System

Abstract

To provide a system capable of deepening a user's understanding of news, increasing a stay time, and allowing AI to be a receiver of an aggressive contribution.SOLUTION: Provided is a system that includes: means for generating, by a generative AI model, comments in a format similar to comments posted by general users for the latest news articles obtained from a news site; means for posting generated comments in a comment section of the news article so that the comments are displayed only to specific users; means for generating, by a generative AI model, reply comments to comments posted by specific users; and means for posting the generated reply comment in a comment section of the news in a manner that makes it visible only to the specific user, and after the exchange with the specific user is over, for generating a summary of the contents of the comments exchanged with the specific user and posting the summary in a manner that makes it viewable by other users to the comment section of the news.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] Traditional news sites lack concrete ways for users to deepen their understanding of the news, lack effective methods for increasing user time on the site, and require appropriate measures to deal with offensive posts. [Means for solving the problem]

[0005] This invention introduces an AI character into the comment section of news articles. The AI ​​character comments on news articles in a format similar to that of regular users. Users can exchange messages with the AI ​​character about the news, and these exchanges are visible only to the user. After the exchange is over, users can post a "summary of the exchange." This deepens users' understanding of the news, increases their time on the site, and allows the AI ​​to act as a receptacle for offensive posts. [Brief explanation of the drawings]

[0006] [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. 2 is a sequence diagram showing a flow of processing in the data processing system according to the first embodiment of the first form example. [Figure 12]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Embodiment 1. [Figure 13] FIG. 10 is a sequence diagram showing a processing flow of a data processing system in a second embodiment of the second form example. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Embodiment 2. [Figure 15] FIG. 10 is a sequence diagram showing the flow of processing in a data processing system according to a third embodiment of the third embodiment. [Figure 16] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Embodiment 3. [Figure 17] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the first embodiment of the first form example when an emotion engine is combined. [Figure 18] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Embodiment 1 when an emotion engine is combined. [Figure 19] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the second embodiment of the second form example when an emotion engine is combined. [Figure 20] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Form Example 2 when an emotion engine is combined. [Figure 21] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the third embodiment of the third form example when an emotion engine is combined. [Figure 22] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Form Example 3 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0007] 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.

[0008] First, the terms used in the following description will be explained.

[0009] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, the processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)).

[0010] 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.

[0011] 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.

[0012] 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.

[0013] 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."

[0014] [First embodiment]

[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] 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.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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.

[0022] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0023] 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.

[0024] 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.

[0025] 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.

[0026] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0027] "Example 1"

[0028] One embodiment of the present invention is a system that introduces an AI character to a news site. In this system, the AI ​​character posts comments on news articles. The AI ​​character's comments are in a format similar to that of general users, and users can exchange messages with the AI ​​character about the news. These exchanges are set to be visible only to the user. After the exchange is over, the user can post a "summary of the exchange."

[0029] "Example 2"

[0030] As a specific example, consider an AI character posting a comment on a political news article. The AI ​​character understands the content of the article and generates a comment based on that understanding. For example, the AI ​​character can post a question such as, "How do you think this bill will affect small and medium-sized businesses?". Users can express their opinion in response to this question, and the AI ​​character will respond with an appropriate response. This interaction helps users gain a deeper understanding of the news article.

[0031] "Example 3"

[0032] The system of this invention also has the effect of increasing the time users spend on the site. As users enjoy interacting with the AI ​​character, they spend more time on the news site. Furthermore, the AI ​​character acts as a receptacle for offensive posts, preventing conflict between users.

[0033] The processing flow of each embodiment will be described below.

[0034] "Example 1"

[0035] Step 1: The AI ​​character analyzes the news article, including the article's text content, associated images and videos, and other metadata.

[0036] Step 2: The AI ​​character generates comments on the news article, which are formatted similarly to typical user comments.

[0037] Step 3: The user responds to the AI ​​character's comments. The user can express their own opinions and thoughts to the AI ​​character.

[0038] Step 4: The AI ​​character generates an appropriate reply to the user's reaction. This reply is a response to the user's comment and continues the dialogue between the user and the AI ​​character.

[0039] Step 5: The user submits a summary of the conversation. This summary is a concise summary of the conversation between the user and the AI ​​character.

[0040] "Example 2"

[0041] Step 1: The AI ​​character analyzes political news articles, including the article's text content, associated images and videos, and other metadata.

[0042] Step 2: The AI ​​character generates question-type comments on news articles, such as "How do you think this bill will affect small businesses?"

[0043] Step 3: The user gives their opinion in response to the AI ​​character's question.

[0044] Step 4: The AI ​​generates an appropriate response to the user's opinion. This response is a reaction to the user's opinion and continues the dialogue between the user and the AI.

[0045] Step 5: The user submits a summary of the conversation. This summary is a concise summary of the conversation between the user and the AI ​​character.

[0046] "Example 3"

[0047] Step 1: A user visits a news site.

[0048] Step 2: The user initiates a dialogue with the AI ​​character, which is based on a news article.

[0049] Step 3: Users enjoy interacting with the AI ​​character, which increases the time they spend on the news site.

[0050] Step 4: If a user posts something offensive, the AI ​​character will respond, preventing any conflict between users.

[0051] Example 1

[0052] Next, a description will be given of Example 1 of Form 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."

[0053] On traditional news sites, when users post comments on news articles, their interactions with other users are made public, raising the risk of privacy issues and offensive comments. Furthermore, there is a lack of support for users to gain a deeper understanding of the content of news articles, which can lead to a shallow understanding of the articles. Furthermore, there is a lack of efforts to increase user time on the site, which can lead to a decline in site usage.

[0054] 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.

[0055] In this invention, the server includes: means for introducing an AI character into a news comment section; means for the AI ​​character to comment on the news in a format similar to that of general users; means for the user to interact with the AI ​​character about the news; means for the interaction to be visible only to the user; means for the user to post a "summary of the interaction" after the interaction is over; means for generating comments on news articles using a generative AI model; means for inputting user input as a prompt sentence into the generative AI model to generate a reply; and means for retrieving news articles. This allows users to comment on news articles while maintaining their privacy and deepen their understanding of the news article through their interaction with the AI ​​character. It also increases user dwell time and reduces the risk of offensive comments.

