System

Abstract

A system is provided.SOLUTION: A system comprising: means for analyzing article information in an electronic device using a generative artificial intelligence model to determine whether it is likely to contain spoiler information; means for hiding spoiler information from display to a user based on the determination; and means for enabling customization of the extent of spoiler information by filtering certain categories and content based on user settings.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] Describe the "problem that the invention aims to solve" and the "means for solving the problem."

[0005] In recent years, article information published on online news sites and social networking sites often contains spoilers for movies, television programs, anime, sporting events, and other content that users are looking forward to. This spoiler information ruins users' enjoyment and degrades their browsing experience. However, manually reviewing the content of each news article and removing spoilers requires a huge amount of effort and is inefficient. Furthermore, it requires considering each user's different preferences for avoiding spoilers, making it difficult to customize the system accordingly. The present invention aims to solve these problems and provide a system that allows users to safely browse the content they want to view without losing any of the content. [Means for solving the problem]

[0006] The present invention provides a system for analyzing the content of news articles in advance using a generative artificial intelligence (AI) model to determine whether they may contain spoilers. The system specifically includes the following means:

[0007] 1. A means for analyzing article information on an electronic device using a generative artificial intelligence model to determine whether it may contain spoilers.

[0008] 2. A means for hiding spoiler information so that it is not displayed to the user based on the result of the determination.

[0009] 3. A means to customize the scope of spoiler information by filtering specific categories or content based on user preferences.

[0010] This allows users to set filters tailored to their interests and avoid articles that may contain spoilers. Additionally, by adding a learning mechanism to improve the analytical accuracy of the generative AI model based on user feedback, the system's accuracy can be improved over time. Furthermore, the system also includes a mechanism to prevent spoilers by replacing parts of articles that may contain spoilers with placeholders. This configuration provides a safe browsing environment without compromising the content users want to see.

[0011] A "generative artificial intelligence model" is a type of computer program that uses machine learning algorithms to analyze text and data and generate specific patterns and predictions.

[0012] "Electronic device" refers to any device used to process and display digital information, such as a computer, smartphone, or tablet.

[0013] "Article information" refers to all information that makes up a news article, such as its title, content, images, and metadata.

[0014] "Spoiler information" refers to important information about the story or ending of a movie, TV program, anime, game, etc., which may spoil the viewer's enjoyment if known in advance.

[0015] "User" means any person or entity that uses the System to view news articles.

[0016] "Filtering" refers to the process of sorting data based on specific criteria to remove irrelevant or unnecessary information.

[0017] "Placeholder" means alternative text or symbols used to hide certain content, such as spoiler information.

[0018] "Customizable" means that a user can change settings or features to suit their preferences and needs.

[0019] "Determining" refers to the process of using a generative artificial intelligence model to analyze article information and determine whether it meets certain conditions.

[0020] "Feedback" refers to information collected from users, such as their ratings and opinions, that can be used to improve systems and models.

[0021] "Learning" refers to the process by which machine learning algorithms improve their analytical accuracy and predictive capabilities based on new data. [Brief explanation of the drawings]

[0022] [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 illustrating 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

[0023] 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.

[0024] First, the terms used in the following description will be explained.

[0025] 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).

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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."

[0030] [First embodiment]

[0031] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0032] 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.

[0033] 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).

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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.

[0038] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0039] 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.

[0040] 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.

[0041] 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.

[0042] 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."

[0043] The present invention relates to a system that analyzes article information, determines whether it contains spoiler information, and hides the spoiler information as necessary, and a specific embodiment thereof will be described below.

[0044] System configuration

[0045] This system consists of a user's device, a server, and a generative artificial intelligence (AI) model. The user uses the device to view news articles and configure spoiler prevention settings. The server receives this configuration information and retrieves and analyzes news articles. The generative AI model is hosted on the server and is used to analyze article information and determine spoiler information.

[0046] Program processing

[0047] 1. User configures spoiler protection:

[0048] The user selects the specific content for which they want to avoid spoilers (for example, "TV Drama A" or "Anime B") from the settings screen on their device, and sends the setting information from the device to the server. The server stores this setting information in association with the user ID.

[0049] 2. The server loads the generative AI model:

[0050] When the server starts up, it loads the generative AI model into memory and prepares it for analyzing article information. This model is trained by a machine learning algorithm and is used to accurately detect spoilers.

[0051] 3. Get news articles and check for spoilers:

[0052] When a user requests a specific news article to be displayed, the server retrieves the headline and content of the relevant news article from the database. The server then inputs the retrieved headline and content into the generative AI model and analyzes whether they contain spoilers. If they are determined to contain spoilers, the server replaces the relevant part with the placeholder "[Spoiler Not Displayed]".

[0053] 4. Customized Display:

[0054] The server reconstructs the article information based on the analysis results and user settings in a way that does not include spoilers, and sends this reconstructed content to the device, which then displays the article to the user in a secure format.

[0055] Specific examples

[0056] As a concrete example, consider the case where a user wants to avoid spoilers for "TV Drama A." When the user makes this setting on their device and registers "TV Drama A" as a spoiler prevention target, the server saves this information. Next, when the user attempts to view news article ID "1234," if the headline of this news article is "Review of the latest episode of 'TV Drama A'," the server uses an AI model to analyze the article content. If the AI ​​model determines that the article content contains spoilers, the server replaces the relevant part with "[Spoiler not displayed]" and sends it to the device. The device then displays the final content to the user.

[0057] Feedback and Learning

[0058] The system also has a function to improve the analysis accuracy of the generative AI model based on user feedback. When users provide feedback on the content of displayed articles, the server collects this information and uses it as retraining data for the AI ​​model. This allows the system's spoiler detection accuracy to improve over time, enabling more accurate filtering.

[0059] Thus, according to the embodiment of the present invention, the user can read news articles with peace of mind and enjoy the content without spoiling the fun.

[0060] The processing flow will be explained below.

[0061] Step 1:

[0062] The user sets up spoiler prevention on the device. The user specifies specific content (e.g., "TV Drama A," "Anime B," etc.) for which they want to avoid spoilers on the device's settings screen.

[0063] Step 2:

[0064] The device sends the setting information to the server, along with the user ID and the specified spoiler prevention target information.

[0065] Step 3:

[0066] The server updates the user settings. Based on the received information, the server saves the spoiler prevention settings associated with the user ID in the database.

[0067] Step 4:

[0068] The server loads the generative AI model. During initialization, the server loads the trained generative AI model into memory, preparing it for analyzing article information.

[0069] Step 5:

[0070] The user requests the display of a news article. The user sends a display request from the terminal to the server specifying a specific news article ID.

[0071] Step 6:

[0072] The device sends a request to the server, which sends the news article ID and user ID information to the server.

[0073] Step 7:

[0074] The server retrieves the news article. The server retrieves the headline and content of the requested article from the database.

[0075] Step 8:

[0076] The server analyzes the article information using the generative AI model. The server inputs the acquired headline and content into the generative AI model and analyzes whether it contains spoilers.

[0077] Step 9:

[0078] The server determines whether the information is a spoiler, and if the generative AI model determines that the information is a spoiler, it identifies the spoiler.

[0079] Step 10:

[0080] The server hides the spoiler information, replaces the spoiler part with the placeholder "[Spoiler hidden]", and modifies the original content.

[0081] Step 11:

[0082] The server sends the revised article information to the terminal. The server sends the revised headline and content to the terminal.

[0083] Step 12:

[0084] The terminal displays the corrected article information to the user. The terminal displays the received headline and the corrected content to the user.

[0085] Step 13:

[0086] When a user views an article and provides feedback, the user transmits their rating and opinion to the server via the terminal.

[0087] Step 14:

[0088] The server collects the feedback and stores it in a database.

[0089] Step 15:

[0090] The server retrains the generative AI model. The server uses the collected feedback data to retrain the generative AI model to improve its accuracy.

[0091] By following the steps above, users can read news articles with peace of mind and enjoy the content without spoiling it.

[0092] Example 1

[0093] 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."

[0094] In recent years, with the increasing number of news articles and reviews distributed over the Internet, users are at a greater risk of coming across spoilers for content they have not yet viewed. In particular, articles about entertainment content such as TV shows, movies, and anime can unintentionally contain spoilers, which can ruin the user's viewing experience. Conventional systems require users to take measures to avoid spoilers themselves, which is cumbersome and unreliable.

[0095] 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.

[0096] In this invention, the server includes a means for a user to use an electronic device to set spoiler prevention settings and transmit the setting information to the server, a means for the server to analyze article information using a generative artificial intelligence model and determine whether it is likely to contain spoiler information, and a means for hiding the spoiler information from the user based on the determination result. This reduces the risk that users will unintentionally come across spoiler information about content they have not viewed, allowing them to read news articles with peace of mind.

[0097] A "user" is a person who uses the system to set up spoiler protection for news articles and view the content.

[0098] "Electronic device" refers to a device used by a user to access and configure the system, including smartphones, PCs, tablets, etc.

[0099] "Server" refers to the central control unit that receives user preferences and retrieves, analyzes, and reconstructs news articles.

[0100] A "generative artificial intelligence model" is a model trained by a machine learning algorithm to analyze text in news articles to detect spoilers.

[0101] "Spoiler information" refers to information that includes important information or developments about content that the user has not yet viewed, and that may spoil the enjoyment of that content.

[0102] "Settings Information" refers to information about specific content or categories related to spoiler prevention that a user enters through the system.

[0103] "Feedback" refers to user-provided ratings and comments on the displayed article content, which helps improve the system's analysis accuracy.

[0104] A "placeholder" is an alternative expression such as "[Spoiler not shown]" that is used to replace a portion of text that contains spoiler information.

[0105] "Filtering" refers to the process of excluding certain types or content based on user settings.

[0106] A "prompt sentence" is an input sentence given to a generative artificial intelligence model that instructs the model on a specific analysis task.

[0107] The present invention is a system that analyzes news articles, determines whether they contain spoilers, and hides the spoilers as necessary. A specific embodiment of the system is described below. The system consists of a user's electronic device, a server, and a generative artificial intelligence model (generative AI model).

[0108] System configuration

[0109] The system includes the following hardware and software:

[0110] 1. User's electronic devices:

[0111] Users access the system using electronic devices such as smartphones, PCs, and tablets.

[0112] Set spoiler protection and read the news article.

[0113] 2. Server:

[0114] The server receives and stores user-submitted configuration information, and also retrieves, analyzes, and reconstructs news articles.

[0115] The server analyzes article information using a generative artificial intelligence model (e.g., BERT or GPT-3).

[0116] 3. Generative AI Models:

[0117] The generative artificial intelligence model analyzes the headlines and content of news articles to determine whether they contain spoilers.

[0118] Specific machine learning libraries used include TensorFlow and PyTorch.

[0119] Program processing

[0120] The system operates in the following steps:

[0121] 1. User settings input:

[0122] The user selects the content for which they want to avoid spoilers (e.g., "TV Drama A") from the interface on their electronic device.

[0123] Check the settings and press the "Save" button to send the settings to the server.

[0124] 2. Server generation AI model load:

[0125] When the server starts up, it loads the generative artificial intelligence model into memory and prepares it for text analysis.

[0126] 3. Get news articles:

[0127] When a user requests that a particular news article be displayed, the server retrieves the appropriate news article from the database.

[0128] 4. Spoiler Check:

[0129] The server inputs the headlines and content of the retrieved news articles into the generative AI model.

[0130] The generative AI model uses prompts to analyze articles and detect spoilers.

[0131] For example, enter the following prompt:

[0132] "If this news article contains spoilers for 'TV Drama A', please detect the relevant parts."

[0133] 5. Customized article generation:

[0134] Based on the analysis results, the server replaces the spoiler information with "[Spoiler not shown]" and reconstructs it.

[0135] 6. View Article:

[0136] The server transmits the reconstructed article to the user's electronic device, where the user can view the article in a secure format.

[0137] Specific examples

[0138] A specific example of operation is shown below.

[0139] example:

[0140] The user configures spoiler protection for "TV Drama A" and sends this information from their device to the server. When the user then requests the display of news article ID "1234," the server retrieves the relevant article from the database. The server then analyzes the article content using a generative AI model and replaces the spoiler portion with "[Spoiler Not Displayed]." Finally, the server sends this reconstructed article to the user's electronic device and displays it to the user.

[0141] The embodiments of the present invention allow users to browse news articles safely while avoiding spoilers. Furthermore, the accuracy of analysis is continually improved based on user feedback, further increasing the usefulness of the system.

[0142] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0143] Program processing flow

[0144] Step 1: Enter user settings

[0145] Input: The user enters the content they want to avoid spoilers for in the device's settings screen.

[0146] Specific behavior:

[0147] Users access an interface on an electronic device such as a smartphone or computer.

[0148] Open the settings screen and select content such as "TV Drama A."

[0149] Click the "Save" button to confirm the settings.

[0150] Output: The configuration information is sent to the server.

[0151] Data processing and calculation:

[0152] The terminal collects the user's setting information and sends it to the server as an HTTP request.

[0153] The setting information includes a user ID and a list of content for which spoilers are to be avoided.

[0154] Step 2: Server generation AI model load

[0155] Input: The server starts or receives a user request.

[0156] Specific behavior:

[0157] The server loads a generative artificial intelligence model (e.g., BERT or GPT-3) into memory when the system starts up.

[0158] Update the model cache as needed.

[0159] Output: The generative AI model is loaded into memory and ready to use.

[0160] Data processing and calculation:

[0161] Use a machine learning library (e.g., TensorFlow or PyTorch) to load the model.

[0162] The model is initialized and preprocessed.

[0163] Step 3: Get news articles

[0164] Input: A user requests that a specific news article be displayed.

[0165] Specific behavior:

[0166] A user sends a request to display a news article from an electronic device.

[0167] The server runs an SQL query against the database to retrieve the headlines and content of the relevant news articles.

[0168] Output: News article headline and content data.

[0169] Data processing and calculation:

[0170] Execute an SQL query to retrieve news article information from the database.

[0171] The acquired data is structured and stored in memory.

[0172] Step 4: Spoiler check

[0173] Input: News article headline and content, user preferences.

[0174] Specific behavior:

[0175] The server inputs the retrieved news articles and user settings into the generative AI model.

[0176] As a concrete example, the following prompt sentence is input to the generative AI model:

[0177] "If this news article contains spoilers for 'TV Drama A', please detect the relevant parts."

[0178] The generative AI model analyzes article information and detects whether it contains spoilers.

[0179] Output: Index information of the parts containing spoilers.

[0180] Data processing and calculation:

[0181] Generate a prompt sentence and input it into the model.

[0182] The model analyzes the text and detects the relevant parts (spoiler information).

[0183] Step 5: Generate a customized article

[0184] Input: spoiler check results (index information) and original news article.

[0185] Specific behavior:

[0186] Based on the analysis results, the server replaces the spoiler information with the placeholder "[Spoiler not shown]".

[0187] The entire reconstructed article is temporarily saved.

[0188] Output: A reconstructed news article.

[0189] Data processing and calculation:

[0190] Perform text processing and replace spoilers with placeholders.

[0191] Generate reconstructed article data.

[0192] Step 6: Viewing articles

[0193] Input: A reconstructed news article.

[0194] Specific behavior:

[0195] The server transmits the reconstructed news article to the user's electronic device.

[0196] The terminal renders and displays the news article data received from the server.

[0197] Output: The news article displayed to the user in a safe format.

[0198] Data processing and calculation:

[0199] Generate an HTTP response containing the reconstructed article data.

[0200] The device converts the received data into a format that can be displayed in a browser or application.

[0201] (Application example 1)

[0202] 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."

[0203] There is a growing need to avoid spoilers when viewing news articles and other content. However, existing systems lack sufficient customization for spoiler prevention, making it difficult for users to safely view content. In addition, there is a lack of systems that can train AI models that reflect user feedback.

[0204] 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.

[0205] In this invention, the server, as an application installed on a smartphone, includes means for retrieving news articles from an API and replacing parts containing spoiler information with "[Spoiler Not Displayed]" before displaying them; means for customizing the scope of spoiler information by filtering specific categories and content based on user settings; and means for performing learning to improve the analytical accuracy of the generative artificial intelligence model based on user feedback. This allows users to safely view articles and content they want to view and reduces the risk of spoilers. A "generative artificial intelligence model" is an AI model trained based on machine learning algorithms that is used to analyze news articles and content and determine spoiler information.

[0206] An "electronic device" is a device such as a smartphone, tablet, or computer that a user uses to view news articles.

[0207] An "API" is an application program interface for obtaining user-specified news article information from an external database.

[0208] "Spoiler information" refers to information about the ending or important scenes of a television drama, movie, anime, etc., and is information that may spoil your enjoyment of the content if you know it in advance.

[0209] A "placeholder" is a replacement string or symbol used to hide parts of a page that contain spoilers. A typical example is "[Spoiler Hide]".

[0210] "User settings" refers to customization that the user can make to avoid spoilers about specific content, and allows the user to specify specific categories and content.

[0211] "Feedback" refers to opinions and evaluations provided by users about the content of displayed articles, and is used as data to improve the analysis accuracy of the AI ​​model.

[0212] This invention relates to a system that detects spoilers in news articles and other content and allows users to avoid them. The system consists of an electronic device such as a smartphone, a server, and a generative artificial intelligence model.

[0213] Program processing

[0214] 1. The server receives a request to view a news article from a user's device (such as a smartphone). The user then uses an application installed on the smartphone to view the news article.

[0215] 2. The server retrieves the news articles using an API, which is an application program interface for retrieving article information from an external database.

[0216] 3. To analyze the article information, the server uses a generative artificial intelligence model, which is trained based on a machine learning algorithm and can accurately determine spoilers contained in the article information.

[0217] 4. The server filters spoilers for specific categories or content based on the user's settings, which the user previously entered in the application's settings screen.

[0218] 5. The server analyzes the article information and replaces any detected spoiler information with a placeholder (e.g., "[Spoiler Not Included]"). This process allows users to view the news article in a safe, spoiler-free format.

[0219] 6. When users provide feedback on the displayed article content, the server collects this feedback, which is used as retraining data for the generative AI model to improve its analysis accuracy.

[0220] Hardware and software used

[0221] Hardware: Smartphone (user device), server (analysis and filtering of article information)

[0222] Software: Python, Requests, Transformers (Hugging Face library)

[0223] Specific examples

[0224] As a concrete example, consider the case where a user wants to avoid spoilers for "Drama A." The user registers "Drama A" as a target for spoiler prevention on the settings screen of an application installed on their smartphone. When the user requests the display of news article ID "1234," the server retrieves the corresponding news article through the API. The content of this news article is analyzed by a generative artificial intelligence model, and if spoiler information is detected, the relevant part is replaced with "[Spoiler not shown]." Finally, the server reconstructs the article information in a secure format and sends it to the device to display to the user.

[0225] Example prompt sentence:

[0226] "Analyze whether an article about the latest episode of Drama A contains spoilers about the ending or important scenes."

[0227] As described above, the embodiment of the present invention allows users to read news articles with peace of mind and prevents spoiler information from ruining the enjoyment of the article.

[0228] The flow of the specific processing in Application Example 1 will be explained with reference to FIG. 12. Step 1:

[0229] Users can set spoiler prevention settings

[0230] On the settings screen of the application installed on the smartphone, the user selects the specific content or category (e.g., Drama A) for which they want to avoid spoilers, and saves the settings. The input is the setting information set by the user, and the output is the setting information sent to the server. This setting information is stored on the server in association with the user ID. Specifically, the user enters keywords for the category they selected (e.g., ending, important scene), and presses the save button.

[0231] Step 2:

[0232] The server receives a request to display a news article.

[0233] The user sends a request to display an article from their smartphone. The input is the ID or URL of the news article specified by the user, and the output is the request being received by the server. Specifically, the user selects a specific article from the list of articles in the application and presses the display button to send the request.

[0234] Step 3:

[0235] The server retrieves news articles using an API

[0236] The server calls an API to retrieve news articles from an external database. The input is the article ID and URL, and the output is the retrieved article information. Specifically, the server sends a request to the API endpoint and parses the returned JSON-formatted article data.

[0237] Step 4:

[0238] The server analyzes the article information using a generated artificial intelligence model

[0239] The server uses a generative artificial intelligence model to analyze whether the acquired article information contains spoilers. The input is the article text, and the output is the result of determining whether spoilers are included. Specifically, the article text is input into the model, and the generated output (whether spoilers are present or not) is obtained. This model is implemented using Hugging Face's Transformers library.

[0240] Step 5:

[0241] The server filters spoilers based on configuration information.

[0242] The server filters article content based on a spoiler prevention keyword list set by the user. The input is the article text and user setting information, and the output is the article text with spoiler information filtered out. Specifically, it detects set keywords (e.g., ending, important scene) from the article text and replaces them with placeholders (e.g., [Spoiler not shown]).

[0243] Step 6:

[0244] The server sends the article to the device after processing to prevent spoilers.

[0245] The server sends the article text, from which spoilers have been filtered, to the user's device. The input is the filtered article text, and the output is safe article information that is displayed on the user's device. Specifically, the server sends the filtered article data to the device as an HTTP response, and the device receives it and displays it on the screen.

