engine, which analyzes the data and determines the user's emotional state in real time.

[1915] Educational content recommendations

[1916] Device: Based on the user's usage history, interests, and emotional state, generative AI selects the most appropriate educational content and notifies the device. For example, if the user is feeling stressed, it will recommend relaxation content, and if the user is concentrating on a learning topic they are interested in, it will recommend learning reinforcement content.

[1917] 3. User-side implementation

[1918] Remote Parental Monitoring

[1919] User (Parent): Parents can monitor their children's smartphone usage and emotional state in real time through a dedicated mobile app or web portal. Data collected includes apps used, websites visited, screen time, and emotional state.

[1920] Providing appropriate feedback

[1921] User (Parent): Parents can provide appropriate feedback with the support of generative AI based on their child's smartphone usage and emotional state. When a parent enters feedback and presses the send button, the feedback is displayed on the child's device in real time. For example, if the child is stressed, they can send an encouraging message, or if the child is concentrating on their studies, they can send a message of encouragement.

[1922] Specific examples

[1923] Website Filtering

[1924] User (child): A child tries to access a new website.

[1925] On the device: Website URLs are matched against a local filtering database and generative AI analyzes the content.

[1926] Generative AI: Determines whether a website contains inappropriate content.

[1927] Device: Blocks website access and displays "This site is inappropriate."

[1928] Educational content recommendations

[1929] User (child): Children are interested in online learning.

[1930] Device: Generative AI recommends appropriate educational content based on a child's interests, past usage history, and emotional state.

[1931] Device: Sends a notification to the parent app saying "New educational content: Math game recommended."

[1932] Real-time emotion monitoring and feedback

[1933] User (child): The emotion engine detects stress levels while children are learning remotely.

[1934] Device: A notification will be sent to the parent's device saying "Your child is stressed."

[1935] User (parent): Create an encouraging message remotely and press the send button, and the message "Good luck!" will appear in real time on the child's device.

[1936] In this way, the system maintains a healthy digital environment for children, while enhancing educational value and providing parents with tools for effective remote monitoring and feedback.

[1937] The processing flow will be explained below.

[1938] Step 1:

[1939] Server: Collects inappropriate content and updates the database. The server collects the latest inappropriate content list from the Internet using scraping or APIs, and periodically updates the filtering database. This database includes inappropriate keywords, URLs, and metadata.

[1940] Step 2:

[1941] Server: Trains the generative AI model. The server trains the generative AI using large amounts of text and media data. Using natural language processing (NLP) techniques and deep learning algorithms (e.g., Transformer-based models), the AI ​​continues to learn based on new data.

[1942] Step 3:

[1943] Device: Downloads the filtering database. The device periodically sends a request to the server to download the latest filtering database and saves it in local storage. This database is used to evaluate the appropriateness of content.

[1944] Step 4:

[1945] On the device: Every time a user attempts to access a website or app, the content is evaluated for appropriateness. The URL is checked against a local filtering database, and generative AI analyzes the content in real time. If it is deemed inappropriate, it is processed in the next step.

[1946] Step 5:

[1947] Device: Blocks inappropriate content. If real-time analysis determines that content is inappropriate, a warning message will be displayed to the user stating "This site is inappropriate." Access will be blocked.

[1948] Step 6:

[1949] Device: Implementation of the emotion engine. The device's camera and microphone are used to capture the user's (child's) facial expressions and voice, and the data is sent to the emotion engine. The emotion engine analyzes the data and determines the user's emotional state in real time.

[1950] Step 7:

[1951] Device: Recommends educational content. Based on the user's usage history, interests, and emotional state, generative AI selects the most appropriate educational content and notifies the device. For example, if the user is feeling stressed, it will recommend relaxation content.

[1952] Step 8:

[1953] Device: Display recommended educational content on the user's (child's) screen. Promote access to educational content via push notifications and dedicated pages.