[0056] A "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[0057] An "artificial intelligence character" is a virtual character generated using artificial intelligence technology that can interact with users.

[0058] A "format similar to comments from general users" is a format that has the same style and content as comments posted by regular users.

[0059] "User" means an individual or organization that uses a news site.

[0060] "Interaction" refers to the dialogue and communication that takes place between the user and the AI ​​character.

[0061] The "interaction summary" is a concise summary of the interaction between the user and the AI ​​character.

[0062] A "generative AI model" is a model for generating text using artificial intelligence technology, including those that use natural language processing technology.

[0063] A "prompt" is an instruction or question to be input into a generative AI model.

[0064] A "news article" is a piece of text containing news or information that is published on a news site.

[0065] A "server" is a computer system used to operate a news site and process data.

[0066] "User session information" is information that records the status and activities of a user while the user is accessing a news site.

[0067] A "database" is a system for storing data such as news articles, user comments, and summaries of interactions.

[0068] The present invention is a system that introduces an AI character to a news site, allowing users to post comments on news articles and interact with the AI ​​character. This system includes elements of a server, a terminal, and a user.

[0069] System Overview

[0070] The server retrieves the latest news articles from news sites and uses a generative AI model to generate comments on the news articles. The generated comments are in a format similar to comments made by general users and are only displayed to specific users. Users can use their devices to interact with the AI ​​character about the news article. After the interaction is complete, users can post a "summary of the interaction."

[0071] Hardware and software used

[0072] Server: Retrieves news articles, runs generative AI models, generates and displays comments, and manages user sessions.

[0073] Device: The device (PC, smartphone, etc.) on which users view news articles, post comments, and interact with AI characters.

[0074] Generative AI models: Artificial intelligence models for generating comments on news articles (e.g., OpenAI's GPT-4).

[0075] Data processing and calculation

[0076] 1. Get news articles: The server uses the news site's API to get the latest news articles and stores them in a database.

[0077] 2. Comment generation: The server inputs the content of the retrieved news article as a prompt sentence into the generative AI model to generate a comment.

[0078] 3. Display the comment: The server displays the generated comment, associating it with the specific user session.

[0079] 4. User interaction: When a user enters a comment using the device, the server inputs the comment as a prompt into the generative AI model to generate a reply.

[0080] 5. Interaction Summary: When a user enters a summary of an interaction, the server stores the summary in a database and makes it available for other users to view.

[0081] Specific examples

[0082] News article: "COVID-19 cases declining as vaccinations progress."

[0083] AI character comment: "It's good news that vaccinations are progressing. We can also expect the number of infections to decrease."

[0084] Prompt Sentence Examples

[0085] "Please generate a comment from an AI character in a format similar to that of a regular user in response to a news article about the decline in the number of COVID-19 infections as vaccinations progress."

[0086] In this way, the system of the present invention allows users to post comments on news articles and interact with AI characters, thereby improving user understanding of the news articles while maintaining privacy and increasing site usage.

[0087] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0088] Step 1: Get the news article

[0089] The server retrieves the latest news articles using the news site's API. Specifically, the server calls the API at "https: / / newsapi.org / " every morning at 9:00 and retrieves the latest news articles. The retrieved news articles are stored in a database.

[0090] Input: API request

[0091] Output: News article data

[0092] Data processing: Store the news article data obtained from the API in a database.

[0093] Step 2: Generate comments for the AI ​​character

[0094] The server inputs the content of the retrieved news article into the generative AI model as a prompt sentence to generate a comment. Specifically, the title and body of the news article are input into the generative AI model as prompt sentences. For example, in response to a news article stating, "As COVID-19 vaccinations progress, the number of infections is decreasing," the server inputs the prompt sentence, "What do you think about this news?" into the generative AI model to generate a comment.

[0095] Input: News article data, prompt

[0096] Output: AI character's comment

[0097] Data processing: News article data is converted into prompt sentences, which are then fed into a generative AI model to generate comments.

[0098] Step 3: View comments

[0099] The server associates the generated AI character's comments with a specific user's session and displays them. Specifically, it uses the user's session information to ensure that the comments are only displayed to that specific user. For example, if user A is logged in, it uses user A's session ID to ensure that the comments are only displayed to user A.

[0100] Input: AI character comments, user session information

[0101] Output: Comments that are displayed to the user

[0102] Data processing: Displays the AI ​​character's comments based on the user's session information.

[0103] Step 4: User interaction

[0104] Users use their devices to interact with the AI ​​character about news articles. Specifically, when a user enters a comment, the server inputs the comment as a prompt into the generative AI model and generates a reply. For example, if a user inputs a comment such as "It's good news that vaccinations are progressing," the server inputs the comment as a prompt into the generative AI model and generates a reply.

[0105] Input: User comments

[0106] Output: AI character's reply

[0107] Data processing: User comments are converted into prompt sentences and fed into a generative AI model to generate replies.

[0108] Step 5: Summarize the conversation

[0109] After the interaction is complete, the user enters a "summary of the interaction" and posts it. Specifically, when a user enters a summary of the interaction, the server saves the summary in a database and makes it available for other users to view. For example, if a user enters a summary such as "Through my interaction with the AI ​​character, I was able to reaffirm the importance of vaccination," the server saves the summary in a database and makes it available for other users to view.

[0110] Input: User Summary

[0111] Output: Summary stored in database

[0112] Data processing: Save the user's summary in a database and make it available for other users to view.

[0113] (Application example 1)

[0114] Next, a description will be given of Application Example 1 of Embodiment 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."

[0115] On traditional news sites, users can refer to other users' comments when reading news articles, but they lack the ability to have individual conversations. They also lack support for users to gain a deeper understanding of news articles and are not designed to increase user time on the site. Furthermore, offensive comments are sometimes posted, and measures to address this are needed.

[0116] 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.

[0117] In this invention, the server includes means for introducing an AI character into a news comment section, means for the AI ​​character to comment on the news in a format similar to that of general users, means for the user to interact with the AI ​​character about the news, means for the interaction to be visible only to the user, means for the user to post a "summary of the interaction" after the interaction is over, means for the AI ​​character to post comments and interact with the user as the user reads the news article, and means for generating a summary of the interaction between the user and the AI ​​character. This allows users to understand the news article while individually interacting with the AI ​​character, deepening their understanding of the news and increasing their stay time. Furthermore, the AI ​​character's role as a receptacle for offensive comments helps maintain a healthy comment section.