[0246] Step 7:

[0247] Collect user feedback and incorporate it into the AI ​​model's training data

[0248] The user provides feedback on the content of the displayed article. This feedback is collected by the server and used as retraining data for the generative AI model. The input is the user's feedback information, and the output is a database in which the feedback is saved. Specifically, the user enters their opinion in the feedback form and presses the submit button, and the server saves the information.

[0249] The above are the specific processing steps in this article filtering system.

[0250] 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.

[0251] The present invention relates to a system that analyzes article information, determines whether it contains spoilers, and hides the spoilers as necessary, and further improves the user experience by combining it with an emotion engine that recognizes the user's emotions. Specific embodiments of the system are described below.

[0252] System configuration

[0253] This system consists of a user's device, a server, a generative artificial intelligence (AI) model, and an emotion engine. The user uses the device to view news articles and configure spoiler prevention settings. The server receives this configuration information and retrieves and analyzes news articles. The generative AI model is hosted on the server and is used to analyze article information and determine spoiler information. The emotion engine detects the user's emotions and provides this information to the server.

[0254] Program processing

[0255] 1. User configures spoiler protection:

[0256] The user selects the specific content for which they want to avoid spoilers (for example, "TV Drama A" or "Anime B") from the settings screen on their device, and sends the setting information from the device to the server. The server stores this setting information in association with the user ID.

[0257] 2. The server loads the generative AI model:

[0258] When the server starts up, it loads the generative AI model into memory and prepares it for analyzing article information. This model is trained by a machine learning algorithm and is used to accurately detect spoilers.

[0259] 3. Get news articles and check for spoilers:

[0260] When a user requests a specific news article to be displayed, the server retrieves the headline and content of the relevant news article from the database. The server then inputs the retrieved headline and content into the generative AI model and analyzes whether they contain spoilers. If they are determined to contain spoilers, the server replaces the relevant part with the placeholder "[Spoiler Not Displayed]".

[0261] 4. Emotion Recognition with Emotion Engine:

[0262] When a user uses a device, the emotion engine recognizes the user's emotional state using sensors (such as a camera or microphone) installed on the device. The recognized emotional information is sent to the server in real time.

[0263] 5. Customized Display:

[0264] The server adjusts the display method of spoiler information based on the analysis results, user settings, and the user's emotional state. Based on these results, the server reconstructs the article information so that it does not contain spoilers. The reconstructed content is sent to the device, which then displays the article to the user in a secure format.

[0265] Specific examples

[0266] As a concrete example, consider the case where a user wants to avoid spoilers for "TV Drama A." When the user makes this setting on their device and registers "TV Drama A" as a spoiler prevention target, the server saves this information. Next, when the user attempts to view news article ID "1234," if the headline of this news article is "Review of the latest episode of 'TV Drama A'," the server will use an AI model to analyze the article content. If the AI ​​model determines that the article content contains spoilers, the server will replace the content with "[Spoiler not shown]." At the same time, if the user's emotional information while viewing the device is "surprise" or "discomfort," the server will further take this information into account and adjust the displayed content, reconstructing the article in a way that better takes the user's emotions into consideration.

[0267] Feedback and Learning

[0268] The system also has a function that improves the analytical accuracy of the generative AI model based on user feedback. When users provide feedback on the content of displayed articles, the server collects this information and uses it as retraining data for the AI ​​model. Feedback reflecting users' emotional information is also collected, and this is used to retrain the model for even more accurate filtering. This allows the system's spoiler detection accuracy to improve over time, enabling more accurate filtering.

[0269] In this way, the embodiments of the present invention not only allow users to read news articles with peace of mind, but also provide optimal information display according to their emotional state, allowing them to enjoy a more comfortable and personalized viewing experience.

[0270] The processing flow will be explained below.

[0271] Step 1:

[0272] The user sets up spoiler prevention on the device. The user specifies specific content (e.g., "TV Drama A," "Anime B," etc.) for which they want to avoid spoilers on the device's settings screen.

[0273] Step 2:

[0274] The device sends the setting information to the server, along with the user ID and the specified spoiler prevention target information.

[0275] Step 3:

[0276] The server updates the user settings. Based on the received information, the server saves the spoiler prevention settings associated with the user ID in the database.

[0277] Step 4:

[0278] The server loads the generative AI model. During initialization, the server loads the trained generative AI model into memory, preparing it for analyzing article information.

[0279] Step 5:

[0280] The user requests the display of a news article. The user sends a display request from the terminal to the server specifying a specific news article ID.

[0281] Step 6:

[0282] The device sends a request to the server, which sends the news article ID and user ID information to the server.

[0283] Step 7:

[0284] The server retrieves the news article. The server retrieves the headline and content of the requested article from the database.

[0285] Step 8:

[0286] The server analyzes the article information using the generative AI model. The server inputs the acquired headline and content into the generative AI model and analyzes whether it contains spoilers.

[0287] Step 9:

[0288] The server determines whether the information is a spoiler, and if the generative AI model determines that the information is a spoiler, it identifies the spoiler.

[0289] Step 10:

[0290] The server hides the spoiler information, replaces the spoiler part with the placeholder "[Spoiler hidden]", and modifies the original content.

[0291] Step 11:

[0292] The emotion engine recognizes the user's emotions while using the device. It uses sensors such as the device's camera and microphone to analyze the user's facial expressions and voice to recognize emotions.

[0293] Step 12:

[0294] The device transmits the recognized emotion information to the server. The device transmits the user's emotional state information recognized in real time to the server.

[0295] Step 13:

[0296] The server reconstructs the article information based on the user's emotional state. The server adjusts the displayed content taking into account the user's emotional state in addition to the analysis results and spoiler prevention settings.

[0297] Step 14:

[0298] The server sends the revised article information to the terminal. The server sends the revised headline and content to the terminal.

[0299] Step 15:

[0300] The terminal displays the corrected article information to the user. The terminal displays the received headline and the corrected content to the user.

[0301] Step 16:

[0302] When a user views an article and provides feedback, the user transmits their rating and opinion to the server via the terminal.

[0303] Step 17:

[0304] The server collects the feedback and stores it in a database.

[0305] Step 18:

[0306] The server retrains the generative AI model. The server uses the collected feedback data to retrain the generative AI model to improve its accuracy.

[0307] Through these steps, users can not only browse news articles with peace of mind, but also enjoy a more comfortable and personalized browsing experience by being provided with optimal information display according to their emotional state.

[0308] Example 2

[0309] 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."

[0310] Conventional news article browsing systems often display articles containing unwanted spoilers, which detracts from the user's browsing experience. Furthermore, because they only implement a uniform approach to preventing spoilers without taking into account the user's emotional state, they are unable to provide optimal information to each individual user. Furthermore, there is a lack of effort to improve the accuracy of generative AI models based on user feedback, limiting the accuracy of spoiler detection.

[0311] 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 receiving spoiler prevention settings from a user terminal and registering specific categories or content as spoiler prevention targets; means for analyzing article information on the server using a generative AI model and determining whether the article information may contain spoiler information; means for hiding spoiler information from the user based on the determination result; and means for customizing the display method of spoiler information based on user settings and emotion information generated by an emotion engine. This allows users to avoid unwanted spoiler information and provides optimal information according to their individual emotional state. Furthermore, by improving the accuracy of the generative AI model based on feedback, more accurate spoiler prevention can be achieved.

[0312] A "user terminal" is an electronic device that is operated by a user to view information and change settings.

[0313] The "spoiler prevention setting" is information that a user sets to avoid spoiler information for a specific content or category.

[0314] A "generative AI model" is a model trained using machine learning algorithms to analyze article information and detect spoilers.

[0315] A "server" is a device that provides services over a network and receives and processes requests from user terminals.

[0316] "Spoiler information" is information that includes important details or the ending of content that the user has not yet seen.

[0317] An "emotion engine" is a software or hardware mechanism for recognizing and analyzing a user's emotional state.

[0318] A "placeholder" is alternative text or symbols that are displayed in place of spoiler information.

[0319] "Feedback" refers to information such as usage experience, impressions, and evaluations provided by users, and is used to improve the system.

[0320] "Analysis accuracy" refers to the degree to which the generative AI model can accurately detect spoiler information.

[0321] "News Article" means article content provided via the Internet or other media.

[0322] "Customization" means changing the system's behavior and display methods according to the user's settings and status.

[0323] The present invention relates to a system that analyzes news articles, determines whether they contain spoilers, and hides the spoilers as necessary. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system further improves the user experience. Specific embodiments of the system are described below.

[0324] System configuration

[0325] This system consists of a user device, a server, a generative AI model, and an emotion engine. The user uses the device to view news articles and configure spoiler prevention settings. The server receives this configuration information and retrieves and analyzes news articles. The generative AI model is hosted on the server and is used to analyze news articles and determine spoiler information. The emotion engine detects the user's emotions and provides this information to the server.

[0326] Program processing

[0327] User-defined spoiler prevention settings

[0328] The user selects the specific content for which they want to avoid spoilers, such as "TV dramas" or "movies," on the device's settings screen. The device then sends the selected setting information to the server, which then associates the received setting information with the user ID and stores it in a database.

[0329] Loading of generated AI models by the server

[0330] At startup, the server loads into memory a generative AI model that is trained using machine learning algorithms and used to analyze news articles and identify spoilers.

[0331] Get news articles and check for spoilers

[0332] When a user requests a specific news article to be displayed, the server retrieves the relevant news article from the database. The headline and content of the retrieved article are input into the generative AI model to determine whether it contains spoilers. If the model detects any part that it deems a spoiler, the server replaces that part with "[Spoiler Not Displayed]."

[0333] Emotion recognition by emotion engine

[0334] While a user is browsing a news article on their device, the emotion engine uses the device's built-in camera and microphone to recognize the user's emotional state, and the recognized emotional information is sent to the server in real time.

[0335] Customized View

[0336] The server adjusts the display method of the spoiler information based on the analysis results, user settings, and the user's emotional information. The reconstructed article content is sent to the device, which then displays the news article to the user in a secure format.

[0337] Specific examples

[0338] As a concrete example, consider the case where a user wants to avoid spoilers for a "TV drama." When the user makes this setting on their device and registers "TV drama" as a spoiler prevention target, the server saves this information. Next, when the user attempts to view news article ID "1234," if the headline of this news article is "Review of the latest episode of 'TV drama'," the server will analyze the article content using the generative AI model. If the generative AI model determines that the article content contains spoilers, the server will replace the content with "[Spoiler not displayed]."

[0339] At the same time, if the emotional information displayed by the user while browsing the device is "surprise" or "discomfort," the server will further take this information into account and adjust the displayed content, reconstructing the article in a way that takes the user's emotions into greater consideration.

[0340] Prompt Sentence Examples

[0341] Examples of input prompts for a generative AI model include:

[0342] "Please analyze what spoilers this news article contains."

[0343] "The user has specified that they would like to avoid spoilers for 'TV drama'. Analyze the content of this news article and replace any spoilers with 'Spoiler Hide'."

[0344] Using these prompts, we can see how the generative AI model parses news articles and appropriately detects and filters spoilers.

[0345] This allows users to avoid unwanted spoilers and obtain optimal information according to their emotional state. Furthermore, by improving the accuracy of the generative AI model based on feedback, even more accurate spoiler prevention can be achieved.

[0346] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0347] Step 1:

[0348] User enters and submits spoiler protection settings

[0349] The user opens the settings screen on the device and selects the specific category or content (e.g., "TV dramas" or "movies") for which they want to avoid spoilers. Using the selected information as input, the device sends the setting information to the server. Specifically, the user taps "Add spoiler-protected content," selects "TV dramas," and presses the "Save" button. The device then sends a request to the server stating, "User ID: 123 has set his preference to avoid spoilers for TV dramas."

[0350] Input: Specific categories or content you want to avoid spoilers for

[0351] Output: Sending configuration information to the server

[0352] Step 2:

[0353] Server-based storage of configuration information

[0354] The server receives the setting information sent from the device, associates it with the user ID, and stores it in the database. Specifically, the server analyzes the data it receives and adds the information "User ID: 123" and "Spoiler prevention target: TV drama" to the user information table in the database.

[0355] Input: Setting information sent from the device

[0356] Output: User preferences stored in the database

[0357] Step 3:

[0358] Loading a generative AI model

[0359] When the server starts up, it loads the generative AI model into memory. This generative AI model is trained using a machine learning algorithm and is used to analyze news articles and determine spoiler information. Specifically, when the server is restarted, the generative AI model loading process is automatically executed and the model is loaded into memory. The system log records "The generative AI model has been loaded."

[0360] Input: Start the server

[0361] Output: The loaded generative AI model

[0362] Step 4:

[0363] Processing a request to retrieve news articles

[0364] A user sends a specific news article ID as input from their device, requesting the server to display that article. The server receives this request, searches for the requested news article in its internal database, and retrieves the headline and content of the corresponding news article. Specifically, when a user taps to display "News Article ID: 1234" on their smartphone app, the device sends a request to the server saying, "I want to display article ID 1234." The server retrieves the article from the database and returns the headline and body data, such as "Review of the latest TV drama episode."

[0365] Input: News article ID

[0366] Output: News article headline and content

[0367] Step 5:

[0368] Analysis of spoiler information in article content

[0369] The server inputs the headline and content of the retrieved news article into the generative AI model and analyzes whether or not it contains spoilers. When the model detects any parts that it determines contain spoilers, the server replaces those parts with "[Spoiler not shown]". Specifically, the server sends the generative AI model a prompt message saying, "Analyze the content of article ID: 1234 and detect any parts that contain spoilers." When the generative AI model responds that the parts contain spoilers, it replaces them with "[Spoiler not shown]".

[0370] Input: News article headline and content

[0371] Output: News article with spoilers replaced

[0372] Step 6:

[0373] Emotion recognition by emotion engine

[0374] While a user is viewing a news article on their device, the emotion engine uses the device's built-in camera and microphone to recognize the user's emotional state. The recognized emotional information is sent to the server in real time. Specifically, while the user is viewing a news article, the device's camera captures the user's facial expressions, and the emotion engine detects the "surprise" expression. This information is sent to the server as "User ID: 123 is in a surprised emotional state."

[0375] Input: Emotion information from the device

[0376] Output: Emotion information sent to the server

[0377] Step 7:

[0378] Customized article display

[0379] The server adjusts the display method of spoiler information based on the analysis results, user settings, and the user's emotional information. The reconstructed article content is sent to the device, and the device displays the news article to the user in a safe format. Specifically, when the server receives that the user is in the "surprise" emotional information state, it applies further filtering to avoid spoilers. The reconstructed article content is sent to the device and displayed as a "safe article."

[0380] Input: Analysis results, user settings, emotional information

[0381] Output: Reconstructed news article

[0382] (Application example 2)

[0383] 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."

[0384] In conventional news article and content distribution services, users have difficulty avoiding unwanted spoilers. Furthermore, the optimal display of information based on the user's emotional state can sometimes impair the browsing experience. Therefore, there is a need for a system that allows users to browse content with peace of mind and that displays information appropriate to their emotional state.

[0385] 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 analyzing article information of the electronic device using a generative artificial intelligence model and determining whether it is likely to contain spoiler information, means for preventing spoilers by replacing part of the article containing spoiler information with a placeholder, emotion recognition means for detecting the emotional state of the user, and means for adjusting the display method of the article information based on the detected emotion information. This allows the user to avoid spoiler information and also makes it possible to display information optimally according to the user's emotional state.

[0386] A "generative artificial intelligence model" is a model that is trained based on machine learning algorithms to analyze text data and determine specific information.

[0387] "Electronic equipment" is a general term for devices used to process and display digital data, including smartphones, tablets, and personal computers.

[0388] "Spoiler information" is information that reveals important developments or endings of content such as movies, dramas, anime, and books in advance.

[0389] A "means for determining" is a method or device for utilizing a generative AI model to analyze text data and identify whether it contains specific information.

[0390] A "placeholder" is a substitute display string that is temporarily used to hide the original information.

[0391] "Emotion recognition means" refers to a method or device for detecting and analyzing emotions from a user's facial expressions, voice, actions, etc.

[0392] "User settings" refers to setting information that allows a user to customize the operating conditions of the system and the display contents based on their own preferences.

[0393] "Customizable means" means a method or device for changing or adjusting the system's functionality or display content based on user settings.

[0394] "Emotion information" is data indicating the emotional state of the user detected by the emotion recognition means.

[0395] The "means for adjusting the display method of article information" is a method or device for changing the content or format of information displayed to a user based on emotion information.

[0396] The system for implementing this invention is composed of a user terminal, a server, a generative AI model, and emotion recognition means. This allows users to browse news articles and other content with peace of mind, and displays information appropriate to their emotional state.

[0397] System configuration

[0398] 1. On the user's device:

[0399] Users browse news articles and other content using electronic devices such as smartphones, tablets, smart glasses, and head-mounted displays. These devices are equipped with emotion recognition sensors (cameras and microphones) to detect the user's emotional state in real time.

[0400] 2. Server:

[0401] The server stores and manages user settings, a news article database, a generative AI model, emotion recognition means, and a feedback database. The server has the following functions:

[0402] Loading the generative AI model: When the server starts up, it loads the generative AI model into memory to analyze article information. This generative AI model is trained based on machine learning algorithms and has the ability to detect spoilers.

[0403] Acquisition and analysis of article information: When a user attempts to view a news article, the server retrieves the relevant news article from the database and passes it to the generative AI model to analyze spoiler information.

[0404] Spoiler hiding: If the analysis reveals that a spoiler is included, the server replaces the relevant part with the placeholder "[Spoiler hidden]" and displays it to the user.

[0405] Use of emotional information: The server collects the user's emotional information sent from the device and adjusts the way article information is displayed based on this information.

[0406] User settings management: Users can set the spoiler categories and content they want to avoid according to their preferences, and the server stores and manages this setting information.

[0407] 3. Emotion recognition means:

[0408] The emotion recognition means analyzes the user's facial expressions and voice data to detect emotional information in real time. This emotional information is sent to the server and used to adjust the display method of article information.

[0409] Specific examples of the embodiment

[0410] For example, if a user wants to avoid spoilers for the movie "Movie X," they can register "Movie X" as a spoiler-protected content on the device's settings screen. The server receives and stores this setting information.

[0411] When a user attempts to view an article related to "Movie X," the server retrieves the article from the database, passes it to the generative AI model, and analyzes it for spoilers. If spoilers are found, the server replaces the relevant part with "[Spoiler Not Displayed]" and delivers it to the user.

[0412] At the same time, if the user's emotional state is surprise or displeasure, the server will adjust the way the article is displayed based on the emotional information, providing the user with a better browsing experience.

[0413] Examples of instructions and prompts

[0414] An example of a prompt sentence when using the sentiment analysis API is as follows:

[0415] python

[0416] Prompt sentence for using the sentiment analysis API

[0417] import requests

[0418] def analyze_sentiment(audio_file_path):

[0419] url = "http: / / sentiment-api.example.com / analyze"

[0420] files = {'file': open(audio_file_path, 'rb')}

[0421] response = requests.post(url, files=files)

[0422] return response.json()

[0423] Analyze audio files to recognize emotions

[0424] emotion_result = analyze_sentiment("path / to / user_audio.wav")

[0425] print(emotion_result)

[0426] This prompt shows how to send an audio file to the sentiment analysis API and receive the sentiment result.

[0427] This allows the user to avoid spoiler information and enjoy optimal information display suited to their own emotional state.

[0428] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0429] Step 1:

[0430] The user sets up spoiler prevention on the device. The user opens the device's settings screen, selects the specific content for which they want to avoid spoilers (e.g., "Movie X"), and enters the setting information. This setting information is sent by the device to the server. The input data consists of the user ID and setting information, which the server receives and stores in a database in association with the user ID.

[0431] Step 2:

[0432] The server loads the generative AI model. At startup, the server loads the generative AI model into memory and prepares it for analyzing article data. This model is trained based on a machine learning algorithm and is used to detect spoilers from news articles. The input data are the parameters of the generative AI model, and the output is the state of the model loaded in memory.

[0433] Step 3:

[0434] A user requests to view an article. When a user uses a device to view a specific news article, the device sends the request to the server. This request includes the article ID. The server receives the user's request and the article ID as input data.

[0435] Step 4:

[0436] The server retrieves and analyzes news articles. The server retrieves the headline and content of the news article corresponding to the specified article ID from the database. The retrieved article content is passed to a generative AI model, which analyzes whether it contains spoilers. The input data is the text data of the news article, and the output data is the spoiler detection results.

[0437] Step 5:

[0438] Concealment of spoiler information. If the generative AI model determines that an article contains spoiler information, it replaces the relevant part with "[Spoiler hidden]." The input data is the spoiler detection result and the text data of the article, and the output data is the text data of the article with the spoiler hidden.

[0439] Step 6:

[0440] Emotion detection using emotion recognition means. When a user uses a device to view an article, sensors (camera and microphone) installed on the device detect the user's emotional state in real time and send that data to the server. The input data is the user's facial expression and voice data, and the output data is the recognized emotional information.