[1954] Step 9:

[1955] User (Parent): Monitor their child's smartphone usage and emotional state in real time using a dedicated mobile app or web portal. Data collected includes apps used, websites visited, screen time, and emotional state.

[1956] Step 10:

[1957] User (Parent): Provides feedback on the child's smartphone usage and emotional state. With the support of generative AI, the parent enters the feedback and presses the send button, and the feedback is displayed in real time on the child's device. For example, if stress is detected, an encouraging message is sent, and if the child is concentrating on their studies, a supportive message is sent.

[1958] Example 2

[1959] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1960] In today's digital environment, there is a lot of inappropriate content available online, which can have a negative impact on children. It is also difficult to understand in real time what content children are interested in and their emotional state, making it difficult to provide appropriate educational content. Furthermore, there is a lack of means for parents to remotely monitor their children's digital behavior and provide appropriate feedback. Therefore, there is a need for a system that integrates the detection and blocking of inappropriate content, analysis of children's emotions, recommendation of educational content, and remote monitoring and feedback provision functions.

[1961] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for detecting and blocking inappropriate content using generative artificial intelligence, means for updating an inappropriate content list based on collected data, means for evaluating the appropriateness of content a user attempts to access based on the inappropriate content list, means for analyzing a user's emotional state using a camera and a microphone, means for recommending educational content based on the user's emotional state and usage status, means for collecting and remotely monitoring the user's usage status and emotional state, and means for parents to provide feedback. This not only enables children to enjoy learning and entertainment in a safe and healthy digital environment, but also enables parents to effectively monitor their children's digital behavior and provide appropriate feedback.

[1962] "Generative AI" is an AI technology that learns using large amounts of data and generates and analyzes content based on new data.

[1963] "Inappropriate Content" means any information or media that may be harmful to children or other users, such as violent, pornographic, or libelous material.

[1964] "Updating" refers to adding new information to existing data or lists and keeping them up to date.

[1965] A "filtering database" is a database that accumulates information about inappropriate content and uses that data to evaluate the appropriateness of the content.

[1966] "Evaluating the appropriateness of content" means determining whether the content a user is attempting to access is inappropriate.

[1967] "Camera and microphone" refers to a photographing and recording device for capturing the user's facial expressions and voice.

[1968] An "emotional state" refers to a user's mental and emotional state, and may include, for example, stress, joy, excitement, and the like.

[1969] "Analyzing" refers to extracting and understanding specific information from collected data.

[1970] "Educational content" is information or media intended to educate and increase the user's knowledge.

[1971] "Usage" refers to information about how users use devices and applications.

[1972] "Enabling remote monitoring" means being able to check the device usage status and the user's emotional state from a remote location.

[1973] "Providing feedback" means conveying advice or comments based on the user's actions or situation.

[1974] An "inappropriate content list" is a collection of information about inappropriate content that exists on the Internet.

[1975] This invention is a system that uses generative artificial intelligence (AI) and an emotion engine to detect and block inappropriate content and maintain a healthy digital environment for children. Furthermore, this system analyzes the emotional state of the user (child), recommends appropriate educational content based on the user's emotion, and provides feedback support to parents. Specific embodiments are described below.

[1976] Server-side implementation

[1977] Collecting and updating inappropriate content lists

[1978] The server periodically collects the latest inappropriate content list from the Internet and updates the database. This list includes inappropriate keywords, website URLs, and metadata. The server retrieves the data using a web scraping library such as Python's BeautifulSoup or Scrapy and stores it in an SQL database (e.g., MySQL or PostgreSQL). This allows the server to always maintain the latest inappropriate content information.

[1979] Generative AI Training

[1980] The server trains the generative AI using large amounts of text and media data. Deep learning frameworks used include TensorFlow and PyTorch, and Transformer-based models (e.g., BERT and GPT-3) are used. Natural language processing (NLP) techniques and deep learning algorithms are used for training, and the AI ​​continues to learn based on new data.