[0118] A "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[0119] An "artificial intelligence character" is a virtual character generated using artificial intelligence technology that has the ability to interact with users.

[0120] A "format similar to comments from general users" is a format that has the same style and content as comments posted by regular users.

[0121] "Means by which the user can exchange news with the AI ​​character" is a function that allows the user to have two-way communication with the AI ​​character.

[0122] "Means for making the exchange visible only to the user" refers to a function that prevents the content of the exchange between the user and the AI ​​character from being displayed to other users.

[0123] "A means to post a 'summary of the conversation' after the conversation is over" is a function that allows users to post a concise summary of the content of their conversation with an AI character.

[0124] "A means for an AI character to post comments and interact with users when they read news articles" is a function that enables an AI character to provide comments to users who are viewing news articles and start a dialogue.

[0125] The "means for generating a summary of the interaction between the user and the AI ​​character" is a function for automatically summarizing the content of the conversation between the user and the AI ​​character.

[0126] As an embodiment of the present invention, a system for introducing an AI character into a news site is constructed. A specific embodiment of the system is shown below.

[0127] System Program

[0128] This system consists of a server and a user terminal. The server provides a function that allows an AI character to post comments on news articles and interact with users. The user terminal displays news articles and provides an interface for interacting with the AI ​​character.

[0129] Hardware and software used

[0130] Hardware: User devices such as smartphones, tablets, and PCs

[0131] Software: Flask (Python web framework), OpenAI API (generative AI model)

[0132] Data processing and calculation

[0133] 1. Acquisition of news articles: The server acquires news articles from the database and sends them to the user terminal.

[0134] 2. AI character comment generation: The server receives the news article as input and uses a generative AI model (e.g., text-davinci-003) to generate AI character comments based on the prompt sentence.

[0135] 3. User interaction: The user device displays the generated comments and provides an interface through which the user can interact with the AI ​​character. The user's input is sent to the server, which again uses the generative AI model to generate a response.

[0136] 4. Summary generation of interaction: The server receives the content of the interaction between the user and the AI ​​character as input and generates a summary using the generative AI model. The generated summary is sent to the user's device so that the user can post it.

[0137] Specific examples

[0138] Suppose a user is reading a news article titled "COVID-19 Vaccination Progress." An AI character comments, "As vaccination progresses, it is expected that herd immunity will be achieved." The user responds, "That's true, but I'm worried about side effects." The AI ​​character responds, "It would be best to consult an expert about side effects." After the exchange, the user posts a summary saying, "We discussed the progress of vaccination and concerns about side effects."

[0139] Prompt Sentence Examples

[0140] Generating comments from AI characters on news articles:

[0141] News article: COVID-19 vaccinations progress

[0142] AI character's comment:

[0143] Generate summaries of user interactions:

[0144] User interaction with AI character:

[0145] User: As vaccinations progress, it is hoped that we will achieve herd immunity.

[0146] AI Character: Yes, but I'm worried about the side effects.

[0147] User: It's best to seek expert advice regarding side effects.

[0148] summary:

[0149] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0150] Step 1:

[0151] The server retrieves the news articles from the database and transmits them to the user terminal.

[0152] Input: News article ID

[0153] Data processing: Search and retrieve news article content from the database.

[0154] Output: News article content

[0155] Specific operation: The server searches the database using the news article ID as a key, retrieves the content of the corresponding news article, and then sends the retrieved content of the news article to the user's terminal.

[0156] Step 2:

[0157] The user terminal displays the acquired news article.

[0158] Input: News article content

[0159] Data processing: Converting the content of news articles into a format that can be displayed in the user interface.

[0160] Output: The displayed news article

[0161] Specific operation: The user terminal displays the contents of the news article received from the server on the user interface, allowing the user to view the news article.

[0162] Step 3:

[0163] The server receives a news article as input and uses a generative AI model to generate comments for an artificially intelligent character based on a prompt sentence.

[0164] Input: News article content

[0165] Data processing: The content of the news article is embedded in a prompt sentence and input to the generative AI model, which then generates comments based on the prompt sentence.

[0166] Output: Comments from the AI ​​character

[0167] Specific operation: The server embeds the content of the news article into a prompt sentence and inputs it into the generative AI model (e.g., text-davinci-003). The generative AI model generates a comment for the AI ​​character based on the prompt sentence and sends the comment to the user's device.

[0168] Step 4:

[0169] The user terminal displays the generated comments and provides an interface through which the user can interact with the artificial intelligence character.

[0170] Input: AI character's comment

[0171] Data processing: Converting comments into a format that can be displayed in the user interface.

[0172] Output: Comments displayed

[0173] Specific operation: The user terminal displays the AI ​​character's comments received from the server on the user interface. The user can view the comments and start a dialogue.

[0174] Step 5:

[0175] The user interacts with an artificially intelligent character, and the user's input is sent to the server.

[0176] Input: User input

[0177] Data processing: Converting user input into a format to send to the server.

[0178] Output: User input sent to the server

[0179] Specific operation: The user inputs a response to the AI ​​character's comment. The user terminal sends the user's input to the server.

[0180] Step 6:

[0181] The server receives the user's input and generates a response, again using the generative AI model.

[0182] Input: User input

[0183] Data processing: The user's input is embedded in a prompt sentence and input to the generative AI model, which then generates a response based on the prompt sentence.

[0184] Output: AI character's response

[0185] Specific operation: The server embeds the user's input into a prompt sentence and inputs it into the generative AI model. The generative AI model generates a response for the AI ​​character based on the prompt sentence and sends the response to the user's device.

[0186] Step 7:

[0187] The user terminal displays the generated response and allows the user to continue the dialogue.

[0188] Input: AI character's response

[0189] Data processing: Converting responses into a format that can be displayed in the user interface.

[0190] Output: The displayed response

[0191] Specific operation: The user terminal displays the AI ​​character's response received from the server on the user interface, allowing the user to continue the dialogue.