[0441] Step 7:

[0442] Adjustment of article display based on emotional information. The server adjusts the display method of article information based on the user's emotional information obtained by the emotion recognition means. For example, if the user expresses emotions such as "surprise" or "discomfort," the server further adjusts the display content of the article and uses appropriate placeholders. The input data is the emotional information and the text data of the concealed article, and the output data is the adjusted article data that is finally displayed to the user.

[0443] Step 8:

[0444] Delivery of the adjusted article. The server finally sends the adjusted article to the device, where the user views it. The input data is the adjusted article data, and the output data is the article displayed on the user's device. This allows the user to avoid spoilers and experience information displayed optimally according to their emotional state.

[0445] In this way, each step works in cooperation with the others, allowing users to browse news articles and other content safely and comfortably.

[0446] 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.

[0447] 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.

[0448] 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.

[0449] [Second embodiment]

[0450] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0451] 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.

[0452] 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).

[0453] 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.

[0454] 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.

[0455] 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).

[0456] 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. 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.

[0457] 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.

[0458] 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.

[0459] 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.

[0460] 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.

[0461] 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."

[0462] The present invention relates to a system that analyzes article information, determines whether it contains spoiler information, and hides the spoiler information as necessary, and a specific embodiment thereof will be described below.

[0463] System configuration

[0464] This system consists of a user's device, a server, and a generative artificial intelligence (AI) model. The user uses the device to view news articles and configure spoiler prevention settings. The server receives this configuration information and retrieves and analyzes news articles. The generative AI model is hosted on the server and is used to analyze article information and determine spoiler information.

[0465] Program processing

[0466] 1. User configures spoiler protection:

[0467] The user selects the specific content for which they want to avoid spoilers (for example, "TV Drama A" or "Anime B") from the settings screen on their device, and sends the setting information from the device to the server. The server stores this setting information in association with the user ID.

[0468] 2. The server loads the generative AI model:

[0469] When the server starts up, it loads the generative AI model into memory and prepares it for analyzing article information. This model is trained by a machine learning algorithm and is used to accurately detect spoilers.

[0470] 3. Get news articles and check for spoilers:

[0471] When a user requests a specific news article to be displayed, the server retrieves the headline and content of the relevant news article from the database. The server then inputs the retrieved headline and content into the generative AI model and analyzes whether they contain spoilers. If they are determined to contain spoilers, the server replaces the relevant part with the placeholder "[Spoiler Not Displayed]".

[0472] 4. Customized Display:

[0473] The server reconstructs the article information based on the analysis results and user settings in a way that does not include spoilers, and sends this reconstructed content to the device, which then displays the article to the user in a secure format.

[0474] Specific examples

[0475] As a concrete example, consider the case where a user wants to avoid spoilers for "TV Drama A." When the user makes this setting on their device and registers "TV Drama A" as a spoiler prevention target, the server saves this information. Next, when the user attempts to view news article ID "1234," if the headline of this news article is "Review of the latest episode of 'TV Drama A'," the server uses an AI model to analyze the article content. If the AI ​​model determines that the article content contains spoilers, the server replaces the relevant part with "[Spoiler not displayed]" and sends it to the device. The device then displays the final content to the user.

[0476] Feedback and Learning

[0477] The system also has a function to improve the analysis accuracy of the generative AI model based on user feedback. When users provide feedback on the content of displayed articles, the server collects this information and uses it as retraining data for the AI ​​model. This allows the system's spoiler detection accuracy to improve over time, enabling more accurate filtering.

[0478] Thus, according to the embodiment of the present invention, the user can read news articles with peace of mind and enjoy the content without spoiling the fun.

[0479] The processing flow will be explained below.

[0480] Step 1:

[0481] The user sets up spoiler prevention on the device. The user specifies specific content (e.g., "TV Drama A," "Anime B," etc.) for which they want to avoid spoilers on the device's settings screen.

[0482] Step 2:

[0483] The device sends the setting information to the server, along with the user ID and the specified spoiler prevention target information.

[0484] Step 3:

[0485] The server updates the user settings. Based on the received information, the server saves the spoiler prevention settings associated with the user ID in the database.

[0486] Step 4:

[0487] The server loads the generative AI model. During initialization, the server loads the trained generative AI model into memory, preparing it for analyzing article information.

[0488] Step 5:

[0489] The user requests the display of a news article. The user sends a display request from the terminal to the server specifying a specific news article ID.

[0490] Step 6:

[0491] The device sends a request to the server, which sends the news article ID and user ID information to the server.

[0492] Step 7:

[0493] The server retrieves the news article. The server retrieves the headline and content of the requested article from the database.

[0494] Step 8:

[0495] The server analyzes the article information using the generative AI model. The server inputs the acquired headline and content into the generative AI model and analyzes whether it contains spoilers.

[0496] Step 9:

[0497] The server determines whether the information is a spoiler, and if the generative AI model determines that the information is a spoiler, it identifies the spoiler.

[0498] Step 10:

[0499] The server hides the spoiler information, replaces the spoiler part with the placeholder "[Spoiler hidden]", and modifies the original content.

[0500] Step 11:

[0501] The server sends the revised article information to the terminal. The server sends the revised headline and content to the terminal.

[0502] Step 12:

[0503] The terminal displays the corrected article information to the user. The terminal displays the received headline and the corrected content to the user.

[0504] Step 13:

[0505] When a user views an article and provides feedback, the user transmits their rating and opinion to the server via the terminal.

[0506] Step 14:

[0507] The server collects the feedback and stores it in a database.

[0508] Step 15:

[0509] The server retrains the generative AI model. The server uses the collected feedback data to retrain the generative AI model to improve its accuracy.

[0510] By following the steps above, users can read news articles with peace of mind and enjoy the content without spoiling it.

[0511] Example 1

[0512] 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."

[0513] In recent years, with the increasing number of news articles and reviews distributed over the Internet, users are at a greater risk of coming across spoilers for content they have not yet viewed. In particular, articles about entertainment content such as TV shows, movies, and anime can unintentionally contain spoilers, which can ruin the user's viewing experience. Conventional systems require users to take measures to avoid spoilers themselves, which is cumbersome and unreliable.

[0514] 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.

[0515] In this invention, the server includes a means for a user to use an electronic device to set spoiler prevention settings and transmit the setting information to the server, a means for the server to analyze article information using a generative artificial intelligence model and determine whether it is likely to contain spoiler information, and a means for hiding the spoiler information from the user based on the determination result. This reduces the risk that users will unintentionally come across spoiler information about content they have not viewed, allowing them to read news articles with peace of mind.

[0516] A "user" is a person who uses the system to set up spoiler protection for news articles and view the content.

[0517] "Electronic device" refers to a device used by a user to access and configure the system, including smartphones, PCs, tablets, etc.

[0518] "Server" refers to the central control unit that receives user preferences and retrieves, analyzes, and reconstructs news articles.

[0519] A "generative artificial intelligence model" is a model trained by a machine learning algorithm to analyze text in news articles to detect spoilers.

[0520] "Spoiler information" refers to information that includes important information or developments about content that the user has not yet viewed, and that may spoil the enjoyment of that content.

[0521] "Settings Information" refers to information about specific content or categories related to spoiler prevention that a user enters through the system.

[0522] "Feedback" refers to user-provided ratings and comments on the displayed article content, which helps improve the system's analysis accuracy.

[0523] A "placeholder" is an alternative expression such as "[Spoiler not shown]" that is used to replace a portion of text that contains spoiler information.

[0524] "Filtering" refers to the process of excluding certain types or content based on user settings.

[0525] A "prompt sentence" is an input sentence given to a generative artificial intelligence model that instructs the model on a specific analysis task.

[0526] The present invention is a system that analyzes news articles, determines whether they contain spoilers, and hides the spoilers as necessary. A specific embodiment of the system is described below. The system consists of a user's electronic device, a server, and a generative artificial intelligence model (generative AI model).

[0527] System configuration

[0528] The system includes the following hardware and software:

[0529] 1. User's electronic devices:

[0530] Users access the system using electronic devices such as smartphones, PCs, and tablets.

[0531] Set spoiler protection and read the news article.

[0532] 2. Server:

[0533] The server receives and stores user-submitted configuration information, and also retrieves, analyzes, and reconstructs news articles.

[0534] The server analyzes article information using a generative artificial intelligence model (e.g., BERT or GPT-3).

[0535] 3. Generative AI Models:

[0536] The generative artificial intelligence model analyzes the headlines and content of news articles to determine whether they contain spoilers.

[0537] Specific machine learning libraries used include TensorFlow and PyTorch.

[0538] Program processing

[0539] The system operates in the following steps:

[0540] 1. User settings input:

[0541] The user selects the content for which they want to avoid spoilers (e.g., "TV Drama A") from the interface on their electronic device.

[0542] Check the settings and press the "Save" button to send the settings to the server.

[0543] 2. Server generation AI model load:

[0544] When the server starts up, it loads the generative artificial intelligence model into memory and prepares it for text analysis.

[0545] 3. Get news articles:

[0546] When a user requests that a particular news article be displayed, the server retrieves the appropriate news article from the database.

[0547] 4. Spoiler Check:

[0548] The server inputs the headlines and content of the retrieved news articles into the generative AI model.

[0549] The generative AI model uses prompts to analyze articles and detect spoilers.

[0550] For example, enter the following prompt:

[0551] "If this news article contains spoilers for 'TV Drama A', please detect the relevant parts."

[0552] 5. Customized article generation:

[0553] Based on the analysis results, the server replaces the spoiler information with "[Spoiler not shown]" and reconstructs it.

[0554] 6. View Article:

[0555] The server transmits the reconstructed article to the user's electronic device, where the user can view the article in a secure format.

[0556] Specific examples

[0557] A specific example of operation is shown below.

[0558] example:

[0559] The user configures spoiler protection for "TV Drama A" and sends this information from their device to the server. When the user then requests the display of news article ID "1234," the server retrieves the relevant article from the database. The server then analyzes the article content using a generative AI model and replaces the spoiler portion with "[Spoiler Not Displayed]." Finally, the server sends this reconstructed article to the user's electronic device and displays it to the user.

[0560] The embodiments of the present invention allow users to browse news articles safely while avoiding spoilers. Furthermore, the accuracy of analysis is continually improved based on user feedback, further increasing the usefulness of the system.

[0561] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0562] Program processing flow

[0563] Step 1: Enter user settings

[0564] Input: The user enters the content they want to avoid spoilers for in the device's settings screen.

[0565] Specific behavior:

[0566] Users access an interface on an electronic device such as a smartphone or computer.

[0567] Open the settings screen and select content such as "TV Drama A."

[0568] Click the "Save" button to confirm the settings.

[0569] Output: The configuration information is sent to the server.

[0570] Data processing and calculation:

[0571] The terminal collects the user's setting information and sends it to the server as an HTTP request.

[0572] The setting information includes a user ID and a list of content for which spoilers are to be avoided.

[0573] Step 2: Server generation AI model load

[0574] Input: The server starts or receives a user request.

[0575] Specific behavior:

[0576] The server loads a generative artificial intelligence model (e.g., BERT or GPT-3) into memory when the system starts up.

[0577] Update the model cache as needed.

[0578] Output: The generative AI model is loaded into memory and ready to use.

[0579] Data processing and calculation:

[0580] Use a machine learning library (e.g., TensorFlow or PyTorch) to load the model.

[0581] The model is initialized and preprocessed.

[0582] Step 3: Get news articles

[0583] Input: A user requests that a specific news article be displayed.

[0584] Specific behavior:

[0585] A user sends a request to display a news article from an electronic device.

[0586] The server runs an SQL query against the database to retrieve the headlines and content of the relevant news articles.

[0587] Output: News article headline and content data.

[0588] Data processing and calculation:

[0589] Execute an SQL query to retrieve news article information from the database.

[0590] The acquired data is structured and stored in memory.

[0591] Step 4: Spoiler check

[0592] Input: News article headline and content, user preferences.

[0593] Specific behavior:

[0594] The server inputs the retrieved news articles and user settings into the generative AI model.

[0595] As a concrete example, the following prompt sentence is input to the generative AI model:

[0596] "If this news article contains spoilers for 'TV Drama A', please detect the relevant parts."

[0597] The generative AI model analyzes article information and detects whether it contains spoilers.

[0598] Output: Index information of the parts containing spoilers.

[0599] Data processing and calculation:

[0600] Generate a prompt sentence and input it into the model.

[0601] The model analyzes the text and detects the relevant parts (spoiler information).

[0602] Step 5: Generate a customized article

[0603] Input: spoiler check results (index information) and original news article.

[0604] Specific behavior:

[0605] Based on the analysis results, the server replaces the spoiler information with the placeholder "[Spoiler not shown]".

[0606] The entire reconstructed article is temporarily saved.

[0607] Output: A reconstructed news article.

[0608] Data processing and calculation:

[0609] Perform text processing and replace spoilers with placeholders.

[0610] Generate reconstructed article data.

[0611] Step 6: Viewing articles

[0612] Input: A reconstructed news article.

[0613] Specific behavior:

[0614] The server transmits the reconstructed news article to the user's electronic device.

[0615] The terminal renders and displays the news article data received from the server.

[0616] Output: The news article displayed to the user in a safe format.

[0617] Data processing and calculation:

[0618] Generate an HTTP response containing the reconstructed article data.

[0619] The device converts the received data into a format that can be displayed in a browser or application.

[0620] (Application example 1)

[0621] 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."

[0622] There is a growing need to avoid spoilers when viewing news articles and other content. However, existing systems lack sufficient customization for spoiler prevention, making it difficult for users to safely view content. In addition, there is a lack of systems that can train AI models that reflect user feedback.

[0623] 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.

[0624] In this invention, the server, as an application installed on a smartphone, includes means for retrieving news articles from an API and replacing parts containing spoiler information with "[Spoiler Not Displayed]" before displaying them; means for customizing the scope of spoiler information by filtering specific categories and content based on user settings; and means for performing learning to improve the analytical accuracy of the generative artificial intelligence model based on user feedback. This allows users to safely view articles and content they want to view and reduces the risk of spoilers. A "generative artificial intelligence model" is an AI model trained based on machine learning algorithms that is used to analyze news articles and content and determine spoiler information.

[0625] An "electronic device" is a device such as a smartphone, tablet, or computer that a user uses to view news articles.

[0626] An "API" is an application program interface for obtaining user-specified news article information from an external database.

[0627] "Spoiler information" refers to information about the ending or important scenes of a television drama, movie, anime, etc., and is information that may spoil your enjoyment of the content if you know it in advance.

[0628] A "placeholder" is a replacement string or symbol used to hide parts of a page that contain spoilers. A typical example is "[Spoiler Hide]".

[0629] "User settings" refers to customization that the user can make to avoid spoilers about specific content, and allows the user to specify specific categories and content.

[0630] "Feedback" refers to opinions and evaluations provided by users about the content of displayed articles, and is used as data to improve the analysis accuracy of the AI ​​model.

[0631] This invention relates to a system that detects spoilers in news articles and other content and allows users to avoid them. The system consists of an electronic device such as a smartphone, a server, and a generative artificial intelligence model.

[0632] Program processing

[0633] 1. The server receives a request to view a news article from a user's device (such as a smartphone). The user then uses an application installed on the smartphone to view the news article.

[0634] 2. The server retrieves the news articles using an API, which is an application program interface for retrieving article information from an external database.

[0635] 3. To analyze the article information, the server uses a generative artificial intelligence model, which is trained based on a machine learning algorithm and can accurately determine spoilers contained in the article information.

[0636] 4. The server filters spoilers for specific categories or content based on the user's settings, which the user previously entered in the application's settings screen.

[0637] 5. The server analyzes the article information and replaces any detected spoiler information with a placeholder (e.g., "[Spoiler Not Included]"). This process allows users to view the news article in a safe, spoiler-free format.

[0638] 6. When users provide feedback on the displayed article content, the server collects this feedback, which is used as retraining data for the generative AI model to improve its analysis accuracy.

[0639] Hardware and software used

[0640] Hardware: Smartphone (user device), server (analysis and filtering of article information)

[0641] Software: Python, Requests, Transformers (Hugging Face library)

[0642] Specific examples

[0643] As a concrete example, consider the case where a user wants to avoid spoilers for "Drama A." The user registers "Drama A" as a target for spoiler prevention on the settings screen of an application installed on their smartphone. When the user requests the display of news article ID "1234," the server retrieves the corresponding news article through the API. The content of this news article is analyzed by a generative artificial intelligence model, and if spoiler information is detected, the relevant part is replaced with "[Spoiler not shown]." Finally, the server reconstructs the article information in a secure format and sends it to the device to display to the user.

[0644] Example prompt sentence:

[0645] "Analyze whether an article about the latest episode of Drama A contains spoilers about the ending or important scenes."

[0646] As described above, the embodiment of the present invention allows users to read news articles with peace of mind and prevents spoiler information from ruining the enjoyment of the article.

[0647] The flow of the specific processing in Application Example 1 will be explained with reference to FIG. 12. Step 1:

[0648] Users can set spoiler prevention settings

[0649] On the settings screen of the application installed on the smartphone, the user selects the specific content or category (e.g., Drama A) for which they want to avoid spoilers, and saves the settings. The input is the setting information set by the user, and the output is the setting information sent to the server. This setting information is stored on the server in association with the user ID. Specifically, the user enters keywords for the category they selected (e.g., ending, important scene), and presses the save button.

[0650] Step 2:

[0651] The server receives a request to display a news article.

[0652] The user sends a request to display an article from their smartphone. The input is the ID or URL of the news article specified by the user, and the output is the request being received by the server. Specifically, the user selects a specific article from the list of articles in the application and presses the display button to send the request.

[0653] Step 3:

[0654] The server retrieves news articles using an API

[0655] The server calls an API to retrieve news articles from an external database. The input is the article ID and URL, and the output is the retrieved article information. Specifically, the server sends a request to the API endpoint and parses the returned JSON-formatted article data.

[0656] Step 4:

[0657] The server analyzes the article information using a generated artificial intelligence model

[0658] The server uses a generative artificial intelligence model to analyze whether the acquired article information contains spoilers. The input is the article text, and the output is the result of determining whether spoilers are included. Specifically, the article text is input into the model, and the generated output (whether spoilers are present or not) is obtained. This model is implemented using Hugging Face's Transformers library.

[0659] Step 5:

[0660] The server filters spoilers based on configuration information.

[0661] The server filters article content based on a spoiler prevention keyword list set by the user. The input is the article text and user setting information, and the output is the article text with spoiler information filtered out. Specifically, it detects set keywords (e.g., ending, important scene) from the article text and replaces them with placeholders (e.g., [Spoiler not shown]).

[0662] Step 6:

[0663] The server sends the article to the device after processing to prevent spoilers.

[0664] The server sends the article text, from which spoilers have been filtered, to the user's device. The input is the filtered article text, and the output is safe article information that is displayed on the user's device. Specifically, the server sends the filtered article data to the device as an HTTP response, and the device receives it and displays it on the screen.

[0665] Step 7:

[0666] Collect user feedback and incorporate it into the AI ​​model's training data

[0667] The user provides feedback on the content of the displayed article. This feedback is collected by the server and used as retraining data for the generative AI model. The input is the user's feedback information, and the output is a database in which the feedback is saved. Specifically, the user enters their opinion in the feedback form and presses the submit button, and the server saves the information.

[0668] The above are the specific processing steps in this article filtering system.

[0669] 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.

[0670] The present invention relates to a system that analyzes article information, determines whether it contains spoilers, and hides the spoilers as necessary, and further improves the user experience by combining it with an emotion engine that recognizes the user's emotions. Specific embodiments of the system are described below.

[0671] System configuration

[0672] This system consists of a user's device, a server, a generative artificial intelligence (AI) model, and an emotion engine. The user uses the device to view news articles and configure spoiler prevention settings. The server receives this configuration information and retrieves and analyzes news articles. The generative AI model is hosted on the server and is used to analyze article information and determine spoiler information. The emotion engine detects the user's emotions and provides this information to the server.

[0673] Program processing

[0674] 1. User configures spoiler protection:

[0675] The user selects the specific content for which they want to avoid spoilers (for example, "TV Drama A" or "Anime B") from the settings screen on their device, and sends the setting information from the device to the server. The server stores this setting information in association with the user ID.

[0676] 2. The server loads the generative AI model:

[0677] When the server starts up, it loads the generative AI model into memory and prepares it for analyzing article information. This model is trained by a machine learning algorithm and is used to accurately detect spoilers.

[0678] 3. Get news articles and check for spoilers:

[0679] When a user requests a specific news article to be displayed, the server retrieves the headline and content of the relevant news article from the database. The server then inputs the retrieved headline and content into the generative AI model and analyzes whether they contain spoilers. If they are determined to contain spoilers, the server replaces the relevant part with the placeholder "[Spoiler Not Displayed]".

[0680] 4. Emotion Recognition with Emotion Engine:

[0681] When a user uses a device, the emotion engine recognizes the user's emotional state using sensors (such as a camera or microphone) installed on the device. The recognized emotional information is sent to the server in real time.