[1981] Terminal side embodiment

[1982] Download the filtering database

[1983] The device periodically sends a request to the server to download the latest filtering database and save it in local storage. The device periodically communicates with the server using the HTTP protocol to keep the database up to date.

[1984] Real-time filtering

[1985] When a user attempts to access a website or app, the device uses a local filtering database and generative AI to analyze the content, detect inappropriate content, and block it. Specifically, an AI model running on the device evaluates the content in real time and blocks access if it is deemed inappropriate. For example, when a user (child) attempts to access a new website, the device compares the URL with the local filtering database and the generative AI analyzes the content. If it is deemed inappropriate, the device blocks access to the website and displays the message, "This site is inappropriate."

[1986] Implementing the Emotion Engine

[1987] The device uses sensors such as a camera and microphone to capture the user's facial expressions and voice and send them to the emotion engine. Using libraries such as FaceAPI and AudioAnalyzer, this data is analyzed in real time to determine the user's emotional state. For example, if the emotion engine detects a child is stressed while learning remotely, the parent's device will receive a notification saying, "Your child is feeling stressed."

[1988] Educational content recommendations

[1989] Based on the user's usage history, interests, and emotional state, the generative AI selects the most suitable educational content and notifies the device. For example, if a user (child) shows interest in online learning, the generative AI will recommend appropriate educational content based on their interests, past usage history, and emotional state. The selected content is notified to the device and a message appears saying, "New educational content: Math game recommended."

[1990] User-side implementation (parent)

[1991] Remote Monitoring

[1992] Parents can monitor their children's smartphone usage and emotional state in real time through a dedicated mobile app or web portal. Collected data is displayed using Firebase and cloud services. This feature allows parents to understand their children's current situation and take appropriate action if necessary.

[1993] Providing feedback

[1994] Parents can provide appropriate feedback with the support of generative AI based on their child's smartphone usage and emotional state. When a parent enters the feedback and presses the send button, the feedback is displayed on the child's device in real time. For example, if the child is stressed, they can send an encouraging message, or if they are concentrating on their studies, they can send a message of support.

[1995] Example prompt sentence:

[1996] Prompt: "Please recommend some new educational content. My child is interested in math and has been feeling stressed lately. I'd like some content that's relaxing yet educational."

[1997] The system will enhance educational value while maintaining a healthy digital environment for children, allowing parents to effectively monitor and provide feedback remotely.

[1998] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1999] Step 1:

[2000] Server: Collects inappropriate content lists from the Internet and updates the database.

[2001] Input: Web page data obtained from the Internet

[2002] Processing: The server uses Python's BeautifulSoup or Scrapy library to perform web scraping and extract inappropriate keywords, URLs, and metadata from web page data. The extracted data is stored in an SQL database (e.g., MySQL or PostgreSQL).

[2003] Output: Updated inappropriate content list

[2004] Specific Actions: The server runs a scheduled job (e.g., once daily) that crawls designated websites and stores inappropriate content information in a database.

[2005] Step 2:

[2006] Server: Trains the generative AI model.

[2007] Input: Text data, media data

[2008] Processing: Use TensorFlow or PyTorch to train Transformer-based models (e.g., BERT or GPT-3). Use natural language processing (NLP) techniques and deep learning algorithms to learn from large amounts of data.

[2009] Output: A trained generative AI model

[2010] What it does: The server loads the training data and trains the deep learning model. After the training period is over, the optimized model is saved and deployed as an inappropriate content detection algorithm.

[2011] Step 3:

[2012] On your device: Download the filtering database.

[2013] Input: Latest filtering database on the server

[2014] Processing: The device communicates with the server using an HTTP request and downloads the latest filtering database.

[2015] Output: Filtering database stored in local storage

[2016] Specific operation: The device executes a periodic job (for example, once a day) and sends an HTTP GET request to the server. The device receives the latest database from the server and stores it in local storage.

[2017] Step 4:

[2018] Device: Filter content in real time.

[2019] Input: URL and content data of the website or app the user is trying to access

[2020] Processing: The device analyzes the content using a local filtering database and generative AI, blocking access if it is deemed inappropriate.