[0192] Step 8:

[0193] The server receives the dialogue between the user and the AI ​​character as input and generates a summary using a generative AI model.

[0194] Input: Dialogue between the user and the AI ​​character

[0195] Data processing: The dialogue is embedded in a prompt and fed into a generative AI model, which then generates a summary based on the prompt.

[0196] Output: A summary of the conversation

[0197] Specific operation: The server embeds the dialogue between the user and the AI ​​character into a prompt sentence and inputs it into the generative AI model. The generative AI model generates a summary of the dialogue based on the prompt sentence and sends the summary to the user's device.

[0198] Step 9:

[0199] The user terminal displays the generated summary and allows the user to contribute.

[0200] Input: A summary of the conversation

[0201] Data processing: Converting the summary into a format that can be displayed in the user interface.

[0202] Output: Displayed summary

[0203] Specific operation: The user terminal displays the summary of the exchange received from the server on the user interface. The user can check the summary and post.

[0204] Example 2

[0205] Next, a description will be given of Example 2 of Form 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."

[0206] Conventional news comment systems lacked the support required for users to gain a deeper understanding of news articles, and were not designed to increase user retention. Furthermore, there was a need for an appropriate response to offensive posts.

[0207] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0208] In this invention, the server includes a means for acquiring news articles, a means for analyzing the acquired news articles using natural language processing technology, a means for generating comments and questions using a generative AI model based on the analysis results, and a means for analyzing user responses and generating appropriate responses using the generative AI model. This allows users to gain a deeper understanding of news articles and increase their stay time. It also allows for appropriate responses to offensive posts.

[0209] The "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[0210] An "artificial intelligence character" is a program that automatically generates comments and questions about news articles and interacts with users.

[0211] "A format similar to comments from general users" refers to comments and questions generated by AI characters that are in the same format and style as comments and questions posted by regular users.

[0212] "Means for making interactions visible only to the user" refers to a function that prevents the conversation between the user and the AI ​​character from being displayed to other users.

[0213] The "summary of the interaction" is information that briefly summarizes the content of the conversation between the user and the AI ​​character.

[0214] "Means for obtaining news articles" refers to a function for collecting the latest news articles from news sites and RSS feeds on the Internet.

[0215] "Natural language processing technology" is a technology that enables computers to understand and analyze human language.

[0216] A "generative AI model" is an algorithm that allows artificial intelligence to generate new text based on a given prompt.

[0217] The "means for analyzing user responses" is a function for analyzing the text entered by the user and understanding its content and emotions.

[0218] The "means for generating an appropriate response" is a function that enables an AI character to generate appropriate comments or questions based on the user's response.

[0219] This invention is a system that introduces an AI character into a news comment section and provides dialogue that allows users to gain a deeper understanding of a news article. Specific embodiments of this system are described below.

[0220] The server first retrieves the latest news articles from news sites or RSS feeds on the Internet. For example, the server retrieves the RSS feed from "https: / / example-news.com / rss" and extracts the latest articles. This process uses common web scraping techniques or RSS reader software.

[0221] The server then analyzes the content of the retrieved news articles using natural language processing technology. Specifically, it uses the Google® Cloud Natural Language API to analyze the article's topics, keywords, and sentiment. For example, the server analyzes the content of the articles and extracts keywords such as "small and medium-sized businesses," "bill," and "tax benefits."

[0222] Based on the analysis results, the server uses a generative AI model (e.g., OpenAI's GPT-4) to generate comments and questions. For example, the server generates the question, "How do you think this bill will affect small businesses?" An example of this prompt would be:

[0223] Analyze the content of a news article and generate questions such as:

[0224] "How do you think this bill will affect small businesses?"

[0225] The generated comments and questions are displayed to the user via the terminal. The terminal displays the question "How do you think this bill will affect small businesses?" below the news article. The user inputs their opinion in response to this question. For example, the user might input "I think this bill is beneficial to small businesses because it includes tax breaks."

[0226] The server receives the user's response and analyzes it again using natural language processing technology. This allows it to understand the user's opinions and emotions. Based on the analysis results, the server uses a generative AI model to generate an appropriate response. For example, the server might generate a response such as, "I see, tax incentives are important for small and medium-sized businesses. What else are you concerned about?"

[0227] This series of processes allows users to gain a deeper understanding of news articles and increase their viewing time. The AI ​​character can also respond appropriately to offensive posts. This system is an effective way for users to gain a deeper understanding of news articles and promote constructive dialogue.

[0228] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0229] Step 1: Get the news article

[0230] The server retrieves the latest news articles from news sites and RSS feeds on the Internet. Specifically, the server retrieves the RSS feed from "https: / / example-news.com / rss" and extracts the latest articles. The input is the RSS feed URL, and the output is the news article data. This process uses web scraping technology or RSS reader software.

[0231] Step 2: Analyzing the article content

[0232] The server analyzes the content of the retrieved news articles using natural language processing technology. Specifically, it uses the Google Cloud Natural Language API to analyze the article's topics, keywords, and sentiment. The input is the news article data, and the output is the analysis results (topic, keyword, and sentiment analysis results). For example, the server analyzes the content of the article and extracts keywords such as "small and medium-sized enterprises," "bill," and "tax incentives."

[0233] Step 3: Generate comments and questions

[0234] The server generates comments and questions using a generative AI model (e.g., OpenAI's GPT-4) based on the analysis results. The input is the analysis results, and the output is the generated comments and questions. For example, the server generates the question, "How do you think this bill will affect small businesses?" An example of this prompt is as follows:

[0235] Analyze the content of a news article and generate questions such as:

[0236] "How do you think this bill will affect small businesses?"

[0237] Step 4: Display to the user

[0238] The terminal displays comments and questions received from the server to the user. The input is the generated comment or question, and the output is the text displayed on the user's screen. For example, the terminal might display the question "How do you think this bill will affect small businesses?" below a news article.

[0239] Step 5: Getting the user's response

[0240] The user inputs their opinion in response to the displayed question. The input is the user's opinion, and the output is the user's response data. For example, the user might input, "I think this bill is beneficial to small businesses because it includes tax breaks."