[0682] 5. Customized Display:

[0683] The server adjusts the display method of spoiler information based on the analysis results, user settings, and the user's emotional state. Based on these results, the server reconstructs the article information so that it does not contain spoilers. The reconstructed content is sent to the device, which then displays the article to the user in a secure format.

[0684] Specific examples

[0685] As a concrete example, consider the case where a user wants to avoid spoilers for "TV Drama A." When the user makes this setting on their device and registers "TV Drama A" as a spoiler prevention target, the server saves this information. Next, when the user attempts to view news article ID "1234," if the headline of this news article is "Review of the latest episode of 'TV Drama A'," the server will use an AI model to analyze the article content. If the AI ​​model determines that the article content contains spoilers, the server will replace the content with "[Spoiler not shown]." At the same time, if the user's emotional information while viewing the device is "surprise" or "discomfort," the server will further take this information into account and adjust the displayed content, reconstructing the article in a way that better takes the user's emotions into consideration.

[0686] Feedback and Learning

[0687] The system also has a function that improves the analytical accuracy of the generative AI model based on user feedback. When users provide feedback on the content of displayed articles, the server collects this information and uses it as retraining data for the AI ​​model. Feedback reflecting users' emotional information is also collected, and this is used to retrain the model for even more accurate filtering. This allows the system's spoiler detection accuracy to improve over time, enabling more accurate filtering.

[0688] In this way, the embodiments of the present invention not only allow users to read news articles with peace of mind, but also provide optimal information display according to their emotional state, allowing them to enjoy a more comfortable and personalized viewing experience.

[0689] The processing flow will be explained below.

[0690] Step 1:

[0691] The user sets up spoiler prevention on the device. The user specifies specific content (e.g., "TV Drama A," "Anime B," etc.) for which they want to avoid spoilers on the device's settings screen.

[0692] Step 2:

[0693] The device sends the setting information to the server, along with the user ID and the specified spoiler prevention target information.

[0694] Step 3:

[0695] The server updates the user settings. Based on the received information, the server saves the spoiler prevention settings associated with the user ID in the database.

[0696] Step 4:

[0697] The server loads the generative AI model. During initialization, the server loads the trained generative AI model into memory, preparing it for analyzing article information.

[0698] Step 5:

[0699] The user requests the display of a news article. The user sends a display request from the terminal to the server specifying a specific news article ID.

[0700] Step 6:

[0701] The device sends a request to the server, which sends the news article ID and user ID information to the server.

[0702] Step 7:

[0703] The server retrieves the news article. The server retrieves the headline and content of the requested article from the database.

[0704] Step 8:

[0705] The server analyzes the article information using the generative AI model. The server inputs the acquired headline and content into the generative AI model and analyzes whether it contains spoilers.

[0706] Step 9:

[0707] The server determines whether the information is a spoiler, and if the generative AI model determines that the information is a spoiler, it identifies the spoiler.

[0708] Step 10:

[0709] The server hides the spoiler information, replaces the spoiler part with the placeholder "[Spoiler hidden]", and modifies the original content.

[0710] Step 11:

[0711] The emotion engine recognizes the user's emotions while using the device. It uses sensors such as the device's camera and microphone to analyze the user's facial expressions and voice to recognize emotions.

[0712] Step 12:

[0713] The device transmits the recognized emotion information to the server. The device transmits the user's emotional state information recognized in real time to the server.

[0714] Step 13:

[0715] The server reconstructs the article information based on the user's emotional state. The server adjusts the displayed content taking into account the user's emotional state in addition to the analysis results and spoiler prevention settings.

[0716] Step 14:

[0717] The server sends the revised article information to the terminal. The server sends the revised headline and content to the terminal.

[0718] Step 15:

[0719] The terminal displays the corrected article information to the user. The terminal displays the received headline and the corrected content to the user.

[0720] Step 16:

[0721] When a user views an article and provides feedback, the user transmits their rating and opinion to the server via the terminal.

[0722] Step 17:

[0723] The server collects the feedback and stores it in a database.

[0724] Step 18:

[0725] The server retrains the generative AI model. The server uses the collected feedback data to retrain the generative AI model to improve its accuracy.

[0726] Through these steps, users can not only browse news articles with peace of mind, but also enjoy a more comfortable and personalized browsing experience by being provided with optimal information display according to their emotional state.

[0727] Example 2

[0728] 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."

[0729] Conventional news article browsing systems often display articles containing unwanted spoilers, which detracts from the user's browsing experience. Furthermore, because they only implement a uniform approach to preventing spoilers without taking into account the user's emotional state, they are unable to provide optimal information to each individual user. Furthermore, there is a lack of effort to improve the accuracy of generative AI models based on user feedback, limiting the accuracy of spoiler detection.

[0730] 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 receiving spoiler prevention settings from a user terminal and registering specific categories or content as spoiler prevention targets; means for analyzing article information on the server using a generative AI model and determining whether the article information may contain spoiler information; means for hiding spoiler information from the user based on the determination result; and means for customizing the display method of spoiler information based on user settings and emotion information generated by an emotion engine. This allows users to avoid unwanted spoiler information and provides optimal information according to their individual emotional state. Furthermore, by improving the accuracy of the generative AI model based on feedback, more accurate spoiler prevention can be achieved.

[0731] A "user terminal" is an electronic device that is operated by a user to view information and change settings.

[0732] The "spoiler prevention setting" is information that a user sets to avoid spoiler information for a specific content or category.

[0733] A "generative AI model" is a model trained using machine learning algorithms to analyze article information and detect spoilers.

[0734] A "server" is a device that provides services over a network and receives and processes requests from user terminals.

[0735] "Spoiler information" is information that includes important details or the ending of content that the user has not yet seen.

[0736] An "emotion engine" is a software or hardware mechanism for recognizing and analyzing a user's emotional state.

[0737] A "placeholder" is alternative text or symbols that are displayed in place of spoiler information.

[0738] "Feedback" refers to information such as usage experience, impressions, and evaluations provided by users, and is used to improve the system.

[0739] "Analysis accuracy" refers to the degree to which the generative AI model can accurately detect spoiler information.

[0740] "News Article" means article content provided via the Internet or other media.

[0741] "Customization" means changing the system's behavior and display methods according to the user's settings and status.

[0742] The present invention relates to a system that analyzes news articles, determines whether they contain spoilers, and hides the spoilers as necessary. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system further improves the user experience. Specific embodiments of the system are described below.

[0743] System configuration

[0744] This system consists of a user device, a server, a generative AI model, and an emotion engine. The user uses the device to view news articles and configure spoiler prevention settings. The server receives this configuration information and retrieves and analyzes news articles. The generative AI model is hosted on the server and is used to analyze news articles and determine spoiler information. The emotion engine detects the user's emotions and provides this information to the server.

[0745] Program processing

[0746] User-defined spoiler prevention settings

[0747] The user selects the specific content for which they want to avoid spoilers, such as "TV dramas" or "movies," on the device's settings screen. The device then sends the selected setting information to the server, which then associates the received setting information with the user ID and stores it in a database.

[0748] Loading of generated AI models by the server

[0749] At startup, the server loads into memory a generative AI model that is trained using machine learning algorithms and used to analyze news articles and identify spoilers.

[0750] Get news articles and check for spoilers

[0751] When a user requests a specific news article to be displayed, the server retrieves the relevant news article from the database. The headline and content of the retrieved article are input into the generative AI model to determine whether it contains spoilers. If the model detects any part that it deems a spoiler, the server replaces that part with "[Spoiler Not Displayed]."

[0752] Emotion recognition by emotion engine

[0753] While a user is browsing a news article on their device, the emotion engine uses the device's built-in camera and microphone to recognize the user's emotional state, and the recognized emotional information is sent to the server in real time.

[0754] Customized View

[0755] The server adjusts the display method of the spoiler information based on the analysis results, user settings, and the user's emotional information. The reconstructed article content is sent to the device, which then displays the news article to the user in a secure format.

[0756] Specific examples

[0757] As a concrete example, consider the case where a user wants to avoid spoilers for a "TV drama." When the user makes this setting on their device and registers "TV drama" as a spoiler prevention target, the server saves this information. Next, when the user attempts to view news article ID "1234," if the headline of this news article is "Review of the latest episode of 'TV drama'," the server will analyze the article content using the generative AI model. If the generative AI model determines that the article content contains spoilers, the server will replace the content with "[Spoiler not displayed]."

[0758] At the same time, if the emotional information displayed by the user while browsing the device is "surprise" or "discomfort," the server will further take this information into account and adjust the displayed content, reconstructing the article in a way that takes the user's emotions into greater consideration.

[0759] Prompt Sentence Examples

[0760] Examples of input prompts for a generative AI model include:

[0761] "Please analyze what spoilers this news article contains."

[0762] "The user has specified that they would like to avoid spoilers for 'TV drama'. Analyze the content of this news article and replace any spoilers with 'Spoiler Hide'."

[0763] Using these prompts, we can see how the generative AI model parses news articles and appropriately detects and filters spoilers.

[0764] This allows users to avoid unwanted spoilers and obtain optimal information according to their emotional state. Furthermore, by improving the accuracy of the generative AI model based on feedback, even more accurate spoiler prevention can be achieved.

[0765] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0766] Step 1:

[0767] User enters and submits spoiler protection settings

[0768] The user opens the settings screen on the device and selects the specific category or content (e.g., "TV dramas" or "movies") for which they want to avoid spoilers. Using the selected information as input, the device sends the setting information to the server. Specifically, the user taps "Add spoiler-protected content," selects "TV dramas," and presses the "Save" button. The device then sends a request to the server stating, "User ID: 123 has set his preference to avoid spoilers for TV dramas."

[0769] Input: Specific categories or content you want to avoid spoilers for

[0770] Output: Sending configuration information to the server

[0771] Step 2:

[0772] Server-based storage of configuration information

[0773] The server receives the setting information sent from the device, associates it with the user ID, and stores it in the database. Specifically, the server analyzes the data it receives and adds the information "User ID: 123" and "Spoiler prevention target: TV drama" to the user information table in the database.

[0774] Input: Setting information sent from the device

[0775] Output: User preferences stored in the database

[0776] Step 3:

[0777] Loading a generative AI model

[0778] When the server starts up, it loads the generative AI model into memory. This generative AI model is trained using a machine learning algorithm and is used to analyze news articles and determine spoiler information. Specifically, when the server is restarted, the generative AI model loading process is automatically executed and the model is loaded into memory. The system log records "The generative AI model has been loaded."

[0779] Input: Start the server

[0780] Output: The loaded generative AI model

[0781] Step 4:

[0782] Processing a request to retrieve news articles

[0783] A user sends a specific news article ID as input from their device, requesting the server to display that article. The server receives this request, searches for the requested news article in its internal database, and retrieves the headline and content of the corresponding news article. Specifically, when a user taps to display "News Article ID: 1234" on their smartphone app, the device sends a request to the server saying, "I want to display article ID 1234." The server retrieves the article from the database and returns the headline and body data, such as "Review of the latest TV drama episode."

[0784] Input: News article ID

[0785] Output: News article headline and content

[0786] Step 5:

[0787] Analysis of spoiler information in article content

[0788] The server inputs the headline and content of the retrieved news article into the generative AI model and analyzes whether or not it contains spoilers. When the model detects any parts that it determines contain spoilers, the server replaces those parts with "[Spoiler not shown]". Specifically, the server sends the generative AI model a prompt message saying, "Analyze the content of article ID: 1234 and detect any parts that contain spoilers." When the generative AI model responds that the parts contain spoilers, it replaces them with "[Spoiler not shown]".

[0789] Input: News article headline and content

[0790] Output: News article with spoilers replaced

[0791] Step 6:

[0792] Emotion recognition by emotion engine

[0793] While a user is viewing a news article on their device, the emotion engine uses the device's built-in camera and microphone to recognize the user's emotional state. The recognized emotional information is sent to the server in real time. Specifically, while the user is viewing a news article, the device's camera captures the user's facial expressions, and the emotion engine detects the "surprise" expression. This information is sent to the server as "User ID: 123 is in a surprised emotional state."

[0794] Input: Emotion information from the device

[0795] Output: Emotion information sent to the server

[0796] Step 7:

[0797] Customized article display

[0798] The server adjusts the display method of spoiler information based on the analysis results, user settings, and the user's emotional information. The reconstructed article content is sent to the device, and the device displays the news article to the user in a safe format. Specifically, when the server receives that the user is in the "surprise" emotional information state, it applies further filtering to avoid spoilers. The reconstructed article content is sent to the device and displayed as a "safe article."

[0799] Input: Analysis results, user settings, emotional information

[0800] Output: Reconstructed news article

[0801] (Application example 2)

[0802] 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."

[0803] In conventional news article and content distribution services, users have difficulty avoiding unwanted spoilers. Furthermore, the optimal display of information based on the user's emotional state can sometimes impair the browsing experience. Therefore, there is a need for a system that allows users to browse content with peace of mind and that displays information appropriate to their emotional state.

[0804] 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 analyzing article information of the electronic device using a generative artificial intelligence model and determining whether it is likely to contain spoiler information, means for preventing spoilers by replacing part of the article containing spoiler information with a placeholder, emotion recognition means for detecting the emotional state of the user, and means for adjusting the display method of the article information based on the detected emotion information. This allows the user to avoid spoiler information and also makes it possible to display information optimally according to the user's emotional state.

[0805] A "generative artificial intelligence model" is a model that is trained based on machine learning algorithms to analyze text data and determine specific information.

[0806] "Electronic equipment" is a general term for devices used to process and display digital data, including smartphones, tablets, and personal computers.

[0807] "Spoiler information" is information that reveals important developments or endings of content such as movies, dramas, anime, and books in advance.

[0808] A "means for determining" is a method or device for utilizing a generative AI model to analyze text data and identify whether it contains specific information.

[0809] A "placeholder" is a substitute display string that is temporarily used to hide the original information.

[0810] "Emotion recognition means" refers to a method or device for detecting and analyzing emotions from a user's facial expressions, voice, actions, etc.

[0811] "User settings" refers to setting information that allows a user to customize the operating conditions of the system and the display contents based on their own preferences.

[0812] "Customizable means" means a method or device for changing or adjusting the system's functionality or display content based on user settings.

[0813] "Emotion information" is data indicating the emotional state of the user detected by the emotion recognition means.

[0814] The "means for adjusting the display method of article information" is a method or device for changing the content or format of information displayed to a user based on emotion information.

[0815] The system for implementing this invention is composed of a user terminal, a server, a generative AI model, and emotion recognition means. This allows users to browse news articles and other content with peace of mind, and displays information appropriate to their emotional state.

[0816] System configuration

[0817] 1. On the user's device:

[0818] Users browse news articles and other content using electronic devices such as smartphones, tablets, smart glasses, and head-mounted displays. These devices are equipped with emotion recognition sensors (cameras and microphones) to detect the user's emotional state in real time.

[0819] 2. Server:

[0820] The server stores and manages user settings, a news article database, a generative AI model, emotion recognition means, and a feedback database. The server has the following functions:

[0821] Loading the generative AI model: When the server starts up, it loads the generative AI model into memory to analyze article information. This generative AI model is trained based on machine learning algorithms and has the ability to detect spoilers.

[0822] Acquisition and analysis of article information: When a user attempts to view a news article, the server retrieves the relevant news article from the database and passes it to the generative AI model to analyze spoiler information.

[0823] Spoiler hiding: If the analysis reveals that a spoiler is included, the server replaces the relevant part with the placeholder "[Spoiler hidden]" and displays it to the user.

[0824] Use of emotional information: The server collects the user's emotional information sent from the device and adjusts the way article information is displayed based on this information.

[0825] User settings management: Users can set the spoiler categories and content they want to avoid according to their preferences, and the server stores and manages this setting information.

[0826] 3. Emotion recognition means:

[0827] The emotion recognition means analyzes the user's facial expressions and voice data to detect emotional information in real time. This emotional information is sent to the server and used to adjust the display method of article information.

[0828] Specific examples of the embodiment

[0829] For example, if a user wants to avoid spoilers for the movie "Movie X," they can register "Movie X" as a spoiler-protected content on the device's settings screen. The server receives and stores this setting information.

[0830] When a user attempts to view an article related to "Movie X," the server retrieves the article from the database, passes it to the generative AI model, and analyzes it for spoilers. If spoilers are found, the server replaces the relevant part with "[Spoiler Not Displayed]" and delivers it to the user.

[0831] At the same time, if the user's emotional state is surprise or displeasure, the server will adjust the way the article is displayed based on the emotional information, providing the user with a better browsing experience.

[0832] Examples of instructions and prompts

[0833] An example of a prompt sentence when using the sentiment analysis API is as follows:

[0834] python

[0835] Prompt sentence for using the sentiment analysis API

[0836] import requests

[0837] def analyze_sentiment(audio_file_path):

[0838] url = "http: / / sentiment-api.example.com / analyze"

[0839] files = {'file': open(audio_file_path, 'rb')}

[0840] response = requests.post(url, files=files)

[0841] return response.json()

[0842] Analyze audio files to recognize emotions

[0843] emotion_result = analyze_sentiment("path / to / user_audio.wav")

[0844] print(emotion_result)

[0845] This prompt shows how to send an audio file to the sentiment analysis API and receive the sentiment result.

[0846] This allows the user to avoid spoiler information and enjoy optimal information display suited to their own emotional state.

[0847] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0848] Step 1:

[0849] The user sets up spoiler prevention on the device. The user opens the device's settings screen, selects the specific content for which they want to avoid spoilers (e.g., "Movie X"), and enters the setting information. This setting information is sent by the device to the server. The input data consists of the user ID and setting information, which the server receives and stores in a database in association with the user ID.

[0850] Step 2:

[0851] The server loads the generative AI model. At startup, the server loads the generative AI model into memory and prepares it for analyzing article data. This model is trained based on a machine learning algorithm and is used to detect spoilers from news articles. The input data are the parameters of the generative AI model, and the output is the state of the model loaded in memory.

[0852] Step 3:

[0853] A user requests to view an article. When a user uses a device to view a specific news article, the device sends the request to the server. This request includes the article ID. The server receives the user's request and the article ID as input data.

[0854] Step 4:

[0855] The server retrieves and analyzes news articles. The server retrieves the headline and content of the news article corresponding to the specified article ID from the database. The retrieved article content is passed to a generative AI model, which analyzes whether it contains spoilers. The input data is the text data of the news article, and the output data is the spoiler detection results.

[0856] Step 5:

[0857] Concealment of spoiler information. If the generative AI model determines that an article contains spoiler information, it replaces the relevant part with "[Spoiler hidden]." The input data is the spoiler detection result and the text data of the article, and the output data is the text data of the article with the spoiler hidden.

[0858] Step 6:

[0859] Emotion detection using emotion recognition means. When a user uses a device to view an article, sensors (camera and microphone) installed on the device detect the user's emotional state in real time and send that data to the server. The input data is the user's facial expression and voice data, and the output data is the recognized emotional information.

[0860] Step 7:

[0861] Adjustment of article display based on emotional information. The server adjusts the display method of article information based on the user's emotional information obtained by the emotion recognition means. For example, if the user expresses emotions such as "surprise" or "discomfort," the server further adjusts the display content of the article and uses appropriate placeholders. The input data is the emotional information and the text data of the concealed article, and the output data is the adjusted article data that is finally displayed to the user.

[0862] Step 8:

[0863] Delivery of the adjusted article. The server finally sends the adjusted article to the device, where the user views it. The input data is the adjusted article data, and the output data is the article displayed on the user's device. This allows the user to avoid spoilers and experience information displayed optimally according to their emotional state.

[0864] In this way, each step works in cooperation with the others, allowing users to browse news articles and other content safely and comfortably.

[0865] 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.

[0866] 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.

[0867] 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.

[0868] [Third embodiment]

[0869] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0870] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0871] 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).

[0872] 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.

[0873] 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.

[0874] 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).

[0875] 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. 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.

[0876] 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.

[0877] 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.

[0878] 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.

[0879] 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.

[0880] 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."

[0881] The present invention relates to a system that analyzes article information, determines whether it contains spoiler information, and hides the spoiler information as necessary, and a specific embodiment thereof will be described below.

[0882] System configuration

[0883] This system consists of a user's device, a server, and a generative artificial intelligence (AI) model. The user uses the device to view news articles and configure spoiler prevention settings. The server receives this configuration information and retrieves and analyzes news articles. The generative AI model is hosted on the server and is used to analyze article information and determine spoiler information.

[0884] Program processing

[0885] 1. User configures spoiler protection:

[0886] The user selects the specific content for which they want to avoid spoilers (for example, "TV Drama A" or "Anime B") from the settings screen on their device, and sends the setting information from the device to the server. The server stores this setting information in association with the user ID.

[0887] 2. The server loads the generative AI model:

[0888] When the server starts up, it loads the generative AI model into memory and prepares it for analyzing article information. This model is trained by a machine learning algorithm and is used to accurately detect spoilers.