[2021] Output: Filtered content (allowed or blocked)

[2022] How it works: When a user opens a browser and accesses a new website, the device sends the URL to a local filtering database and generative AI for analysis. If it is deemed inappropriate, the device blocks access and displays the message "This site is inappropriate."

[2023] Step 5:

[2024] Device: Uses camera and microphone to analyze the user's emotional state.

[2025] Input: User's facial expression data (still image or video), voice data

[2026] Processing: The device sends this data to the emotion engine, which analyzes it using FaceAPI and AudioAnalyzer to determine the user's emotional state (stress, joy, excitement, etc.).

[2027] Output: Information about the user's emotional state

[2028] How it works: The device's camera captures the user's face and the microphone records their voice. The emotion engine analyzes this data in real time to determine their emotional state. For example, if the user is feeling stressed, that information is stored on the device as an analysis result.

[2029] Step 6:

[2030] Device: Recommend educational content.

[2031] Input: User usage history, interest data, emotional state data

[2032] Processing: The generative AI selects the most appropriate educational content based on the above input data.

[2033] Output: Recommended educational content

[2034] Specific operation: The device uses generative AI to select the most suitable educational content based on the user's usage history, interests, and emotional state. The selected content is notified to the device, and a message appears saying, "New educational content: Math game recommended."

[2035] Step 7:

[2036] Parents (users): Remotely monitor their children's smartphone usage and emotional state.

[2037] Input: Child usage data, emotional state data

[2038] Processing: Parents can view this data in real time via a dedicated mobile app or web portal.

[2039] Output: Displayed usage and emotional state

[2040] What it does: When parents open the mobile app, they can see their child's current usage and emotional state in real time, allowing them to take necessary actions.

[2041] Step 8:

[2042] Parents (users): Provide feedback such as encouragement and advice.

[2043] Input: Child usage and emotional state data, parent-entered feedback

[2044] What happens: Parents create feedback in the app or web portal, press the submit button, and the feedback is sent in real time to their child's device.

[2045] Output: Feedback message displayed on the child's device

[2046] How it works: Parents can create an encouraging message such as "Good luck!" using the app and press the "Send" button, and the message will instantly appear on their child's device.

[2047] Through each of these steps, the system maintains a healthy digital environment for children, enhances educational value, and allows parents to effectively monitor and provide feedback remotely.

[2048] (Application example 2)

[2049] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2050] In recent years, the use of the Internet and digital devices has become more widespread, increasing the risk that children will be exposed to inappropriate media content. There is also a need for a method to recommend educationally valuable media to children based on their emotional state while maintaining a healthy digital environment. It is also difficult for parents to remotely monitor their children's usage and emotional state in real time and provide appropriate feedback. Therefore, the present invention aims to solve these issues by utilizing generative artificial intelligence and an emotion engine, and to provide a system that provides educational value while maintaining a healthy digital environment for children.

[2051] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for detecting and blocking inappropriate media using generative artificial intelligence, means for updating an inappropriate media list based on collected data, means for evaluating the appropriateness of media a user attempts to access based on the inappropriate media list, means for recommending educational media, means for collecting and remotely monitoring user usage, means for parents to provide feedback, and means for analyzing a user's emotional state and recommending appropriate media according to that emotional state. This reduces the risk of children being exposed to inappropriate media and makes it possible to provide media with educational value according to the user's emotional state. It also makes it easy for parents to remotely monitor their children's usage and emotional state in real time and provide appropriate feedback.

[2052] "Generative AI" is a technology that uses natural language processing and deep learning algorithms to automatically generate content based on new data, and then analyze and evaluate it.

[2053] "Inappropriate media" is digital content, such as text, images, audio, or video, that is deemed harmful or inappropriate for children.

[2054] An "inappropriate media list" is a list containing inappropriate keywords, URLs, and metadata collected from across the Internet.