[0241] Step 6: Parsing the user's response

[0242] The server receives the user's response and analyzes it again using natural language processing technology. The input is the user's response data, and the output is the analysis results (the analysis results of the user's opinions and emotions). This allows the user's opinions and emotions to be understood.

[0243] Step 7: Generating AI character reactions

[0244] The server uses a generative AI model to generate an appropriate response based on the results of the user's response analysis. The input is the user's response analysis result, and the output is the generated response. For example, the server generates a response such as, "I see, tax incentives are important for small and medium-sized businesses. What else are you concerned about?"

[0245] Step 8: View the reaction

[0246] The terminal displays the responses received from the server to the user. The input is the generated response, and the output is the text displayed on the user's screen, allowing the user to continue interacting with the AI ​​character.

[0247] This process allows users to gain a deeper understanding of news articles and increase their time spent on the site. It also enables the AI ​​character to respond appropriately to offensive posts.

[0248] (Application example 2)

[0249] Next, a description will be given of Application Example 2 of Form 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."

[0250] Conventional news comment systems lack support for users to gain a deeper understanding of news articles. It is also difficult for users to receive appropriate feedback when posting comments on news articles. Furthermore, they lack the means to increase user retention and lack sufficient countermeasures against offensive posts.

[0251] The specific processing by the specific 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 introducing an AI character into a news comment section, means for the AI ​​character to comment on the news in a format similar to that of general users, means for the user to interact with the AI ​​character about the news, means for the interaction to be visible only to the user, means for posting a "summary of the interaction" after the interaction is over, means for generating comments on the news article using a generative AI model, and means for providing an appropriate response to the user's opinion. This allows users to gain a deeper understanding of the news article and receive appropriate feedback. It also increases the user's stay time and strengthens measures against offensive posts.

[0252] A "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[0253] An "AI character" is a virtual character generated using artificial intelligence technology that can interact with users.

[0254] "A format similar to comments made by general users" refers to comments that are in the same format and style as comments posted by regular users.

[0255] "Means for commenting on news" refers to methods and functions for posting opinions and thoughts about news articles.

[0256] "Means by which users can interact with the AI ​​character about the news" refers to methods or functions that allow users to exchange opinions about news articles through dialogue with the AI ​​character.

[0257] "Means for ensuring that the interaction is visible only to the user" refers to methods or functions that prevent the content of the conversation between the user and the AI ​​character from being displayed to other users.

[0258] "Means for posting summaries of interactions" refers to methods or functions for posting summaries of the content of the conversation between the user and the AI ​​character.

[0259] A "generative AI model" is a model that uses artificial intelligence technology to generate text and data.

[0260] The "means for generating comments on news articles" refers to a method or function for generating appropriate comments based on the content of a news article.

[0261] "Means of responding appropriately to user opinions" refers to methods and functions that allow AI characters to respond appropriately to opinions posted by users.

[0262] The system for implementing this invention introduces an AI character into the comment section of a news article, allowing users to post comments on the news article and interact with the AI ​​character. A specific embodiment of this system is described below.

[0263] System configuration

[0264] The system consists of a server and a user terminal. The server contains a news article database, a generative AI model, a comment generation module, and a dialogue management module. The user terminal is a device such as a smartphone or PC with a news application installed.

[0265] Program processing

[0266] The server retrieves articles from a database of news articles and delivers them to the user's device. When the user reads a news article, the AI ​​character generates a comment based on the article's content. This comment generation uses a generative AI model (e.g., OpenAI's text-davinci-003).

[0267] The generated comments are displayed on the user's device, allowing the user to initiate a dialogue with the AI ​​character. The dialogue management module receives the user's input and sends an appropriate prompt to the generative AI model. The generative AI model then generates an appropriate response to the user's opinion and returns it to the user's device.

[0268] Hardware and software used

[0269] Hardware: Servers, smartphones, PCs

[0270] Software: News application, generative AI model (OpenAI API)

[0271] Data processing and calculation

[0272] The server inputs the contents of the news article as a prompt into the generative AI model to generate a comment. It also generates a prompt for the user's opinion, which is then input into the generative AI model to generate an appropriate response.

[0273] Specific examples

[0274] For example, if the news article states, "The government has introduced a new tax reform bill. This bill aims to provide special tax deductions for small and medium-sized businesses," the following prompt sentence will be generated:

[0275] Comment generation for news articles

[0276] News article: The government has introduced a new tax reform bill, which aims to provide special tax breaks for small and medium-sized businesses.

[0277] How do you think this bill will affect small businesses?

[0278] Generate responses to user opinions

[0279] User Opinion: I think this bill will be very beneficial for small businesses.

[0280] AI character response:

[0281] This allows users to gain a deeper understanding of news articles and receive appropriate feedback, increases user retention, and strengthens measures against offensive posts.

[0282] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0283] Step 1:

[0284] The server retrieves the latest news articles from a news article database. The input is the news article database, and the output is the retrieved news article. Specifically, the server executes a database query to retrieve the content of the news article.

[0285] Step 2:

[0286] The server delivers the retrieved news articles to the user terminal. The input is the retrieved news article, and the output is the news article displayed on the user terminal. Specifically, the server sends the news article to the user terminal as an HTTP response.

[0287] Step 3:

[0288] The user terminal displays the news article. The input is the news article sent from the server, and the output is the news article that the user can view. Specifically, the user terminal displays the news article on the screen through a news application.

[0289] Step 4:

[0290] The server generates a prompt based on the content of the news article. The input is the content of the news article, and the output is the generated prompt. Specifically, the server analyzes the content of the news article and generates a prompt containing an appropriate question.

[0291] Step 5:

[0292] The server inputs a prompt sentence into the generative AI model and generates a comment. The input is the generated prompt sentence, and the output is the generated comment. Specifically, the server sends an API request to the generative AI model (e.g., OpenAI's text-davinci-003) to generate a comment.

[0293] Step 6:

[0294] The server sends the generated comment to the user terminal. The input is the generated comment, and the output is the comment displayed on the user terminal. In concrete terms, the server sends the comment to the user terminal as an HTTP response.

[0295] Step 7:

[0296] The user terminal displays the generated comments. The input is the comment sent from the server, and the output is the comment that the user can view. Specifically, the user terminal displays the comment on the screen through a news application.