[0889] 3. Get news articles and check for spoilers:

[0890] When a user requests a specific news article to be displayed, the server retrieves the headline and content of the relevant news article from the database. The server then inputs the retrieved headline and content into the generative AI model and analyzes whether they contain spoilers. If they are determined to contain spoilers, the server replaces the relevant part with the placeholder "[Spoiler Not Displayed]".

[0891] 4. Customized Display:

[0892] The server reconstructs the article information based on the analysis results and user settings in a way that does not include spoilers, and sends this reconstructed content to the device, which then displays the article to the user in a secure format.

[0893] Specific examples

[0894] As a concrete example, consider the case where a user wants to avoid spoilers for "TV Drama A." When the user makes this setting on their device and registers "TV Drama A" as a spoiler prevention target, the server saves this information. Next, when the user attempts to view news article ID "1234," if the headline of this news article is "Review of the latest episode of 'TV Drama A'," the server uses an AI model to analyze the article content. If the AI ​​model determines that the article content contains spoilers, the server replaces the relevant part with "[Spoiler not displayed]" and sends it to the device. The device then displays the final content to the user.

[0895] Feedback and Learning

[0896] The system also has a function to improve the analysis accuracy of the generative AI model based on user feedback. When users provide feedback on the content of displayed articles, the server collects this information and uses it as retraining data for the AI ​​model. This allows the system's spoiler detection accuracy to improve over time, enabling more accurate filtering.

[0897] Thus, according to the embodiment of the present invention, the user can read news articles with peace of mind and enjoy the content without spoiling the fun.

[0898] The processing flow will be explained below.

[0899] Step 1:

[0900] The user sets up spoiler prevention on the device. The user specifies specific content (e.g., "TV Drama A," "Anime B," etc.) for which they want to avoid spoilers on the device's settings screen.

[0901] Step 2:

[0902] The device sends the setting information to the server, along with the user ID and the specified spoiler prevention target information.

[0903] Step 3:

[0904] The server updates the user settings. Based on the received information, the server saves the spoiler prevention settings associated with the user ID in the database.

[0905] Step 4:

[0906] The server loads the generative AI model. During initialization, the server loads the trained generative AI model into memory, preparing it for analyzing article information.

[0907] Step 5:

[0908] The user requests the display of a news article. The user sends a display request from the terminal to the server specifying a specific news article ID.

[0909] Step 6:

[0910] The device sends a request to the server, which sends the news article ID and user ID information to the server.

[0911] Step 7:

[0912] The server retrieves the news article. The server retrieves the headline and content of the requested article from the database.

[0913] Step 8:

[0914] The server analyzes the article information using the generative AI model. The server inputs the acquired headline and content into the generative AI model and analyzes whether it contains spoilers.

[0915] Step 9:

[0916] The server determines whether the information is a spoiler, and if the generative AI model determines that the information is a spoiler, it identifies the spoiler.

[0917] Step 10:

[0918] The server hides the spoiler information, replaces the spoiler part with the placeholder "[Spoiler hidden]", and modifies the original content.

[0919] Step 11:

[0920] The server sends the revised article information to the terminal. The server sends the revised headline and content to the terminal.

[0921] Step 12:

[0922] The terminal displays the corrected article information to the user. The terminal displays the received headline and the corrected content to the user.

[0923] Step 13:

[0924] When a user views an article and provides feedback, the user transmits their rating and opinion to the server via the terminal.

[0925] Step 14:

[0926] The server collects the feedback and stores it in a database.

[0927] Step 15:

[0928] The server retrains the generative AI model. The server uses the collected feedback data to retrain the generative AI model to improve its accuracy.

[0929] By following the steps above, users can read news articles with peace of mind and enjoy the content without spoiling it.

[0930] Example 1

[0931] 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."

[0932] In recent years, with the increasing number of news articles and reviews distributed over the Internet, users are at a greater risk of coming across spoilers for content they have not yet viewed. In particular, articles about entertainment content such as TV shows, movies, and anime can unintentionally contain spoilers, which can ruin the user's viewing experience. Conventional systems require users to take measures to avoid spoilers themselves, which is cumbersome and unreliable.

[0933] 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.

[0934] In this invention, the server includes a means for a user to use an electronic device to set spoiler prevention settings and transmit the setting information to the server, a means for the server to analyze article information using a generative artificial intelligence model and determine whether it is likely to contain spoiler information, and a means for hiding the spoiler information from the user based on the determination result. This reduces the risk that users will unintentionally come across spoiler information about content they have not viewed, allowing them to read news articles with peace of mind.

[0935] A "user" is a person who uses the system to set up spoiler protection for news articles and view the content.

[0936] "Electronic device" refers to a device used by a user to access and configure the system, including smartphones, PCs, tablets, etc.

[0937] "Server" refers to the central control unit that receives user preferences and retrieves, analyzes, and reconstructs news articles.

[0938] A "generative artificial intelligence model" is a model trained by a machine learning algorithm to analyze text in news articles to detect spoilers.

[0939] "Spoiler information" refers to information that includes important information or developments about content that the user has not yet viewed, and that may spoil the enjoyment of that content.

[0940] "Settings Information" refers to information about specific content or categories related to spoiler prevention that a user enters through the system.

[0941] "Feedback" refers to user-provided ratings and comments on the displayed article content, which helps improve the system's analysis accuracy.

[0942] A "placeholder" is an alternative expression such as "[Spoiler not shown]" that is used to replace a portion of text that contains spoiler information.

[0943] "Filtering" refers to the process of excluding certain types or content based on user settings.

[0944] A "prompt sentence" is an input sentence given to a generative artificial intelligence model that instructs the model on a specific analysis task.

[0945] The present invention is a system that analyzes news articles, determines whether they contain spoilers, and hides the spoilers as necessary. A specific embodiment of the system is described below. The system consists of a user's electronic device, a server, and a generative artificial intelligence model (generative AI model).

[0946] System configuration

[0947] The system includes the following hardware and software:

[0948] 1. User's electronic devices:

[0949] Users access the system using electronic devices such as smartphones, PCs, and tablets.

[0950] Set spoiler protection and read the news article.

[0951] 2. Server:

[0952] The server receives and stores user-submitted configuration information, and also retrieves, analyzes, and reconstructs news articles.

[0953] The server analyzes article information using a generative artificial intelligence model (e.g., BERT or GPT-3).

[0954] 3. Generative AI Models:

[0955] The generative artificial intelligence model analyzes the headlines and content of news articles to determine whether they contain spoilers.

[0956] Specific machine learning libraries used include TensorFlow and PyTorch.

[0957] Program processing

[0958] The system operates in the following steps:

[0959] 1. User settings input:

[0960] The user selects the content for which they want to avoid spoilers (e.g., "TV Drama A") from the interface on their electronic device.

[0961] Check the settings and press the "Save" button to send the settings to the server.

[0962] 2. Server generation AI model load:

[0963] When the server starts up, it loads the generative artificial intelligence model into memory and prepares it for text analysis.

[0964] 3. Get news articles:

[0965] When a user requests that a particular news article be displayed, the server retrieves the appropriate news article from the database.

[0966] 4. Spoiler Check:

[0967] The server inputs the headlines and content of the retrieved news articles into the generative AI model.

[0968] The generative AI model uses prompts to analyze articles and detect spoilers.

[0969] For example, enter the following prompt:

[0970] "If this news article contains spoilers for 'TV Drama A', please detect the relevant parts."

[0971] 5. Customized article generation:

[0972] Based on the analysis results, the server replaces the spoiler information with "[Spoiler not shown]" and reconstructs it.

[0973] 6. View Article:

[0974] The server transmits the reconstructed article to the user's electronic device, where the user can view the article in a secure format.

[0975] Specific examples

[0976] A specific example of operation is shown below.

[0977] example:

[0978] The user configures spoiler protection for "TV Drama A" and sends this information from their device to the server. When the user then requests the display of news article ID "1234," the server retrieves the relevant article from the database. The server then analyzes the article content using a generative AI model and replaces the spoiler portion with "[Spoiler Not Displayed]." Finally, the server sends this reconstructed article to the user's electronic device and displays it to the user.

[0979] The embodiments of the present invention allow users to browse news articles safely while avoiding spoilers. Furthermore, the accuracy of analysis is continually improved based on user feedback, further increasing the usefulness of the system.

[0980] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0981] Program processing flow

[0982] Step 1: Enter user settings

[0983] Input: The user enters the content they want to avoid spoilers for in the device's settings screen.

[0984] Specific behavior:

[0985] Users access an interface on an electronic device such as a smartphone or computer.

[0986] Open the settings screen and select content such as "TV Drama A."

[0987] Click the "Save" button to confirm the settings.

[0988] Output: The configuration information is sent to the server.

[0989] Data processing and calculation:

[0990] The terminal collects the user's setting information and sends it to the server as an HTTP request.

[0991] The setting information includes a user ID and a list of content for which spoilers are to be avoided.

[0992] Step 2: Server generation AI model load

[0993] Input: The server starts or receives a user request.

[0994] Specific behavior:

[0995] The server loads a generative artificial intelligence model (e.g., BERT or GPT-3) into memory when the system starts up.

[0996] Update the model cache as needed.

[0997] Output: The generative AI model is loaded into memory and ready to use.

[0998] Data processing and calculation:

[0999] Use a machine learning library (e.g., TensorFlow or PyTorch) to load the model.

[1000] The model is initialized and preprocessed.

[1001] Step 3: Get news articles

[1002] Input: A user requests that a specific news article be displayed.

[1003] Specific behavior:

[1004] A user sends a request to display a news article from an electronic device.

[1005] The server runs an SQL query against the database to retrieve the headlines and content of the relevant news articles.

[1006] Output: News article headline and content data.

[1007] Data processing and calculation:

[1008] Execute an SQL query to retrieve news article information from the database.

[1009] The acquired data is structured and stored in memory.

[1010] Step 4: Spoiler check

[1011] Input: News article headline and content, user preferences.

[1012] Specific behavior:

[1013] The server inputs the retrieved news articles and user settings into the generative AI model.

[1014] As a concrete example, the following prompt sentence is input to the generative AI model:

[1015] "If this news article contains spoilers for 'TV Drama A', please detect the relevant parts."

[1016] The generative AI model analyzes article information and detects whether it contains spoilers.

[1017] Output: Index information of the parts containing spoilers.

[1018] Data processing and calculation:

[1019] Generate a prompt sentence and input it into the model.

[1020] The model analyzes the text and detects the relevant parts (spoiler information).

[1021] Step 5: Generate a customized article

[1022] Input: spoiler check results (index information) and original news article.

[1023] Specific behavior:

[1024] Based on the analysis results, the server replaces the spoiler information with the placeholder "[Spoiler not shown]".

[1025] The entire reconstructed article is temporarily saved.

[1026] Output: A reconstructed news article.

[1027] Data processing and calculation:

[1028] Perform text processing and replace spoilers with placeholders.

[1029] Generate reconstructed article data.

[1030] Step 6: Viewing articles

[1031] Input: A reconstructed news article.

[1032] Specific behavior:

[1033] The server transmits the reconstructed news article to the user's electronic device.

[1034] The terminal renders and displays the news article data received from the server.

[1035] Output: The news article displayed to the user in a safe format.

[1036] Data processing and calculation:

[1037] Generate an HTTP response containing the reconstructed article data.

[1038] The device converts the received data into a format that can be displayed in a browser or application.

[1039] (Application example 1)

[1040] 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."

[1041] There is a growing need to avoid spoilers when viewing news articles and other content. However, existing systems lack sufficient customization for spoiler prevention, making it difficult for users to safely view content. In addition, there is a lack of systems that can train AI models that reflect user feedback.

[1042] 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.

[1043] In this invention, the server, as an application installed on a smartphone, includes means for retrieving news articles from an API and replacing parts containing spoiler information with "[Spoiler Not Displayed]" before displaying them; means for customizing the scope of spoiler information by filtering specific categories and content based on user settings; and means for performing learning to improve the analytical accuracy of the generative artificial intelligence model based on user feedback. This allows users to safely view articles and content they want to view and reduces the risk of spoilers. A "generative artificial intelligence model" is an AI model trained based on machine learning algorithms that is used to analyze news articles and content and determine spoiler information.

[1044] An "electronic device" is a device such as a smartphone, tablet, or computer that a user uses to view news articles.

[1045] An "API" is an application program interface for obtaining user-specified news article information from an external database.

[1046] "Spoiler information" refers to information about the ending or important scenes of a television drama, movie, anime, etc., and is information that may spoil your enjoyment of the content if you know it in advance.

[1047] A "placeholder" is a replacement string or symbol used to hide parts of a page that contain spoilers. A typical example is "[Spoiler Hide]".

[1048] "User settings" refers to customization that the user can make to avoid spoilers about specific content, and allows the user to specify specific categories and content.

[1049] "Feedback" refers to opinions and evaluations provided by users about the content of displayed articles, and is used as data to improve the analysis accuracy of the AI ​​model.

[1050] This invention relates to a system that detects spoilers in news articles and other content and allows users to avoid them. The system consists of an electronic device such as a smartphone, a server, and a generative artificial intelligence model.

[1051] Program processing

[1052] 1. The server receives a request to view a news article from a user's device (such as a smartphone). The user then uses an application installed on the smartphone to view the news article.

[1053] 2. The server retrieves the news articles using an API, which is an application program interface for retrieving article information from an external database.

[1054] 3. To analyze the article information, the server uses a generative artificial intelligence model, which is trained based on a machine learning algorithm and can accurately determine spoilers contained in the article information.

[1055] 4. The server filters spoilers for specific categories or content based on the user's settings, which the user previously entered in the application's settings screen.

[1056] 5. The server analyzes the article information and replaces any detected spoiler information with a placeholder (e.g., "[Spoiler Not Included]"). This process allows users to view the news article in a safe, spoiler-free format.

[1057] 6. When users provide feedback on the displayed article content, the server collects this feedback, which is used as retraining data for the generative AI model to improve its analysis accuracy.

[1058] Hardware and software used

[1059] Hardware: Smartphone (user device), server (analysis and filtering of article information)

[1060] Software: Python, Requests, Transformers (Hugging Face library)

[1061] Specific examples

[1062] As a concrete example, consider the case where a user wants to avoid spoilers for "Drama A." The user registers "Drama A" as a target for spoiler prevention on the settings screen of an application installed on their smartphone. When the user requests the display of news article ID "1234," the server retrieves the corresponding news article through the API. The content of this news article is analyzed by a generative artificial intelligence model, and if spoiler information is detected, the relevant part is replaced with "[Spoiler not shown]." Finally, the server reconstructs the article information in a secure format and sends it to the device to display to the user.

[1063] Example prompt sentence:

[1064] "Analyze whether an article about the latest episode of Drama A contains spoilers about the ending or important scenes."

[1065] As described above, the embodiment of the present invention allows users to read news articles with peace of mind and prevents spoiler information from ruining the enjoyment of the article.

[1066] The flow of the specific processing in Application Example 1 will be explained with reference to FIG. 12. Step 1:

[1067] Users can set spoiler prevention settings

[1068] On the settings screen of the application installed on the smartphone, the user selects the specific content or category (e.g., Drama A) for which they want to avoid spoilers, and saves the settings. The input is the setting information set by the user, and the output is the setting information sent to the server. This setting information is stored on the server in association with the user ID. Specifically, the user enters keywords for the category they selected (e.g., ending, important scene), and presses the save button.

[1069] Step 2:

[1070] The server receives a request to display a news article.

[1071] The user sends a request to display an article from their smartphone. The input is the ID or URL of the news article specified by the user, and the output is the request being received by the server. Specifically, the user selects a specific article from the list of articles in the application and presses the display button to send the request.

[1072] Step 3:

[1073] The server retrieves news articles using an API

[1074] The server calls an API to retrieve news articles from an external database. The input is the article ID and URL, and the output is the retrieved article information. Specifically, the server sends a request to the API endpoint and parses the returned JSON-formatted article data.

[1075] Step 4:

[1076] The server analyzes the article information using a generated artificial intelligence model

[1077] The server uses a generative artificial intelligence model to analyze whether the acquired article information contains spoilers. The input is the article text, and the output is the result of determining whether spoilers are included. Specifically, the article text is input into the model, and the generated output (whether spoilers are present or not) is obtained. This model is implemented using Hugging Face's Transformers library.

[1078] Step 5:

[1079] The server filters spoilers based on configuration information.

[1080] The server filters article content based on a spoiler prevention keyword list set by the user. The input is the article text and user setting information, and the output is the article text with spoiler information filtered out. Specifically, it detects set keywords (e.g., ending, important scene) from the article text and replaces them with placeholders (e.g., [Spoiler not shown]).

[1081] Step 6:

[1082] The server sends the article to the device after processing to prevent spoilers.

[1083] The server sends the article text, from which spoilers have been filtered, to the user's device. The input is the filtered article text, and the output is safe article information that is displayed on the user's device. Specifically, the server sends the filtered article data to the device as an HTTP response, and the device receives it and displays it on the screen.

[1084] Step 7:

[1085] Collect user feedback and incorporate it into the AI ​​model's training data

[1086] The user provides feedback on the content of the displayed article. This feedback is collected by the server and used as retraining data for the generative AI model. The input is the user's feedback information, and the output is a database in which the feedback is saved. Specifically, the user enters their opinion in the feedback form and presses the submit button, and the server saves the information.

[1087] The above are the specific processing steps in this article filtering system.

[1088] 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.

[1089] The present invention relates to a system that analyzes article information, determines whether it contains spoilers, and hides the spoilers as necessary, and further improves the user experience by combining it with an emotion engine that recognizes the user's emotions. Specific embodiments of the system are described below.

[1090] System configuration

[1091] This system consists of a user's device, a server, a generative artificial intelligence (AI) model, and an emotion engine. The user uses the device to view news articles and configure spoiler prevention settings. The server receives this configuration information and retrieves and analyzes news articles. The generative AI model is hosted on the server and is used to analyze article information and determine spoiler information. The emotion engine detects the user's emotions and provides this information to the server.

[1092] Program processing

[1093] 1. User configures spoiler protection:

[1094] The user selects the specific content for which they want to avoid spoilers (for example, "TV Drama A" or "Anime B") from the settings screen on their device, and sends the setting information from the device to the server. The server stores this setting information in association with the user ID.

[1095] 2. The server loads the generative AI model:

[1096] When the server starts up, it loads the generative AI model into memory and prepares it for analyzing article information. This model is trained by a machine learning algorithm and is used to accurately detect spoilers.

[1097] 3. Get news articles and check for spoilers:

[1098] When a user requests a specific news article to be displayed, the server retrieves the headline and content of the relevant news article from the database. The server then inputs the retrieved headline and content into the generative AI model and analyzes whether they contain spoilers. If they are determined to contain spoilers, the server replaces the relevant part with the placeholder "[Spoiler Not Displayed]".

[1099] 4. Emotion Recognition with Emotion Engine:

[1100] When a user uses a device, the emotion engine recognizes the user's emotional state using sensors (such as a camera or microphone) installed on the device. The recognized emotional information is sent to the server in real time.

[1101] 5. Customized Display:

[1102] The server adjusts the display method of spoiler information based on the analysis results, user settings, and the user's emotional state. Based on these results, the server reconstructs the article information so that it does not contain spoilers. The reconstructed content is sent to the device, which then displays the article to the user in a secure format.

[1103] Specific examples

[1104] As a concrete example, consider the case where a user wants to avoid spoilers for "TV Drama A." When the user makes this setting on their device and registers "TV Drama A" as a spoiler prevention target, the server saves this information. Next, when the user attempts to view news article ID "1234," if the headline of this news article is "Review of the latest episode of 'TV Drama A'," the server will use an AI model to analyze the article content. If the AI ​​model determines that the article content contains spoilers, the server will replace the content with "[Spoiler not shown]." At the same time, if the user's emotional information while viewing the device is "surprise" or "discomfort," the server will further take this information into account and adjust the displayed content, reconstructing the article in a way that better takes the user's emotions into consideration.

[1105] Feedback and Learning

[1106] The system also has a function that improves the analytical accuracy of the generative AI model based on user feedback. When users provide feedback on the content of displayed articles, the server collects this information and uses it as retraining data for the AI ​​model. Feedback reflecting users' emotional information is also collected, and this is used to retrain the model for even more accurate filtering. This allows the system's spoiler detection accuracy to improve over time, enabling more accurate filtering.

[1107] In this way, the embodiments of the present invention not only allow users to read news articles with peace of mind, but also provide optimal information display according to their emotional state, allowing them to enjoy a more comfortable and personalized viewing experience.

[1108] The processing flow will be explained below.

[1109] Step 1:

[1110] The user sets up spoiler prevention on the device. The user specifies specific content (e.g., "TV Drama A," "Anime B," etc.) for which they want to avoid spoilers on the device's settings screen.

[1111] Step 2:

[1112] The device sends the setting information to the server, along with the user ID and the specified spoiler prevention target information.

[1113] Step 3:

[1114] The server updates the user settings. Based on the received information, the server saves the spoiler prevention settings associated with the user ID in the database.

[1115] Step 4:

[1116] The server loads the generative AI model. During initialization, the server loads the trained generative AI model into memory, preparing it for analyzing article information.