[2055] "Educational media" refers to digital content such as text, images, audio, and video that supports children's education and learning.

[2056] "Emotional state" is data that indicates the psychological state of the user, detected using sensors such as a camera or microphone.

[2057] "Remote monitoring" is a feature that allows parents to monitor their children's digital device usage in real time from a distance using dedicated devices or applications.

[2058] "Feedback" is the act of parents sending messages of encouragement or warning based on their children's digital device usage and emotional state.

[2059] "Recommendation" is the act of generative artificial intelligence selecting and presenting appropriate media content based on the user's interests and emotional state.

[2060] MODE FOR CARRYING OUT THE INVENTION

[2061] This invention combines generative artificial intelligence (AI) and an emotion engine to detect and block inappropriate media in virtual stores, recommend products based on emotional states, and provide remote monitoring and feedback. The system ensures a safe digital environment for children and recommends appropriate products based on their emotional states.

[2062] System Configuration

[2063] server

[2064] Collecting and updating inappropriate media lists

[2065] Generative AI Training

[2066] Managing the filtering database

[2067] Terminal

[2068] Real-time detection and blocking of inappropriate media

[2069] Analysis of user emotional states using an emotion engine

[2070] Educational and relevant product recommendations

[2071] Remote monitoring data collection

[2072] User

[2073] Parents can monitor their children's usage and emotional state in real time through a dedicated application.

[2074] Parents provide appropriate feedback

[2075] Server-side implementation

[2076] The server performs the following functions:

[2077] Collecting and updating inappropriate media lists

[2078] The server periodically collects the latest inappropriate media list from the Internet and updates the database, which includes inappropriate keywords, website URLs, and metadata, using web scraping technology and APIs.

[2079] Generative AI Training

[2080] The server trains the generative AI using large amounts of text and media data, using natural language processing (NLP) techniques and deep learning algorithms (e.g., Transformer-based models). The trained AI continues to learn based on new data, achieving highly accurate detection capabilities.

[2081] Terminal side embodiment

[2082] The terminal performs the following functions:

[2083] Download the filtering database

[2084] The device periodically sends a request to the server to download the latest filtering database and store it in local storage.

[2085] Real-time filtering

[2086] Every time a user (child) tries to access a product in a virtual store, the content is analyzed using a local filtering database and generative AI to detect and block inappropriate content.

[2087] Implementing the Emotion Engine

[2088] The device uses sensors such as a camera and microphone to capture the user's (child's) facial expressions and voice, and sends the data to the emotion engine, which analyzes the data and determines the user's emotional state in real time.

[2089] Educational content recommendations

[2090] Based on the user's usage history, interests, and emotional state, generative AI selects the most suitable educational content and products and notifies the device. For example, if the user is feeling stressed, it will recommend relaxation products, and if the user is concentrating on a learning content that interests them, it will recommend learning reinforcement content.

[2091] User-Side Embodiment

[2092] Remote Parental Monitoring

[2093] Parents can monitor their children's smartphone usage and emotional state in real time through a dedicated mobile app or web portal. Data collected includes apps used, websites visited, screen time, and emotional state.

[2094] Providing appropriate feedback

[2095] Parents can provide appropriate feedback with the support of generative AI based on their child's smartphone usage and emotional state. When a parent enters the feedback and presses the send button, the feedback is displayed on the child's device in real time. For example, if the child is stressed, they can send an encouraging message, or if they are concentrating on their studies, they can send a message of support.

[2096] Specific examples

[2097] Website Filtering

[2098] A user (child) attempts to access a new website. The inappropriate media detected is processed as follows:

[2099] The URL is checked against a local filtering database

[2100] Generative AI analyzes content, and if it determines that a website is inappropriate media, it blocks it from being displayed and notifies users that "this site is inappropriate."

[2101] Educational content recommendations

[2102] Users (children) become interested in online learning

[2103] Generative AI suggests appropriate educational content and products based on user interests and past usage history

[2104] A notification is sent to the parent app saying, "New educational content: Math games recommended."