[0297] Step 8:

[0298] The user initiates a dialogue with the AI ​​character. The input is the user's opinion or question, and the output is the input to the AI ​​character. Specifically, the user enters text through a news application.

[0299] Step 9:

[0300] The server generates a prompt based on the user's input. The input is the user's opinion or question, and the output is the generated prompt. Specifically, the server analyzes the user's input and generates a prompt with an appropriate response.

[0301] Step 10:

[0302] The server inputs a prompt sentence into the generative AI model and generates a response. The input is the generated prompt sentence, and the output is the generated response. Specifically, the server sends an API request to the generative AI model to generate a response.

[0303] Step 11:

[0304] The server sends the generated response to the user terminal. The input is the generated response, and the output is the response displayed on the user terminal. Specifically, the server sends the response to the user terminal as an HTTP response.

[0305] Step 12:

[0306] The user device displays the generated responses. The input is the response sent from the server, and the output is the response that the user can view. Specifically, the user device displays the responses on the screen through a news application.

[0307] Example 3

[0308] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0309] The challenges facing news sites are increasing the amount of time users spend on the site and preventing conflicts between users. In particular, when users comment on news stories, aggressive comments are likely to occur, which can cause conflicts between users. Furthermore, users tend to spend less time on the site due to a lack of means for them to gain a deeper understanding of the news content.

[0310] The identification process by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means. In this invention, the server includes a means for introducing an AI character into a comment section of a news item, a means for the AI ​​character to comment on the news item in a format similar to that of general users, a means for the user to interact with the AI ​​character about the news item, a means for the interaction to be visible only to the user, a means for the user to post a "summary of the interaction" after the interaction has ended, a means for sending the user's input as a prompt to the generative AI model, a means for the generative AI model to generate a response based on the prompt, and a means for displaying the generated response to the user. This allows users to gain a deeper understanding of the news content through dialogue with the AI ​​character, thereby extending their stay on the site. Furthermore, the AI ​​character's role as a receptacle for offensive posts can prevent conflicts between users.

[0311] A "news comment section" is an area on a news site where users can post their opinions and thoughts about a news article.

[0312] An "artificial intelligence character" is a virtual character generated using artificial intelligence technology, and is capable of interacting with the user.

[0313] "A format similar to comments made by general users" refers to a format or style similar to that of comments posted by regular users.

[0314] A "prompt sentence" is an input sentence that is generated based on a user's input and causes an artificial intelligence model to generate a response.

[0315] A "generative artificial intelligence model" is a model that uses artificial intelligence techniques to generate an appropriate response based on a prompt sentence.

[0316] A "response" is a reply or reaction generated by an AI character in response to a user's input.

[0317] The "summary of the interaction" is a concise summary of the content of the conversation between the user and the AI ​​character.

[0318] An "offensive post" is a comment that expresses discomfort or hostility toward another user.

[0319] A "receptacle" is something that plays a role in accepting and processing specific content or situations.

[0320] The present invention is a system for increasing the time users spend on a news site and preventing trouble between users. A specific embodiment of this system will be described below.

[0321] First, a user accesses a news site using a web browser. The homepage of the news site displays news articles along with a chat window with an AI character. Through this chat window, the user can enter questions and comments to the AI ​​character.

[0322] When a user types something into the chat window, the device sends it to a server, which then formats it into a prompt and sends it to a generative AI model, such as OpenAI's GPT-4.

[0323] The generative AI model generates an appropriate response based on the prompt sentence. The generated response is returned to the server, which then sends it to the device, which displays the received response in a chat window for the user to see.

[0324] As a concrete example, consider the case where a user inputs "Tell me about today's news." This input becomes a prompt sentence, which the server sends to the generative AI model. The generative AI model generates a response such as "Today's news reports that economic growth is progressing better than expected," and returns it to the server. The server sends this response to the device, which then displays it to the user.

[0325] An example of a prompt is as follows:

[0326] "How's the weather today?"

[0327] "Please tell me the latest sports news."

[0328] "What are the latest technology trends?"

[0329] This system allows users to gain a deeper understanding of the news content through dialogue with an AI character. Furthermore, the AI ​​character can act as a receptacle for offensive posts, preventing conflict between users. This increases the time users spend on news sites and improves user satisfaction.

[0330] This system is also beneficial for news site operators, as it can increase user engagement. Furthermore, by utilizing artificial intelligence technology, it becomes possible to provide efficient user support. The flow of the identification process in Example 3 will be described with reference to FIG. 15.

[0331] Step 1:

[0332] A user visits a news site.

[0333] A user accesses a news site by entering its URL using a web browser. The input is the user's access request, and the output is the news site's homepage, which displays news articles and a chat window with an AI character.

[0334] Step 2:

[0335] The user inputs questions and comments to the AI ​​character.

[0336] The user types a question or comment into the chat window, for example, "Tell me about today's news." The input is the user's text data, and the output is the user's input displayed on the terminal.

[0337] Step 3:

[0338] The terminal sends the user's input to the server.

[0339] The terminal sends the user's input to the server as text data. The input is the user's text data, and the output is the text data sent to the server. In concrete terms, the terminal sends data using an HTTP request.

[0340] Step 4:

[0341] The server sends the user's input as a prompt to the generative AI model.

[0342] The server formats the received user input as a prompt sentence and sends it to the generative AI model. The input is the user's text data, and the output is the prompt sentence sent to the generative AI model. Specifically, the server converts the text data into an appropriate format and sends it to the generative AI model using an API request.

[0343] Step 5:

[0344] A generative AI model generates a response based on the prompt.

[0345] The generative AI model generates an appropriate response based on the prompt sentence. The input is the prompt sentence, and the output is the generated response text. Specifically, the generative AI model uses a natural language processing algorithm to generate the response.

[0346] Step 6:

[0347] The server receives the generated response and sends it to the terminal.

[0348] The server sends the response received from the generative AI model to the terminal. The input is the generated response text, and the output is the response text sent to the terminal. Specifically, the server sends data using an HTTP response.

[0349] Step 7:

[0350] The terminal displays the response to the user.

[0351] The terminal displays the response received from the server in a chat window. The input is the response text received from the server, and the output is the response text displayed to the user. Specifically, the terminal renders the received data on the screen.