[1117] Step 5:

[1118] The user requests the display of a news article. The user sends a display request from the terminal to the server specifying a specific news article ID.

[1119] Step 6:

[1120] The device sends a request to the server, which sends the news article ID and user ID information to the server.

[1121] Step 7:

[1122] The server retrieves the news article. The server retrieves the headline and content of the requested article from the database.

[1123] Step 8:

[1124] The server analyzes the article information using the generative AI model. The server inputs the acquired headline and content into the generative AI model and analyzes whether it contains spoilers.

[1125] Step 9:

[1126] The server determines whether the information is a spoiler, and if the generative AI model determines that the information is a spoiler, it identifies the spoiler.

[1127] Step 10:

[1128] The server hides the spoiler information, replaces the spoiler part with the placeholder "[Spoiler hidden]", and modifies the original content.

[1129] Step 11:

[1130] The emotion engine recognizes the user's emotions while using the device. It uses sensors such as the device's camera and microphone to analyze the user's facial expressions and voice to recognize emotions.

[1131] Step 12:

[1132] The device transmits the recognized emotion information to the server. The device transmits the user's emotional state information recognized in real time to the server.

[1133] Step 13:

[1134] The server reconstructs the article information based on the user's emotional state. The server adjusts the displayed content taking into account the user's emotional state in addition to the analysis results and spoiler prevention settings.

[1135] Step 14:

[1136] The server sends the revised article information to the terminal. The server sends the revised headline and content to the terminal.

[1137] Step 15:

[1138] The terminal displays the corrected article information to the user. The terminal displays the received headline and the corrected content to the user.

[1139] Step 16:

[1140] When a user views an article and provides feedback, the user transmits their rating and opinion to the server via the terminal.

[1141] Step 17:

[1142] The server collects the feedback and stores it in a database.

[1143] Step 18:

[1144] The server retrains the generative AI model. The server uses the collected feedback data to retrain the generative AI model to improve its accuracy.

[1145] Through these steps, users can not only browse news articles with peace of mind, but also enjoy a more comfortable and personalized browsing experience by being provided with optimal information display according to their emotional state.

[1146] Example 2

[1147] 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."

[1148] Conventional news article browsing systems often display articles containing unwanted spoilers, which detracts from the user's browsing experience. Furthermore, because they only implement a uniform approach to preventing spoilers without taking into account the user's emotional state, they are unable to provide optimal information to each individual user. Furthermore, there is a lack of effort to improve the accuracy of generative AI models based on user feedback, limiting the accuracy of spoiler detection.

[1149] 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 receiving spoiler prevention settings from a user terminal and registering specific categories or content as spoiler prevention targets; means for analyzing article information on the server using a generative AI model and determining whether the article information may contain spoiler information; means for hiding spoiler information from the user based on the determination result; and means for customizing the display method of spoiler information based on user settings and emotion information generated by an emotion engine. This allows users to avoid unwanted spoiler information and provides optimal information according to their individual emotional state. Furthermore, by improving the accuracy of the generative AI model based on feedback, more accurate spoiler prevention can be achieved.

[1150] A "user terminal" is an electronic device that is operated by a user to view information and change settings.

[1151] The "spoiler prevention setting" is information that a user sets to avoid spoiler information for a specific content or category.

[1152] A "generative AI model" is a model trained using machine learning algorithms to analyze article information and detect spoilers.

[1153] A "server" is a device that provides services over a network and receives and processes requests from user terminals.

[1154] "Spoiler information" is information that includes important details or the ending of content that the user has not yet seen.

[1155] An "emotion engine" is a software or hardware mechanism for recognizing and analyzing a user's emotional state.

[1156] A "placeholder" is alternative text or symbols that are displayed in place of spoiler information.

[1157] "Feedback" refers to information such as usage experience, impressions, and evaluations provided by users, and is used to improve the system.

[1158] "Analysis accuracy" refers to the degree to which the generative AI model can accurately detect spoiler information.

[1159] "News Article" means article content provided via the Internet or other media.

[1160] "Customization" means changing the system's behavior and display methods according to the user's settings and status.

[1161] The present invention relates to a system that analyzes news articles, determines whether they contain spoilers, and hides the spoilers as necessary. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system further improves the user experience. Specific embodiments of the system are described below.

[1162] System configuration

[1163] This system consists of a user device, a server, a generative AI model, and an emotion engine. The user uses the device to view news articles and configure spoiler prevention settings. The server receives this configuration information and retrieves and analyzes news articles. The generative AI model is hosted on the server and is used to analyze news articles and determine spoiler information. The emotion engine detects the user's emotions and provides this information to the server.

[1164] Program processing

[1165] User-defined spoiler prevention settings

[1166] The user selects the specific content for which they want to avoid spoilers, such as "TV dramas" or "movies," on the device's settings screen. The device then sends the selected setting information to the server, which then associates the received setting information with the user ID and stores it in a database.

[1167] Loading of generated AI models by the server

[1168] At startup, the server loads into memory a generative AI model that is trained using machine learning algorithms and used to analyze news articles and identify spoilers.

[1169] Get news articles and check for spoilers

[1170] When a user requests a specific news article to be displayed, the server retrieves the relevant news article from the database. The headline and content of the retrieved article are input into the generative AI model to determine whether it contains spoilers. If the model detects any part that it deems a spoiler, the server replaces that part with "[Spoiler Not Displayed]."

[1171] Emotion recognition by emotion engine

[1172] While a user is browsing a news article on their device, the emotion engine uses the device's built-in camera and microphone to recognize the user's emotional state, and the recognized emotional information is sent to the server in real time.

[1173] Customized View

[1174] The server adjusts the display method of the spoiler information based on the analysis results, user settings, and the user's emotional information. The reconstructed article content is sent to the device, which then displays the news article to the user in a secure format.

[1175] Specific examples

[1176] As a concrete example, consider the case where a user wants to avoid spoilers for a "TV drama." When the user makes this setting on their device and registers "TV drama" as a spoiler prevention target, the server saves this information. Next, when the user attempts to view news article ID "1234," if the headline of this news article is "Review of the latest episode of 'TV drama'," the server will analyze the article content using the generative AI model. If the generative AI model determines that the article content contains spoilers, the server will replace the content with "[Spoiler not displayed]."

[1177] At the same time, if the emotional information displayed by the user while browsing the device is "surprise" or "discomfort," the server will further take this information into account and adjust the displayed content, reconstructing the article in a way that takes the user's emotions into greater consideration.

[1178] Prompt Sentence Examples

[1179] Examples of input prompts for a generative AI model include:

[1180] "Please analyze what spoilers this news article contains."

[1181] "The user has specified that they would like to avoid spoilers for 'TV drama'. Analyze the content of this news article and replace any spoilers with 'Spoiler Hide'."

[1182] Using these prompts, we can see how the generative AI model parses news articles and appropriately detects and filters spoilers.

[1183] This allows users to avoid unwanted spoilers and obtain optimal information according to their emotional state. Furthermore, by improving the accuracy of the generative AI model based on feedback, even more accurate spoiler prevention can be achieved.

[1184] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1185] Step 1:

[1186] User enters and submits spoiler protection settings

[1187] The user opens the settings screen on the device and selects the specific category or content (e.g., "TV dramas" or "movies") for which they want to avoid spoilers. Using the selected information as input, the device sends the setting information to the server. Specifically, the user taps "Add spoiler-protected content," selects "TV dramas," and presses the "Save" button. The device then sends a request to the server stating, "User ID: 123 has set his preference to avoid spoilers for TV dramas."

[1188] Input: Specific categories or content you want to avoid spoilers for

[1189] Output: Sending configuration information to the server

[1190] Step 2:

[1191] Server-based storage of configuration information

[1192] The server receives the setting information sent from the device, associates it with the user ID, and stores it in the database. Specifically, the server analyzes the data it receives and adds the information "User ID: 123" and "Spoiler prevention target: TV drama" to the user information table in the database.

[1193] Input: Setting information sent from the device

[1194] Output: User preferences stored in the database

[1195] Step 3:

[1196] Loading a generative AI model

[1197] When the server starts up, it loads the generative AI model into memory. This generative AI model is trained using a machine learning algorithm and is used to analyze news articles and determine spoiler information. Specifically, when the server is restarted, the generative AI model loading process is automatically executed and the model is loaded into memory. The system log records "The generative AI model has been loaded."

[1198] Input: Start the server

[1199] Output: The loaded generative AI model

[1200] Step 4:

[1201] Processing a request to retrieve news articles

[1202] A user sends a specific news article ID as input from their device, requesting the server to display that article. The server receives this request, searches for the requested news article in its internal database, and retrieves the headline and content of the corresponding news article. Specifically, when a user taps to display "News Article ID: 1234" on their smartphone app, the device sends a request to the server saying, "I want to display article ID 1234." The server retrieves the article from the database and returns the headline and body data, such as "Review of the latest TV drama episode."

[1203] Input: News article ID

[1204] Output: News article headline and content

[1205] Step 5:

[1206] Analysis of spoiler information in article content

[1207] The server inputs the headline and content of the retrieved news article into the generative AI model and analyzes whether or not it contains spoilers. When the model detects any parts that it determines contain spoilers, the server replaces those parts with "[Spoiler not shown]". Specifically, the server sends the generative AI model a prompt message saying, "Analyze the content of article ID: 1234 and detect any parts that contain spoilers." When the generative AI model responds that the parts contain spoilers, it replaces them with "[Spoiler not shown]".

[1208] Input: News article headline and content

[1209] Output: News article with spoilers replaced

[1210] Step 6:

[1211] Emotion recognition by emotion engine

[1212] While a user is viewing a news article on their device, the emotion engine uses the device's built-in camera and microphone to recognize the user's emotional state. The recognized emotional information is sent to the server in real time. Specifically, while the user is viewing a news article, the device's camera captures the user's facial expressions, and the emotion engine detects the "surprise" expression. This information is sent to the server as "User ID: 123 is in a surprised emotional state."

[1213] Input: Emotion information from the device

[1214] Output: Emotion information sent to the server

[1215] Step 7:

[1216] Customized article display

[1217] The server adjusts the display method of spoiler information based on the analysis results, user settings, and the user's emotional information. The reconstructed article content is sent to the device, and the device displays the news article to the user in a safe format. Specifically, when the server receives that the user is in the "surprise" emotional information state, it applies further filtering to avoid spoilers. The reconstructed article content is sent to the device and displayed as a "safe article."

[1218] Input: Analysis results, user settings, emotional information

[1219] Output: Reconstructed news article

[1220] (Application example 2)

[1221] 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."

[1222] In conventional news article and content distribution services, users have difficulty avoiding unwanted spoilers. Furthermore, the optimal display of information based on the user's emotional state can sometimes impair the browsing experience. Therefore, there is a need for a system that allows users to browse content with peace of mind and that displays information appropriate to their emotional state.

[1223] 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 analyzing article information of the electronic device using a generative artificial intelligence model and determining whether it is likely to contain spoiler information, means for preventing spoilers by replacing part of the article containing spoiler information with a placeholder, emotion recognition means for detecting the emotional state of the user, and means for adjusting the display method of the article information based on the detected emotion information. This allows the user to avoid spoiler information and also makes it possible to display information optimally according to the user's emotional state.

[1224] A "generative artificial intelligence model" is a model that is trained based on machine learning algorithms to analyze text data and determine specific information.

[1225] "Electronic equipment" is a general term for devices used to process and display digital data, including smartphones, tablets, and personal computers.

[1226] "Spoiler information" is information that reveals important developments or endings of content such as movies, dramas, anime, and books in advance.

[1227] A "means for determining" is a method or device for utilizing a generative AI model to analyze text data and identify whether it contains specific information.

[1228] A "placeholder" is a substitute display string that is temporarily used to hide the original information.

[1229] "Emotion recognition means" refers to a method or device for detecting and analyzing emotions from a user's facial expressions, voice, actions, etc.

[1230] "User settings" refers to setting information that allows a user to customize the operating conditions of the system and the display contents based on their own preferences.

[1231] "Customizable means" means a method or device for changing or adjusting the system's functionality or display content based on user settings.

[1232] "Emotion information" is data indicating the emotional state of the user detected by the emotion recognition means.

[1233] The "means for adjusting the display method of article information" is a method or device for changing the content or format of information displayed to a user based on emotion information.

[1234] The system for implementing this invention is composed of a user terminal, a server, a generative AI model, and emotion recognition means. This allows users to browse news articles and other content with peace of mind, and displays information appropriate to their emotional state.

[1235] System configuration

[1236] 1. On the user's device:

[1237] Users browse news articles and other content using electronic devices such as smartphones, tablets, smart glasses, and head-mounted displays. These devices are equipped with emotion recognition sensors (cameras and microphones) to detect the user's emotional state in real time.

[1238] 2. Server:

[1239] The server stores and manages user settings, a news article database, a generative AI model, emotion recognition means, and a feedback database. The server has the following functions:

[1240] Loading the generative AI model: When the server starts up, it loads the generative AI model into memory to analyze article information. This generative AI model is trained based on machine learning algorithms and has the ability to detect spoilers.

[1241] Acquisition and analysis of article information: When a user attempts to view a news article, the server retrieves the relevant news article from the database and passes it to the generative AI model to analyze spoiler information.

[1242] Spoiler hiding: If the analysis reveals that a spoiler is included, the server replaces the relevant part with the placeholder "[Spoiler hidden]" and displays it to the user.

[1243] Use of emotional information: The server collects the user's emotional information sent from the device and adjusts the way article information is displayed based on this information.

[1244] User settings management: Users can set the spoiler categories and content they want to avoid according to their preferences, and the server stores and manages this setting information.

[1245] 3. Emotion recognition means:

[1246] The emotion recognition means analyzes the user's facial expressions and voice data to detect emotional information in real time. This emotional information is sent to the server and used to adjust the display method of article information.

[1247] Specific examples of the embodiment

[1248] For example, if a user wants to avoid spoilers for the movie "Movie X," they can register "Movie X" as a spoiler-protected content on the device's settings screen. The server receives and stores this setting information.

[1249] When a user attempts to view an article related to "Movie X," the server retrieves the article from the database, passes it to the generative AI model, and analyzes it for spoilers. If spoilers are found, the server replaces the relevant part with "[Spoiler Not Displayed]" and delivers it to the user.

[1250] At the same time, if the user's emotional state is surprise or displeasure, the server will adjust the way the article is displayed based on the emotional information, providing the user with a better browsing experience.

[1251] Examples of instructions and prompts

[1252] An example of a prompt sentence when using the sentiment analysis API is as follows:

[1253] python

[1254] Prompt sentence for using the sentiment analysis API

[1255] import requests

[1256] def analyze_sentiment(audio_file_path):

[1257] url = "http: / / sentiment-api.example.com / analyze"

[1258] files = {'file': open(audio_file_path, 'rb')}

[1259] response = requests.post(url, files=files)

[1260] return response.json()

[1261] Analyze audio files to recognize emotions

[1262] emotion_result = analyze_sentiment("path / to / user_audio.wav")

[1263] print(emotion_result)

[1264] This prompt shows how to send an audio file to the sentiment analysis API and receive the sentiment result.

[1265] This allows the user to avoid spoiler information and enjoy optimal information display suited to their own emotional state.

[1266] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1267] Step 1:

[1268] The user sets up spoiler prevention on the device. The user opens the device's settings screen, selects the specific content for which they want to avoid spoilers (e.g., "Movie X"), and enters the setting information. This setting information is sent by the device to the server. The input data consists of the user ID and setting information, which the server receives and stores in a database in association with the user ID.

[1269] Step 2:

[1270] The server loads the generative AI model. At startup, the server loads the generative AI model into memory and prepares it for analyzing article data. This model is trained based on a machine learning algorithm and is used to detect spoilers from news articles. The input data are the parameters of the generative AI model, and the output is the state of the model loaded in memory.

[1271] Step 3:

[1272] A user requests to view an article. When a user uses a device to view a specific news article, the device sends the request to the server. This request includes the article ID. The server receives the user's request and the article ID as input data.

[1273] Step 4:

[1274] The server retrieves and analyzes news articles. The server retrieves the headline and content of the news article corresponding to the specified article ID from the database. The retrieved article content is passed to a generative AI model, which analyzes whether it contains spoilers. The input data is the text data of the news article, and the output data is the spoiler detection results.

[1275] Step 5:

[1276] Concealment of spoiler information. If the generative AI model determines that an article contains spoiler information, it replaces the relevant part with "[Spoiler hidden]." The input data is the spoiler detection result and the text data of the article, and the output data is the text data of the article with the spoiler hidden.

[1277] Step 6:

[1278] Emotion detection using emotion recognition means. When a user uses a device to view an article, sensors (camera and microphone) installed on the device detect the user's emotional state in real time and send that data to the server. The input data is the user's facial expression and voice data, and the output data is the recognized emotional information.

[1279] Step 7:

[1280] Adjustment of article display based on emotional information. The server adjusts the display method of article information based on the user's emotional information obtained by the emotion recognition means. For example, if the user expresses emotions such as "surprise" or "discomfort," the server further adjusts the display content of the article and uses appropriate placeholders. The input data is the emotional information and the text data of the concealed article, and the output data is the adjusted article data that is finally displayed to the user.

[1281] Step 8:

[1282] Delivery of the adjusted article. The server finally sends the adjusted article to the device, where the user views it. The input data is the adjusted article data, and the output data is the article displayed on the user's device. This allows the user to avoid spoilers and experience information displayed optimally according to their emotional state.

[1283] In this way, each step works in cooperation with the others, allowing users to browse news articles and other content safely and comfortably.

[1284] 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.

[1285] 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.

[1286] 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.

[1287] [Fourth embodiment]

[1288] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1289] 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.

[1290] 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).

[1291] 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.

[1292] 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.

[1293] 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).

[1294] 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. 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.

[1295] 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.

[1296] 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.

[1297] 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.

[1298] 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.

[1299] 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.

[1300] 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."

[1301] The present invention relates to a system that analyzes article information, determines whether it contains spoiler information, and hides the spoiler information as necessary, and a specific embodiment thereof will be described below.

[1302] System configuration

[1303] This system consists of a user's device, a server, and a generative artificial intelligence (AI) model. The user uses the device to view news articles and configure spoiler prevention settings. The server receives this configuration information and retrieves and analyzes news articles. The generative AI model is hosted on the server and is used to analyze article information and determine spoiler information.

[1304] Program processing

[1305] 1. User configures spoiler protection:

[1306] The user selects the specific content for which they want to avoid spoilers (for example, "TV Drama A" or "Anime B") from the settings screen on their device, and sends the setting information from the device to the server. The server stores this setting information in association with the user ID.

[1307] 2. The server loads the generative AI model:

[1308] When the server starts up, it loads the generative AI model into memory and prepares it for analyzing article information. This model is trained by a machine learning algorithm and is used to accurately detect spoilers.

[1309] 3. Get news articles and check for spoilers:

[1310] When a user requests a specific news article to be displayed, the server retrieves the headline and content of the relevant news article from the database. The server then inputs the retrieved headline and content into the generative AI model and analyzes whether they contain spoilers. If they are determined to contain spoilers, the server replaces the relevant part with the placeholder "[Spoiler Not Displayed]".

[1311] 4. Customized Display:

[1312] The server reconstructs the article information based on the analysis results and user settings in a way that does not include spoilers, and sends this reconstructed content to the device, which then displays the article to the user in a secure format.

[1313] Specific examples

[1314] As a concrete example, consider the case where a user wants to avoid spoilers for "TV Drama A." When the user makes this setting on their device and registers "TV Drama A" as a spoiler prevention target, the server saves this information. Next, when the user attempts to view news article ID "1234," if the headline of this news article is "Review of the latest episode of 'TV Drama A'," the server uses an AI model to analyze the article content. If the AI ​​model determines that the article content contains spoilers, the server replaces the relevant part with "[Spoiler not displayed]" and sends it to the device. The device then displays the final content to the user.

[1315] Feedback and Learning

[1316] The system also has a function to improve the analysis accuracy of the generative AI model based on user feedback. When users provide feedback on the content of displayed articles, the server collects this information and uses it as retraining data for the AI ​​model. This allows the system's spoiler detection accuracy to improve over time, enabling more accurate filtering.

[1317] Thus, according to the embodiment of the present invention, the user can read news articles with peace of mind and enjoy the content without spoiling the fun.

[1318] The processing flow will be explained below.

[1319] Step 1:

[1320] The user sets up spoiler prevention on the device. The user specifies specific content (e.g., "TV Drama A," "Anime B," etc.) for which they want to avoid spoilers on the device's settings screen.

[1321] Step 2:

[1322] The device sends the setting information to the server, along with the user ID and the specified spoiler prevention target information.

[1323] Step 3:

[1324] The server updates the user settings. Based on the received information, the server saves the spoiler prevention settings associated with the user ID in the database.

[1325] Step 4:

[1326] The server loads the generative AI model. During initialization, the server loads the trained generative AI model into memory, preparing it for analyzing article information.

[1327] Step 5:

[1328] The user requests the display of a news article. The user sends a display request from the terminal to the server specifying a specific news article ID.