[2105] Real-time emotion monitoring and feedback

[2106] The emotion engine detects that the user (child) is feeling stressed during remote learning.

[2107] Parents receive a notification on their device that their child is feeling stressed.

[2108] Parents can remotely create encouraging messages and press the send button, which will then appear on their child's device with the message "Good luck!"

[2109] Prompt Sentence Examples

[2110] Prompt the generative AI model:

[2111] What products do you recommend for users when they are feeling stressed?

[2112] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2113] Step 1:

[2114] The server collects and periodically updates an inappropriate media list from the Internet. This list includes inappropriate keywords, URLs, and metadata. The server uses web scraping technology and public APIs to collect the latest data from the Internet and adds it to the filtering database, thereby keeping the inappropriate media list up to date.

[2115] Input: Data on the Internet

[2116] Output: Updated inappropriate media list

[2117] Step 2:

[2118] The server uses large amounts of text and media data to train the generative AI, using natural language processing (NLP) techniques and deep learning algorithms. For example, a Transformer-based model is used to train the AI ​​to detect inappropriate media with high accuracy. The trained AI model then continues to learn and improve based on new data.

[2119] Input: Large amounts of text and media data

[2120] Output: A trained generative AI model

[2121] Step 3:

[2122] The device periodically sends a request to the server to download the latest filtering database and save it in local storage, so that the device can always use the latest inappropriate media list.

[2123] Input: A request to the server

[2124] Output: Locally stored filtering database

[2125] Step 4:

[2126] When a user (child) tries to access a product in the virtual store, the device analyzes the content using a local filtering database and generative AI. The generative AI evaluates the product's text, images, and multimedia data to determine whether it contains inappropriate elements. If the generative AI determines that the product is inappropriate, it blocks the product and notifies the user, "This product is inappropriate."

[2127] Input: The product data the user is trying to access

[2128] Output: Inappropriate media blocking notification

[2129] Step 5:

[2130] The device uses sensors such as a camera and microphone to capture the user's (child's) facial expressions and voice, and sends the data to the emotion engine, which analyzes the sensor data and determines the user's emotional state in real time.

[2131] Input: Sensor data from camera and microphone

[2132] Output: Parsed user's emotional state

[2133] Step 6:

[2134] The device uses generative AI to recommend appropriate educational content and products based on the user's usage history, interests, and emotional state. For example, if the user is feeling stressed, it will recommend relaxation products, and if the user is concentrating on a learning topic they are interested in, it will recommend learning reinforcement content.

[2135] Input: User usage history, interests, emotional state

[2136] Output: Recommended educational content and products

[2137] Step 7:

[2138] The device collects the user's usage and emotional state and transmits it in real time to a dedicated parental application, through which parents can monitor their child's current usage and emotional state.

[2139] Input: User usage status, emotional state data

[2140] Output: Real-time data displayed in a dedicated parent application

[2141] Step 8:

[2142] The user (parent) creates a feedback message in a dedicated application based on the child's usage and emotional state, and presses the send button. The feedback is then displayed in real time on the child's device. For example, if the child is stressed, an encouraging message can be sent.

[2143] Input: Feedback message entered by the parent

[2144] Output: Messages displayed on the child's device in real time

[2145] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[2146] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[2147] 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 the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[2148] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[2149] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[2150] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[2151] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[2152] Human emotions are based on various balances, such as posture and blood sugar levels. Wh...

Claims

1. A means of detecting and blocking inappropriate content using generative artificial intelligence; and a means for updating the inappropriate content list based on the collected data; and means for evaluating the appropriateness of content that a user attempts to access based on the inappropriate content list; a means of recommending educational content; A means for collecting and remotely monitoring user usage; a means for parents to provide feedback; A system including:

2. 10. The system of claim 1, having the ability to detect and block inappropriate content using generative artificial intelligence.

3. 2. The system according to claim 1, further comprising a function for collecting user usage information and enabling remote monitoring.

Images