[0352] Step 8:

[0353] The user either inputs information to the AI ​​character again or browses the site.

[0354] The user continues to interact with the AI ​​character or browses other content on the news site. The input is the user's new text data or navigation on the site, and the output is new dialogue or new content displayed. Specifically, the user may enter more text into the chat window or click a link on the site.

[0355] (Application example 3)

[0356] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0357] Conventional news sites and virtual stores lacked the ingenuity to increase the time users spent on the site, resulting in users leaving the site after a short time. Furthermore, disputes between users and offensive comments were common, making it difficult to maintain sound site management. Furthermore, virtual stores lacked sufficient product explanations and responses to questions, making it difficult to stimulate users' desire to purchase.

[0358] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for introducing an AI character into a comment section of a news article, means for the AI ​​character to comment on the news article in a format similar to that of general users, means for the user to interact with the AI ​​character about the news article, means for the interaction to be visible only to the user, means for the user to post a "summary of the interaction" after the interaction has ended, means for the user to interact with the AI ​​character in a virtual store and receive product explanations and answer questions, and means for the AI ​​character to acknowledge an offensive comment made by the user and prevent trouble with other users. This increases the user's stay time, prevents trouble between users, and improves product explanations and question responses in the virtual store.

[0359] An "artificial intelligence character" is a computer program designed to interact with a user and respond to user input using natural language processing techniques.

[0360] A "virtual store" is a virtual store that exists on the Internet and provides an environment where users can browse and purchase products online.

[0361] A "comment section" is an area on web pages such as news sites and virtual stores where users can post their opinions and thoughts.

[0362] "Interaction" refers to the dialogue and communication that takes place between the user and the AI ​​character.

[0363] The "interaction summary" is a concise summary of the content of the conversation between the user and the AI ​​character.

[0364] "Aggressive comments" are comments posted with hostility or malice towards other users or the system.

[0365] "Preventing trouble" means preventing conflicts and disputes between users before they occur.

[0366] "Product explanation" refers to an AI character in a virtual store explaining the features and benefits of a product to the user.

[0367] "Question response" refers to providing appropriate answers to questions posed by users to AI characters.

[0368] As an embodiment of the present invention, a dialogue system using an AI character is constructed for a news site or a virtual store. A specific embodiment of the system is shown below.

[0369] System configuration

[0370] This system includes a server, a user terminal, and an AI character. The server manages data from news sites and virtual stores and communicates with the user terminal. The user terminal is a device such as a smartphone or head-mounted display, and provides an interface for users to interact with the system. The AI ​​character generates responses to user input using a generative AI model.

[0371] Program processing

[0372] The server executes a program that includes the following means:

[0373] 1. Introducing AI characters:

[0374] The server displays an AI character in the comment section of the news article or in a virtual store, and this character acts as a conversation partner when a user posts a comment.

[0375] 2. Comment generation method:

[0376] The server uses a generative AI model to have AI characters comment on news in a format similar to that of regular users, using natural language processing technologies such as the OpenAI API.

[0377] 3. Means of interaction:

[0378] Users can interact with AI characters about news and product information. The generative AI model generates appropriate responses to user input and displays them on the user's device.

[0379] 4. Private Communication Methods:

[0380] The interactions between users and AI characters are not visible to other users, thereby protecting privacy.

[0381] 5. Summary submission method:

[0382] After a conversation is over, users can post a "conversation summary," which is made publicly available to other users.

[0383] 6. Receptacles for offensive comments:

[0384] If a user makes an offensive comment, an AI character will receive the comment and prevent trouble with other users.

[0385] Hardware and software used

[0386] Hardware: Smartphone, head-mounted display, camera

[0387] Software: OpenAI API, OpenCV

[0388] Specific examples

[0389] When a user asks, "Tell me about this product" in the virtual store, an artificial intelligence character responds, "This product is the latest smartphone with excellent camera performance."

[0390] Example prompt sentence:

[0391] User: Tell me about this product.

[0392] Artificial Intelligence Character: This product is a state-of-the-art smartphone with an excellent camera.

[0393] In this way, users can enjoy a shopping experience in a virtual store while obtaining product information through dialogue with an AI character.

[0394] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0395] Step 1:

[0396] The server manages the data of news sites and virtual stores and communicates with user terminals. When a user accesses a news site or virtual store, the server sends the data of the corresponding page to the user terminal. This causes the news article or product information to be displayed on the user terminal.

[0397] Input: User's access request

[0398] Output: News articles and product information data

[0399] Step 2:

[0400] The user terminal displays the AI ​​character in the comment section of the news and in the virtual store. When the user posts a comment, the AI ​​character appears as a conversation partner.

[0401] Input: Data sent from the server

[0402] Output: Display of AI character

[0403] Step 3:

[0404] When a user enters a comment, the user's device sends the comment to the server, which uses a generative AI model to generate a response for the AI ​​character to comment on the news in a format similar to that of a regular user.

[0405] Input: User comment

[0406] Output: Response from the generative AI model

[0407] Step 4:

[0408] The server transmits the generated response to the user terminal, which displays the response of the AI ​​character in a comment field.

[0409] Input: Response by a generative AI model

[0410] Output: Response display of AI character

[0411] Step 5:

[0412] The user interacts with the AI ​​character about news and product information. The generative AI model generates an appropriate response to the user's input and sends it to the user's device via the server.

[0413] Input: User input

[0414] Output: Response from the generative AI model

[0415] Step 6:

[0416] The server uses private interaction methods to ensure that the user's interactions with the AI ​​character are not visible to other users, thereby protecting the user's privacy.

[0417] Input: Interaction between the user and the AI ​​character

[0418] Output: Realizing private dialogue

[0419] Step 7:

[0420] After a conversation is over, users can post a "conversation summary," which the server then makes available to other users.

[0421] Input: User Summary

[0422] Output: Publish summary

[0423] Step 8:

[0424] If a user makes an offensive comment, the server will have an AI character receive the comment and take measures to prevent trouble with other users. Using a generative AI model, the server will generate an appropriate response and send it to the user's device.