[1329] Step 6:

[1330] The device sends a request to the server, which sends the news article ID and user ID information to the server.

[1331] Step 7:

[1332] The server retrieves the news article. The server retrieves the headline and content of the requested article from the database.

[1333] Step 8:

[1334] The server analyzes the article information using the generative AI model. The server inputs the acquired headline and content into the generative AI model and analyzes whether it contains spoilers.

[1335] Step 9:

[1336] The server determines whether the information is a spoiler, and if the generative AI model determines that the information is a spoiler, it identifies the spoiler.

[1337] Step 10:

[1338] The server hides the spoiler information, replaces the spoiler part with the placeholder "[Spoiler hidden]", and modifies the original content.

[1339] Step 11:

[1340] The server sends the revised article information to the terminal. The server sends the revised headline and content to the terminal.

[1341] Step 12:

[1342] The terminal displays the corrected article information to the user. The terminal displays the received headline and the corrected content to the user.

[1343] Step 13:

[1344] When a user views an article and provides feedback, the user transmits their rating and opinion to the server via the terminal.

[1345] Step 14:

[1346] The server collects the feedback and stores it in a database.

[1347] Step 15:

[1348] The server retrains the generative AI model. The server uses the collected feedback data to retrain the generative AI model to improve its accuracy.

[1349] By following the steps above, users can read news articles with peace of mind and enjoy the content without spoiling it.

[1350] Example 1

[1351] 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."

[1352] In recent years, with the increasing number of news articles and reviews distributed over the Internet, users are at a greater risk of coming across spoilers for content they have not yet viewed. In particular, articles about entertainment content such as TV shows, movies, and anime can unintentionally contain spoilers, which can ruin the user's viewing experience. Conventional systems require users to take measures to avoid spoilers themselves, which is cumbersome and unreliable.

[1353] 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.

[1354] In this invention, the server includes a means for a user to use an electronic device to set spoiler prevention settings and transmit the setting information to the server, a means for the server to analyze article information using a generative artificial intelligence model and determine whether it is likely to contain spoiler information, and a means for hiding the spoiler information from the user based on the determination result. This reduces the risk that users will unintentionally come across spoiler information about content they have not viewed, allowing them to read news articles with peace of mind.

[1355] A "user" is a person who uses the system to set up spoiler protection for news articles and view the content.

[1356] "Electronic device" refers to a device used by a user to access and configure the system, including smartphones, PCs, tablets, etc.

[1357] "Server" refers to the central control unit that receives user preferences and retrieves, analyzes, and reconstructs news articles.

[1358] A "generative artificial intelligence model" is a model trained by a machine learning algorithm to analyze text in news articles to detect spoilers.

[1359] "Spoiler information" refers to information that includes important information or developments about content that the user has not yet viewed, and that may spoil the enjoyment of that content.

[1360] "Settings Information" refers to information about specific content or categories related to spoiler prevention that a user enters through the system.

[1361] "Feedback" refers to user-provided ratings and comments on the displayed article content, which helps improve the system's analysis accuracy.

[1362] A "placeholder" is an alternative expression such as "[Spoiler not shown]" that is used to replace a portion of text that contains spoiler information.

[1363] "Filtering" refers to the process of excluding certain types or content based on user settings.

[1364] A "prompt sentence" is an input sentence given to a generative artificial intelligence model that instructs the model on a specific analysis task.

[1365] The present invention is a system that analyzes news articles, determines whether they contain spoilers, and hides the spoilers as necessary. A specific embodiment of the system is described below. The system consists of a user's electronic device, a server, and a generative artificial intelligence model (generative AI model).

[1366] System configuration

[1367] The system includes the following hardware and software:

[1368] 1. User's electronic devices:

[1369] Users access the system using electronic devices such as smartphones, PCs, and tablets.

[1370] Set spoiler protection and read the news article.

[1371] 2. Server:

[1372] The server receives and stores user-submitted configuration information, and also retrieves, analyzes, and reconstructs news articles.

[1373] The server analyzes article information using a generative artificial intelligence model (e.g., BERT or GPT-3).

[1374] 3. Generative AI Models:

[1375] The generative artificial intelligence model analyzes the headlines and content of news articles to determine whether they contain spoilers.

[1376] Specific machine learning libraries used include TensorFlow and PyTorch.

[1377] Program processing

[1378] The system operates in the following steps:

[1379] 1. User settings input:

[1380] The user selects the content for which they want to avoid spoilers (e.g., "TV Drama A") from the interface on their electronic device.

[1381] Check the settings and press the "Save" button to send the settings to the server.

[1382] 2. Server generation AI model load:

[1383] When the server starts up, it loads the generative artificial intelligence model into memory and prepares it for text analysis.

[1384] 3. Get news articles:

[1385] When a user requests that a particular news article be displayed, the server retrieves the appropriate news article from the database.

[1386] 4. Spoiler Check:

[1387] The server inputs the headlines and content of the retrieved news articles into the generative AI model.

[1388] The generative AI model uses prompts to analyze articles and detect spoilers.

[1389] For example, enter the following prompt:

[1390] "If this news article contains spoilers for 'TV Drama A', please detect the relevant parts."

[1391] 5. Customized article generation:

[1392] Based on the analysis results, the server replaces the spoiler information with "[Spoiler not shown]" and reconstructs it.

[1393] 6. View Article:

[1394] The server transmits the reconstructed article to the user's electronic device, where the user can view the article in a secure format.

[1395] Specific examples

[1396] A specific example of operation is shown below.

[1397] example:

[1398] The user configures spoiler protection for "TV Drama A" and sends this information from their device to the server. When the user then requests the display of news article ID "1234," the server retrieves the relevant article from the database. The server then analyzes the article content using a generative AI model and replaces the spoiler portion with "[Spoiler Not Displayed]." Finally, the server sends this reconstructed article to the user's electronic device and displays it to the user.

[1399] The embodiments of the present invention allow users to browse news articles safely while avoiding spoilers. Furthermore, the accuracy of analysis is continually improved based on user feedback, further increasing the usefulness of the system.

[1400] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1401] Program processing flow

[1402] Step 1: Enter user settings

[1403] Input: The user enters the content they want to avoid spoilers for in the device's settings screen.

[1404] Specific behavior:

[1405] Users access an interface on an electronic device such as a smartphone or computer.

[1406] Open the settings screen and select content such as "TV Drama A."

[1407] Click the "Save" button to confirm the settings.

[1408] Output: The configuration information is sent to the server.

[1409] Data processing and calculation:

[1410] The terminal collects the user's setting information and sends it to the server as an HTTP request.

[1411] The setting information includes a user ID and a list of content for which spoilers are to be avoided.

[1412] Step 2: Server generation AI model load

[1413] Input: The server starts or receives a user request.

[1414] Specific behavior:

[1415] The server loads a generative artificial intelligence model (e.g., BERT or GPT-3) into memory when the system starts up.

[1416] Update the model cache as needed.

[1417] Output: The generative AI model is loaded into memory and ready to use.

[1418] Data processing and calculation:

[1419] Use a machine learning library (e.g., TensorFlow or PyTorch) to load the model.

[1420] The model is initialized and preprocessed.

[1421] Step 3: Get news articles

[1422] Input: A user requests that a specific news article be displayed.

[1423] Specific behavior:

[1424] A user sends a request to display a news article from an electronic device.

[1425] The server runs an SQL query against the database to retrieve the headlines and content of the relevant news articles.

[1426] Output: News article headline and content data.

[1427] Data processing and calculation:

[1428] Execute an SQL query to retrieve news article information from the database.

[1429] The acquired data is structured and stored in memory.

[1430] Step 4: Spoiler check

[1431] Input: News article headline and content, user preferences.

[1432] Specific behavior:

[1433] The server inputs the retrieved news articles and user settings into the generative AI model.

[1434] As a concrete example, the following prompt sentence is input to the generative AI model:

[1435] "If this news article contains spoilers for 'TV Drama A', please detect the relevant parts."

[1436] The generative AI model analyzes article information and detects whether it contains spoilers.

[1437] Output: Index information of the parts containing spoilers.

[1438] Data processing and calculation:

[1439] Generate a prompt sentence and input it into the model.

[1440] The model analyzes the text and detects the relevant parts (spoiler information).

[1441] Step 5: Generate a customized article

[1442] Input: spoiler check results (index information) and original news article.

[1443] Specific behavior:

[1444] Based on the analysis results, the server replaces the spoiler information with the placeholder "[Spoiler not shown]".

[1445] The entire reconstructed article is temporarily saved.

[1446] Output: A reconstructed news article.

[1447] Data processing and calculation:

[1448] Perform text processing and replace spoilers with placeholders.

[1449] Generate reconstructed article data.

[1450] Step 6: Viewing articles

[1451] Input: A reconstructed news article.

[1452] Specific behavior:

[1453] The server transmits the reconstructed news article to the user's electronic device.

[1454] The terminal renders and displays the news article data received from the server.

[1455] Output: The news article displayed to the user in a safe format.

[1456] Data processing and calculation:

[1457] Generate an HTTP response containing the reconstructed article data.

[1458] The device converts the received data into a format that can be displayed in a browser or application.

[1459] (Application example 1)

[1460] 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."

[1461] There is a growing need to avoid spoilers when viewing news articles and other content. However, existing systems lack sufficient customization for spoiler prevention, making it difficult for users to safely view content. In addition, there is a lack of systems that can train AI models that reflect user feedback.

[1462] 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.

[1463] In this invention, the server, as an application installed on a smartphone, includes means for retrieving news articles from an API and replacing parts containing spoiler information with "[Spoiler Not Displayed]" before displaying them; means for customizing the scope of spoiler information by filtering specific categories and content based on user settings; and means for performing learning to improve the analytical accuracy of the generative artificial intelligence model based on user feedback. This allows users to safely view articles and content they want to view and reduces the risk of spoilers. A "generative artificial intelligence model" is an AI model trained based on machine learning algorithms that is used to analyze news articles and content and determine spoiler information.

[1464] An "electronic device" is a device such as a smartphone, tablet, or computer that a user uses to view news articles.

[1465] An "API" is an application program interface for obtaining user-specified news article information from an external database.

[1466] "Spoiler information" refers to information about the ending or important scenes of a television drama, movie, anime, etc., and is information that may spoil your enjoyment of the content if you know it in advance.

[1467] A "placeholder" is a replacement string or symbol used to hide parts of a page that contain spoilers. A typical example is "[Spoiler Hide]".

[1468] "User settings" refers to customization that the user can make to avoid spoilers about specific content, and allows the user to specify specific categories and content.

[1469] "Feedback" refers to opinions and evaluations provided by users about the content of displayed articles, and is used as data to improve the analysis accuracy of the AI ​​model.

[1470] This invention relates to a system that detects spoilers in news articles and other content and allows users to avoid them. The system consists of an electronic device such as a smartphone, a server, and a generative artificial intelligence model.

[1471] Program processing

[1472] 1. The server receives a request to view a news article from a user's device (such as a smartphone). The user then uses an application installed on the smartphone to view the news article.

[1473] 2. The server retrieves the news articles using an API, which is an application program interface for retrieving article information from an external database.

[1474] 3. To analyze the article information, the server uses a generative artificial intelligence model, which is trained based on a machine learning algorithm and can accurately determine spoilers contained in the article information.

[1475] 4. The server filters spoilers for specific categories or content based on the user's settings, which the user previously entered in the application's settings screen.

[1476] 5. The server analyzes the article information and replaces any detected spoiler information with a placeholder (e.g., "[Spoiler Not Included]"). This process allows users to view the news article in a safe, spoiler-free format.

[1477] 6. When users provide feedback on the displayed article content, the server collects this feedback, which is used as retraining data for the generative AI model to improve its analysis accuracy.

[1478] Hardware and software used

[1479] Hardware: Smartphone (user device), server (analysis and filtering of article information)

[1480] Software: Python, Requests, Transformers (Hugging Face library)

[1481] Specific examples

[1482] As a concrete example, consider the case where a user wants to avoid spoilers for "Drama A." The user registers "Drama A" as a target for spoiler prevention on the settings screen of an application installed on their smartphone. When the user requests the display of news article ID "1234," the server retrieves the corresponding news article through the API. The content of this news article is analyzed by a generative artificial intelligence model, and if spoiler information is detected, the relevant part is replaced with "[Spoiler not shown]." Finally, the server reconstructs the article information in a secure format and sends it to the device to display to the user.

[1483] Example prompt sentence:

[1484] "Analyze whether an article about the latest episode of Drama A contains spoilers about the ending or important scenes."

[1485] As described above, the embodiment of the present invention allows users to read news articles with peace of mind and prevents spoiler information from ruining the enjoyment of the article.

[1486] The flow of the specific processing in Application Example 1 will be explained with reference to FIG. 12. Step 1:

[1487] Users can set spoiler prevention settings

[1488] On the settings screen of the application installed on the smartphone, the user selects the specific content or category (e.g., Drama A) for which they want to avoid spoilers, and saves the settings. The input is the setting information set by the user, and the output is the setting information sent to the server. This setting information is stored on the server in association with the user ID. Specifically, the user enters keywords for the category they selected (e.g., ending, important scene), and presses the save button.

[1489] Step 2:

[1490] The server receives a request to display a news article.

[1491] The user sends a request to display an article from their smartphone. The input is the ID or URL of the news article specified by the user, and the output is the request being received by the server. Specifically, the user selects a specific article from the list of articles in the application and presses the display button to send the request.

[1492] Step 3:

[1493] The server retrieves news articles using an API

[1494] The server calls an API to retrieve news articles from an external database. The input is the article ID and URL, and the output is the retrieved article information. Specifically, the server sends a request to the API endpoint and parses the returned JSON-formatted article data.

[1495] Step 4:

[1496] The server analyzes the article information using a generated artificial intelligence model

[1497] The server uses a generative artificial intelligence model to analyze whether the acquired article information contains spoilers. The input is the article text, and the output is the result of determining whether spoilers are included. Specifically, the article text is input into the model, and the generated output (whether spoilers are present or not) is obtained. This model is implemented using Hugging Face's Transformers library.

[1498] Step 5:

[1499] The server filters spoilers based on configuration information.

[1500] The server filters article content based on a spoiler prevention keyword list set by the user. The input is the article text and user setting information, and the output is the article text with spoiler information filtered out. Specifically, it detects set keywords (e.g., ending, important scene) from the article text and replaces them with placeholders (e.g., [Spoiler not shown]).

[1501] Step 6:

[1502] The server sends the article to the device after processing to prevent spoilers.

[1503] The server sends the article text, from which spoilers have been filtered, to the user's device. The input is the filtered article text, and the output is safe article information that is displayed on the user's device. Specifically, the server sends the filtered article data to the device as an HTTP response, and the device receives it and displays it on the screen.

[1504] Step 7:

[1505] Collect user feedback and incorporate it into the AI ​​model's training data

[1506] The user provides feedback on the content of the displayed article. This feedback is collected by the server and used as retraining data for the generative AI model. The input is the user's feedback information, and the output is a database in which the feedback is saved. Specifically, the user enters their opinion in the feedback form and presses the submit button, and the server saves the information.

[1507] The above are the specific processing steps in this article filtering system.

[1508] 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.

[1509] The present invention relates to a system that analyzes article information, determines whether it contains spoilers, and hides the spoilers as necessary, and further improves the user experience by combining it with an emotion engine that recognizes the user's emotions. Specific embodiments of the system are described below.

[1510] System configuration

[1511] This system consists of a user's device, a server, a generative artificial intelligence (AI) model, and an emotion engine. The user uses the device to view news articles and configure spoiler prevention settings. The server receives this configuration information and retrieves and analyzes news articles. The generative AI model is hosted on the server and is used to analyze article information and determine spoiler information. The emotion engine detects the user's emotions and provides this information to the server.

[1512] Program processing

[1513] 1. User configures spoiler protection:

[1514] The user selects the specific content for which they want to avoid spoilers (for example, "TV Drama A" or "Anime B") from the settings screen on their device, and sends the setting information from the device to the server. The server stores this setting information in association with the user ID.

[1515] 2. The server loads the generative AI model:

[1516] When the server starts up, it loads the generative AI model into memory and prepares it for analyzing article information. This model is trained by a machine learning algorithm and is used to accurately detect spoilers.

[1517] 3. Get news articles and check for spoilers:

[1518] When a user requests a specific news article to be displayed, the server retrieves the headline and content of the relevant news article from the database. The server then inputs the retrieved headline and content into the generative AI model and analyzes whether they contain spoilers. If they are determined to contain spoilers, the server replaces the relevant part with the placeholder "[Spoiler Not Displayed]".

[1519] 4. Emotion Recognition with Emotion Engine:

[1520] When a user uses a device, the emotion engine recognizes the user's emotional state using sensors (such as a camera or microphone) installed on the device. The recognized emotional information is sent to the server in real time.

[1521] 5. Customized Display:

[1522] The server adjusts the display method of spoiler information based on the analysis results, user settings, and the user's emotional state. Based on these results, the server reconstructs the article information so that it does not contain spoilers. The reconstructed content is sent to the device, which then displays the article to the user in a secure format.

[1523] Specific examples

[1524] As a concrete example, consider the case where a user wants to avoid spoilers for "TV Drama A." When the user makes this setting on their device and registers "TV Drama A" as a spoiler prevention target, the server saves this information. Next, when the user attempts to view news article ID "1234," if the headline of this news article is "Review of the latest episode of 'TV Drama A'," the server will use an AI model to analyze the article content. If the AI ​​model determines that the article content contains spoilers, the server will replace the content with "[Spoiler not shown]." At the same time, if the user's emotional information while viewing the device is "surprise" or "discomfort," the server will further take this information into account and adjust the displayed content, reconstructing the article in a way that better takes the user's emotions into consideration.

[1525] Feedback and Learning

[1526] The system also has a function that improves the analytical accuracy of the generative AI model based on user feedback. When users provide feedback on the content of displayed articles, the server collects this information and uses it as retraining data for the AI ​​model. Feedback reflecting users' emotional information is also collected, and this is used to retrain the model for even more accurate filtering. This allows the system's spoiler detection accuracy to improve over time, enabling more accurate filtering.

[1527] In this way, the embodiments of the present invention not only allow users to read news articles with peace of mind, but also provide optimal information display according to their emotional state, allowing them to enjoy a more comfortable and personalized viewing experience.

[1528] The processing flow will be explained below.

[1529] Step 1:

[1530] The user sets up spoiler prevention on the device. The user specifies specific content (e.g., "TV Drama A," "Anime B," etc.) for which they want to avoid spoilers on the device's settings screen.

[1531] Step 2:

[1532] The device sends the setting information to the server, along with the user ID and the specified spoiler prevention target information.

[1533] Step 3:

[1534] The server updates the user settings. Based on the received information, the server saves the spoiler prevention settings associated with the user ID in the database.

[1535] Step 4:

[1536] The server loads the generative AI model. During initialization, the server loads the trained generative AI model into memory, preparing it for analyzing article information.

[1537] Step 5:

[1538] The user requests the display of a news article. The user sends a display request from the terminal to the server specifying a specific news article ID.

[1539] Step 6:

[1540] The device sends a request to the server, which sends the news article ID and user ID information to the server.

[1541] Step 7:

[1542] The server retrieves the news article. The server retrieves the headline and content of the requested article from the database.

[1543] Step 8:

[1544] The server analyzes the article information using the generative AI model. The server inputs the acquired headline and content into the generative AI model and analyzes whether it contains spoilers.

[1545] Step 9:

[1546] The server determines whether the information is a spoiler, and if the generative AI model determines that the information is a spoiler, it identifies the spoiler.

[1547] Step 10:

[1548] The server hides the spoiler information, replaces the spoiler part with the placeholder "[Spoiler hidden]", and modifies the original content.

[1549] Step 11:

[1550] The emotion engine recognizes the user's emotions while using the device. It uses sensors such as the device's camera and microphone to analyze the user's facial expressions and voice to recognize emotions.

[1551] Step 12:

[1552] The device transmits the recognized emotion information to the server. The device transmits the user's emotional state information recognized in real time to the server.

[1553] Step 13:

[1554] The server reconstructs the article information based on the user's emotional state. The server adjusts the displayed content taking into account the user's emotional state in addition to the analysis results and spoiler prevention settings.

[1555] Step 14:

[1556] The server sends the revised article information to the terminal. The server sends the revised headline and content to the terminal.

[1557] Step 15:

[1558] The terminal displays the corrected article information to the user. The terminal displays the received headline and the corrected content to the user.

[1559] Step 16:

[1560] When a user views an article and provides feedback, the user transmits their rating and opinion to the server via the terminal.

[1561] Step 17:

[1562] The server collects the feedback and stores it in a database.

[1563] Step 18:

[1564] The server retrains the generative AI model. The server uses the collected feedback data to retrain the generative AI model to improve its accuracy.

[1565] Through these steps, users can not only browse news articles with peace of mind, but also enjoy a more comfortable and personalized browsing experience by being provided with optimal information display according to their emotional state.

[1566] Example 2

[1567] 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."

[1568] Conventional news article browsing systems often display articles containing unwanted spoilers, which detracts from the user's browsing experience. Furthermore, because they only implement a uniform approach to preventing spoilers without taking into account the user's emotional state, they are unable to provide optimal information to each individual user. Furthermore, there is a lack of effort to improve the accuracy of generative AI models based on user feedback, limiting the accuracy of spoiler detection.

[1569] 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 receiving spoiler prevention settings from a user terminal and registering specific categories or content as spoiler prevention targets; means for analyzing article information on the server using a generative AI model and determining whether the article information may contain spoiler information; means for hiding spoiler information from the user based on the determination result; and means for customizing the display method of spoiler information based on user settings and emotion information generated by an emotion engine. This allows users to avoid unwanted spoiler information and provides optimal information according to their individual emotional state. Furthermore, by improving the accuracy of the generative AI model based on feedback, more accurate spoiler prevention can be achieved.

[1570] A "user terminal" is an electronic device that is operated by a user to view information and change settings.

[1571] The "spoiler prevention setting" is information that a user sets to avoid spoiler information for a specific content or category.

[1572] A "generative AI model" is a model trained using machine learning algorithms to analyze article information and detect spoilers.

[1573] A "server" is a device that provides services over a network and receives and processes requests from user terminals.

[1574] "Spoiler information" is information that includes important details or the ending of content that the user has not yet seen.

[1575] An "emotion engine" is a software or hardware mechanism for recognizing and analyzing a user's emotional state.

[1576] A "placeholder" is alternative text or symbols that are displayed in place of spoiler information.

[1577] "Feedback" refers to information such as usage experience, impressions, and evaluations provided by users, and is used to improve the system.

[1578] "Analysis accuracy" refers to the degree to which the generative AI model can accurately detect spoiler information.

[1579] "News Article" means article content provided via the Internet or other media.

[1580] "Customization" means changing the system's behavior and display methods according to the user's settings and status.

[1581] The present invention relates to a system that analyzes news articles, determines whether they contain spoilers, and hides the spoilers as necessary. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system further improves the user experience. Specific embodiments of the system are described below.

[1582] System configuration

[1583] This system consists of a user device, a server, a generative AI model, and an emotion engine. The user uses the device to view news articles and configure spoiler prevention settings. The server receives this configuration information and retrieves and analyzes news articles. The generative AI model is hosted on the server and is used to analyze news articles and determine spoiler information. The emotion engine detects the user's emotions and provides this information to the server.

[1584] Program processing

[1585] User-defined spoiler prevention settings

[1586] The user selects the specific content for which they want to avoid spoilers, such as "TV dramas" or "movies," on the device's settings screen. The device then sends the selected setting information to the server, which then associates the received setting information with the user ID and stores it in a database.

[1587] Loading of generated AI models by the server

[1588] At startup, the server loads into memory a generative AI model that is trained using machine learning algorithms and used to analyze news articles and identify spoilers.

[1589] Get news articles and check for spoilers

[1590] When a user requests a specific news article to be displayed, the server retrieves the relevant news article from the database. The headline and content of the retrieved article are input into the generative AI model to determine whether it contains spoilers. If the model detects any part that it deems a spoiler, the server replaces that part with "[Spoiler Not Displayed]."

[1591] Emotion recognition by emotion engine

[1592] While a user is browsing a news article on their device, the emotion engine uses the device's built-in camera and microphone to recognize the user's emotional state, and the recognized emotional information is sent to the server in real time.

[1593] Customized View

[1594] The server adjusts the display method of the spoiler information based on the analysis results, user settings, and the user's emotional information. The reconstructed article content is sent to the device, which then displays the news article to the user in a secure format.

[1595] Specific examples

[1596] As a concrete example, consider the case where a user wants to avoid spoilers for a "TV drama." When the user makes this setting on their device and registers "TV drama" as a spoiler prevention target, the server saves this information. Next, when the user attempts to view news article ID "1234," if the headline of this news article is "Review of the latest episode of 'TV drama'," the server will analyze the article content using the generative AI model. If the generative AI model determines that the article content contains spoilers, the server will replace the content with "[Spoiler not displayed]."

[1597] At the same time, if the emotional information displayed by the user while browsing the device is "surprise" or "discomfort," the server will further take this information into account and adjust the displayed content, reconstructing the article in a way that takes the user's emotions into greater consideration.

[1598] Prompt Sentence Examples

[1599] Examples of input prompts for a generative AI model include:

[1600] "Please analyze what spoilers this news article contains."

[1601] "The user has specified that they would like to avoid spoilers for 'TV drama'. Analyze the content of this news article and replace any spoilers with 'Spoiler Hide'."

[1602] Using these prompts, we can see how the generative AI model parses news articles and appropriately detects and filters spoilers.

[1603] This allows users to avoid unwanted spoilers and obtain optimal information according to their emotional state. Furthermore, by improving the accuracy of the generative AI model based on feedback, even more accurate spoiler prevention can be achieved.

[1604] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1605] Step 1:

[1606] User enters and submits spoiler protection settings

[1607] The user opens the settings screen on the device and selects the specific category or content (e.g., "TV dramas" or "movies") for which they want to avoid spoilers. Using the selected information as input, the device sends the setting information to the server. Specifically, the user taps "Add spoiler-protected content," selects "TV dramas," and presses the "Save" button. The device then sends a request to the server stating, "User ID: 123 has set his preference to avoid spoilers for TV dramas."

[1608] Input: Specific categories or content you want to avoid spoilers for

[1609] Output: Sending configuration information to the server

[1610] Step 2:

[1611] Server-based storage of configuration information

[1612] The server receives the setting information sent from the device, associates it with the user ID, and stores it in the database. Specifically, the server analyzes the data it receives and adds the information "User ID: 123" and "Spoiler prevention target: TV drama" to the user information table in the database.

[1613] Input: Setting information sent from the device

[1614] Output: User preferences stored in the database

[1615] Step 3:

[1616] Loading a generative AI model

[1617] When the server starts up, it loads the generative AI model into memory. This generative AI model is trained using a machine learning algorithm and is used to analyze news articles and determine spoiler information. Specifically, when the server is restarted, the generative AI model loading process is automatically executed and the model is loaded into memory. The system log records "The generative AI model has been loaded."

[1618] Input: Start the server

[1619] Output: The loaded generative AI model

[1620] Step 4:

[1621] Processing a request to retrieve news articles

[1622] A user sends a specific news article ID as input from their device, requesting the server to display that article. The server receives this request, searches for the requested news article in its internal database, and retrieves the headline and content of the corresponding news article. Specifically, when a user taps to display "News Article ID: 1234" on their smartphone app, the device sends a request to the server saying, "I want to display article ID 1234." The server retrieves the article from the database and returns the headline and body data, such as "Review of the latest TV drama episode."

[1623] Input: News article ID

[1624] Output: News article headline and content

[1625] Step 5:

[1626] Analysis of spoiler information in article content

[1627] The server inputs the headline and content of the retrieved news article into the generative AI model and analyzes whether or not it contains spoilers. When the model detects any parts that it determines contain spoilers, the server replaces those parts with "[Spoiler not shown]". Specifically, the server sends the generative AI model a prompt message saying, "Analyze the content of article ID: 1234 and detect any parts that contain spoilers." When the generative AI model responds that the parts contain spoilers, it replaces them with "[Spoiler not shown]".

[1628] Input: News article headline and content

[1629] Output: News article with spoilers replaced

[1630] Step 6:

[1631] Emotion recognition by emotion engine

[1632] While a user is viewing a news article on their device, the emotion engine uses the device's built-in camera and microphone to recognize the user's emotional state. The recognized emotional information is sent to the server in real time. Specifically, while the user is viewing a news article, the device's camera captures the user's facial expressions, and the emotion engine detects the "surprise" expression. This information is sent to the server as "User ID: 123 is in a surprised emotional state."

[1633] Input: Emotion information from the device

[1634] Output: Emotion information sent to the server

[1635] Step 7:

[1636] Customized article display

[1637] The server adjusts the display method of spoiler information based on the analysis results, user settings, and the user's emotional information. The reconstructed article content is sent to the device, and the device displays the news article to the user in a safe format. Specifically, when the server receives that the user is in the "surprise" emotional information state, it applies further filtering to avoid spoilers. The reconstructed article content is sent to the device and displayed as a "safe article."

[1638] Input: Analysis results, user settings, emotional information

[1639] Output: Reconstructed news article

[1640] (Application example 2)

[1641] 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."

[1642] In conventional news article and content distribution services, users have difficulty avoiding unwanted spoilers. Furthermore, the optimal display of information based on the user's emotional state can sometimes impair the browsing experience. Therefore, there is a need for a system that allows users to browse content with peace of mind and that displays information appropriate to their emotional state.

[1643] 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 analyzing article information of the electronic device using a generative artificial intelligence model and determining whether it is likely to contain spoiler information, means for preventing spoilers by replacing part of the article containing spoiler information with a placeholder, emotion recognition means for detecting the emotional state of the user, and means for adjusting the display method of the article information based on the detected emotion information. This allows the user to avoid spoiler information and also makes it possible to display information optimally according to the user's emotional state.

[1644] A "generative artificial intelligence model" is a model that is trained based on machine learning algorithms to analyze text data and determine specific information.

[1645] "Electronic equipment" is a general term for devices used to process and display digital data, including smartphones, tablets, and personal computers.

[1646] "Spoiler information" is information that reveals important developments or endings of content such as movies, dramas, anime, and books in advance.

[1647] A "means for determining" is a method or device for utilizing a generative AI model to analyze text data and identify whether it contains specific information.

[1648] A "placeholder" is a substitute display string that is temporarily used to hide the original information.

[1649] "Emotion recognition means" refers to a method or device for detecting and analyzing emotions from a user's facial expressions, voice, actions, etc.

[1650] "User settings" refers to setting information that allows a user to customize the operating conditions of the system and the display contents based on their own preferences.

[1651] "Customizable means" means a method or device for changing or adjusting the system's functionality or display content based on user settings.

[1652] "Emotion information" is data indicating the emotional state of the user detected by the emotion recognition means.

[1653] The "means for adjusting the display method of article information" is a method or device for changing the content or format of information displayed to a user based on emotion information.

[1654] The system for implementing this invention is composed of a user terminal, a server, a generative AI model, and emotion recognition means. This allows users to browse news articles and other content with peace of mind, and displays information appropriate to their emotional state.

[1655] System configuration

[1656] 1. On the user's device:

[1657] Users browse news articles and other content using electronic devices such as smartphones, tablets, smart glasses, and head-mounted displays. These devices are equipped with emotion recognition sensors (cameras and microphones) to detect the user's emotional state in real time.

[1658] 2. Server:

[1659] The server stores and manages user settings, a news article database, a generative AI model, emotion recognition means, and a feedback database. The server has the following functions:

[1660] Loading the generative AI model: When the server starts up, it loads the generative AI model into memory to analyze article information. This generative AI model is trained based on machine learning algorithms and has the ability to detect spoilers.

[1661] Acquisition and analysis of article information: When a user attempts to view a news article, the server retrieves the relevant news article from the database and passes it to the generative AI model to analyze spoiler information.

[1662] Spoiler hiding: If the analysis reveals that a spoiler is included, the server replaces the relevant part with the placeholder "[Spoiler hidden]" and displays it to the user.

[1663] Use of emotional information: The server collects the user's emotional information sent from the device and adjusts the way article information is displayed based on this information.

[1664] User settings management: Users can set the spoiler categories and content they want to avoid according to their preferences, and the server stores and manages this setting information.

[1665] 3. Emotion recognition means:

[1666] The emotion recognition means analyzes the user's facial expressions and voice data to detect emotional information in real time. This emotional information is sent to the server and used to adjust the display method of article information.

[1667] Specific examples of the embodiment

[1668] For example, if a user wants to avoid spoilers for the movie "Movie X," they can register "Movie X" as a spoiler-protected content on the device's settings screen. The server receives and stores this setting information.

[1669] When a user attempts to view an article related to "Movie X," the server retrieves the article from the database, passes it to the generative AI model, and analyzes it for spoilers. If spoilers are found, the server replaces the relevant part with "[Spoiler Not Displayed]" and delivers it to the user.

[1670] At the same time, if the user's emotional state is surprise or displeasure, the server will adjust the way the article is displayed based on the emotional information, providing the user with a better browsing experience.

[1671] Examples of instructions and prompts

[1672] An example of a prompt sentence when using the sentiment analysis API is as follows:

[1673] python

[1674] Prompt sentence for using the sentiment analysis API

[1675] import requests

[1676] def analyze_sentiment(audio_file_path):

[1677] url = "http: / / sentiment-api.example.com / analyze"

[1678] files = {'file': open(audio_file_path, 'rb')}

[1679] response = requests.post(url, files=files)

[1680] return response.json()

[1681] Analyze audio files to recognize emotions

[1682] emotion_result = analyze_sentiment("path / to / user_audio.wav")

[1683] print(emotion_result)

[1684] This prompt shows how to send an audio file to the sentiment analysis API and receive the sentiment result.

[1685] This allows the user to avoid spoiler information and enjoy optimal information display suited to their own emotional state.

[1686] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1687] Step 1:

[1688] The user sets up spoiler prevention on the device. The user opens the device's settings screen, selects the specific content for which they want to avoid spoilers (e.g., "Movie X"), and enters the setting information. This setting information is sent by the device to the server. The input data consists of the user ID and setting information, which the server receives and stores in a database in association with the user ID.

[1689] Step 2:

[1690] The server loads the generative AI model. At startup, the server loads the generative AI model into memory and prepares it for analyzing article data. This model is trained based on a machine learning algorithm and is used to detect spoilers from news articles. The input data are the parameters of the generative AI model, and the output is the state of the model loaded in memory.

[1691] Step 3:

[1692] A user requests to view an article. When a user uses a device to view a specific news article, the device sends the request to the server. This request includes the article ID. The server receives the user's request and the article ID as input data.

[1693] Step 4:

[1694] The server retrieves and analyzes news articles. The server retrieves the headline and content of the news article corresponding to the specified article ID from the database. The retrieved article content is passed to a generative AI model, which analyzes whether it contains spoilers. The input data is the text data of the news article, and the output data is the spoiler detection results.

[1695] Step 5:

[1696] Concealment of spoiler information. If the generative AI model determines that an article contains spoiler information, it replaces the relevant part with "[Spoiler hidden]." The input data is the spoiler detection result and the text data of the article, and the output data is the text data of the article with the spoiler hidden.

[1697] Step 6:

[1698] Emotion detection using emotion recognition means. When a user uses a device to view an article, sensors (camera and microphone) installed on the device detect the user's emotional state in real time and send that data to the server. The input data is the user's facial expression and voice data, and the output data is the recognized emotional information.

[1699] Step 7:

[1700] Adjustment of article display based on emotional information. The server adjusts the display method of article information based on the user's emotional information obtained by the emotion recognition means. For example, if the user expresses emotions such as "surprise" or "discomfort," the server further adjusts the display content of the article and uses appropriate placeholders. The input data is the emotional information and the text data of the concealed article, and the output data is the adjusted article data that is finally displayed to the user.

[1701] Step 8:

[1702] Delivery of the adjusted article. The server finally sends the adjusted article to the device, where the user views it. The input data is the adjusted article data, and the output data is the article displayed on the user's device. This allows the user to avoid spoilers and experience information displayed optimally according to their emotional state.

[1703] In this way, each step works in cooperation with the others, allowing users to browse news articles and other content safely and comfortably.

[1704] 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.

[1705] 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.

[1706] 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.

[1707] 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.

[1708] 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.

[1709] 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.

[1710] 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).

[1711] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1712] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1713] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1714] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1715] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1716] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1717] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1718] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1719] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1720] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1721] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1722] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1723] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1724] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1725] The following is further disclosed regarding the above embodiment.

[1726] (Claim 1)

[1727] Analyzing article information for electronic devices using a generative artificial intelligence model;

[1728] Determine if the information may contain spoilers

[1729] Means and

[1730] Based on the result of the determination, the spoiler information is hidden so as not to be displayed to the user.

[1731] Means and

[1732] Customize the scope of spoilers by filtering specific categories and content based on user preferences

[1733] Means and

[1734] A system including:

[1735] (Claim 2)

[1736] Based on user feedback, the generative AI model learns to improve its analytical accuracy.

[1737] The system of claim 1 further comprising: a means for detecting a temperature change of the sensor;

[1738] (Claim 3)

[1739] Prevent spoilers by replacing parts of articles that may contain spoilers with placeholders

[1740] The system of claim 1 further comprising: a means for detecting a temperature change of the sensor;

[1741] "Example 1"

[1742] (Claim 1)

[1743] The user sets spoiler prevention settings using an electronic device and transmits the setting information to the server.

[1744] Means and

[1745] The server analyzes the article information using a generative artificial intelligence model to determine whether it may contain spoilers.

[1746] Means and

[1747] Based on the result of the determination, the spoiler information is hidden so as not to be displayed to the user.

[1748] Means and

[1749] The server allows users to customize the scope of spoiler information by filtering specific types and content based on their preferences.

[1750] Means and

[1751] When a user requests the display of a specific news article, the server retrieves the relevant news article, reconstructs it, and sends it to the device.

[1752] Means and

[1753] A system including:

[1754] (Claim 2)

[1755] Based on user feedback, the generative AI model learns to improve its analytical accuracy.

[1756] The system of claim 1 further comprising: a means for detecting a temperature change of the sensor;

[1757] (Claim 3)

[1758] Prevent spoilers by replacing parts of articles that may contain spoilers with placeholders

[1759] The system of claim 1 further comprising: a means for detecting a temperature change of the sensor;

[1760] "Application Example 1"

[1761] Okay, now let's rewrite the claims according to the following format:

[1762] (Claim 1)

[1763] Analyzing article information for electronic devices using a generative artificial intelligence model;

[1764] Determine if the information may contain spoilers

[1765] Means and

[1766] Based on the result of the determination, the spoiler information is hidden so as not to be displayed to the user.

[1767] Means and

[1768] Customize the scope of spoilers by filtering specific categories and content based on user preferences

[1769] Means and

[1770] As an application installed on a smartphone, it retrieves news articles from an API and replaces any parts containing spoilers with "[Spoiler Not Displayed]" before displaying them.

[1771] A system including:

[1772] (Claim 2)

[1773] Based on user feedback, the generative AI model learns to improve its analytical accuracy.

[1774] The system of claim 1 further comprising: a means for detecting a temperature change of the sensor;

[1775] (Claim 3)

[1776] Prevent spoilers by replacing parts of articles that may contain spoilers with placeholders

[1777] The system of claim 1 further comprising: a means for detecting a temperature change of the sensor;

[1778] "Example 2: Combining Emotion Engines"

[1779] (Claim 1)

[1780] A means for receiving spoiler prevention settings from a user terminal and registering specific categories or content as spoiler prevention targets;

[1781] A means for analyzing article information on a server using a generative AI model and determining whether it may contain spoilers;

[1782] a means for hiding spoiler information so that it is not displayed to a user based on the determination result;

[1783] A means for customizing the way spoiler information is displayed based on user settings and emotion information generated by an emotion engine;

[1784] A system including:

[1785] (Claim 2)

[1786] 10. The system of claim 1, further comprising means for performing learning to improve the analytical accuracy of the generative artificial intelligence model based on feedback from a user.

[1787] (Claim 3)

[1788] 10. The system of claim 1, further comprising means for preventing spoilers by replacing portions of articles that may contain spoiler information with placeholders.

[1789] "Application example 2 when combining emotion engines"

[1790] (Claim 1)

[1791] Analyzing article information for electronic devices using a generative artificial intelligence model;

[1792] Determine if the information may contain spoilers

[1793] Means and

[1794] Based on the result of the determination, the spoiler information is hidden so as not to be displayed to the user.

[1795] Means and

[1796] Customize the scope of spoilers by filtering specific categories and content based on user preferences

[1797] Means and

[1798] emotion recognition means for detecting an emotional state of a user;

[1799] The method of displaying article information is adjusted based on the emotion information detected by the emotion recognition means.

[1800] Means and

[1801] A system including:

[1802] (Claim 2)

[1803] Based on user feedback, the generative AI model learns to improve its analytical accuracy.

[1804] The system of claim 1 further comprising: a means for detecting a temperature change of the sensor;

[1805] (Claim 3)

[1806] Prevent spoilers by replacing parts of articles that may contain spoilers with placeholders

[1807] The system of claim 1 further comprising: a means for detecting a temperature change of the sensor; [Explanation of symbols]

[1808] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. Analyzing article information for electronic devices using a generative artificial intelligence model; Determine if the information may contain spoilers Means and Based on the result of the determination, the spoiler information is hidden so as not to be displayed to the user. Means and Customize the scope of spoilers by filtering specific categories and content based on user preferences Means and A system including:

2. Based on user feedback, the generative AI model learns to improve its analytical accuracy. The system of claim 1 further comprising: a means for detecting a change in a frequency of the signal;

3. Prevent spoilers by replacing parts of articles that may contain spoilers with placeholders The system of claim 1 further comprising: a means for detecting a change in a frequency of the signal;

Images