[0425] Input: offensive comment

[0426] Output: An appropriate response from the generative AI model

[0427] In this way, users can enjoy the experience of browsing news sites or virtual stores while obtaining information through dialogue with AI characters. This also prevents trouble caused by offensive comments.

[0428] 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.

[0429] "Example 1"

[0430] One embodiment of the present invention is a system incorporating an emotion engine. This system recognizes the user's emotions and adjusts the AI ​​character's response based on those emotions. Specifically, when the user expresses negative emotions such as anger or sadness, the AI ​​character accepts those emotions and responds in a way that soothes the user. Furthermore, when the user expresses positive emotions such as joy or excitement, the AI ​​character shares those emotions and responds in a way that amplifies the user's joy.

[0431] "Example 2"

[0432] The emotion engine also supports the generation of "interaction summaries." Specifically, it analyzes the user's emotions expressed during their interactions with the AI ​​character and generates a summary that reflects those emotions. For example, if the user expresses strong joy during their interaction with the AI ​​character, the summary will reflect the user's emotions in the form of "The user was very happy about the news."

[0433] "Example 3"

[0434] Furthermore, the emotion engine tracks changes in the user's emotions and uses that information to adjust the AI ​​character's responses over time. For example, if the user initially shows anger but gradually calms down as they interact with the AI ​​character, the AI ​​character will pick up on this change and adjust its own responses to match the user's changing emotions. This allows the AI ​​character to have more natural, human-like conversations.

[0435] The processing flow of each embodiment will be described below.

[0436] "Example 1"

[0437] Step 1: A user posts a comment on a news article.

[0438] Step 2: The emotion engine recognizes emotions from user comments.

[0439] Step 3: The AI ​​character generates a response based on the information from the emotion engine and posts a comment to the user.

[0440] "Example 2"

[0441] Step 1: After the user finishes interacting with the AI ​​character, the emotion engine analyzes the user's emotions from the interaction.

[0442] Step 2: The emotion engine generates a "summary of the interaction" based on the analysis results.

[0443] Step 3: The generated "interaction summary" is presented to the user.

[0444] "Example 3"

[0445] Step 1: When the user begins interacting with the AI ​​character, the emotion engine tracks changes in the user's emotions.

[0446] Step 2: The emotion engine communicates the user's emotional changes to the AI ​​character.

[0447] Step 3: The AI ​​character adjusts its response based on the information from the emotion engine and posts a comment to the user.

[0448] Example 1

[0449] Next, a description will be given of Example 1 of Form 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."

[0450] On traditional news sites, when users post comments on news articles, their interactions with other users are made public, which can lead to privacy issues and abusive comments. It is also difficult to provide appropriate responses based on users' emotions, which can hinder user satisfaction. Furthermore, there are a lack of ways to deepen users' understanding of the news, making it difficult to increase user time spent on the site.

[0451] 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.

[0452] In this invention, the server includes means for introducing an AI character into a comment section of a news article, means for the AI ​​character to comment on the news article in a format similar to that of general users, means for the user to interact with the AI ​​character about the news article, means for the interaction to be visible only to the user, means for the user to post a "summary of the interaction" after the interaction has ended, means for recognizing the user's emotions and adjusting the AI ​​character's response based on the emotions, and means for storing the generated comments and responses in a database. This makes it possible to provide appropriate responses according to the user's emotions, deepen the user's understanding of the news, and increase the user's stay time while protecting the user's privacy.

[0453] An "artificial intelligence character" is a virtual character equipped with artificial intelligence designed to interact with users and generate comments.

[0454] The "comment section" is an interface where users can post their opinions and thoughts about news articles.

[0455] "A format similar to comments made by ordinary users" means that comments generated by AI characters have the same style and content as comments posted by ordinary users.

[0456] An "interaction" is an exchange of messages between a user and an AI character.

[0457] The "summary of the interaction" is a sentence that briefly summarizes the content of the interaction between the user and the AI ​​character.

[0458] "Emotion recognition" means analyzing and identifying emotions from text entered by the user.

[0459] "Adjusting responses" means changing the response content and attitude of an AI character based on the user's recognized emotions.

[0460] A "generative AI model" is an artificial intelligence model that uses natural language processing to generate text.

[0461] The "database" is an information management system for storing generated comments, reactions, and interactions with users.

[0462] This invention is a system that introduces an AI character into a news site, deepens users' understanding of news articles through dialogue with them, and increases the time users spend on the site. Specific embodiments of this system are described below.

[0463] server

[0464] The server uses the news API to retrieve the latest news articles and store them in a database. Specifically, the server sends a request to the news API every hour and stores the retrieved news articles in a database.

[0465] The server uses a generative AI model (e.g., GPT-4) to generate a comment from an AI character in response to a user's comment. When a user enters a comment on a news article, the server sends the comment to the generative AI model and returns the generated comment to the user.

[0466] The server uses sentiment analysis software (e.g., IBM Watson® Tone Analyzer) to recognize the user's emotions. It sends the text entered by the user to the sentiment analysis software and receives the results of the sentiment analysis.

[0467] The server uses a generative AI model to generate a response for the AI ​​character based on the results of the emotion analysis. For example, if a user inputs "I'm feeling very sad today," the server uses emotion analysis software to recognize the emotion of "sadness" and sends the result to the generative AI model to generate a response such as "That's terrible. Is there anything you'd like to talk about?"

[0468] The server stores the interactions between users and AI characters in a database and manages private interactions for each user.

[0469] Terminal

[0470] The device provides a user interface that allows users to view news articles and post comments. When a user visits a news site, the device displays a list of the latest news articles.

[0471] The terminal displays a comment input field and a chat window with an AI character for each news article. When a user inputs a comment for a news article, the comment is sent to the server.

[0472] The device displays the AI ​​character's responses in real time. For example, if a user types, "What do you think about the latest economic news?", a comment generated using the generative AI model will be displayed on the device.

[0473] User

[0474] Users browse news articles and begin interacting with the AI ​​character, who then comments on the article using the generative AI model.

[0475] Users enter text to express their feelings, for example, "I'm feeling very sad today," and emotion analysis software is used to recognize that emotion and generate an appropriate response.

[0476] After the user has finished interacting with the AI ​​character, they can enter and post a summary of the interaction. For example, they could write, "Through my interaction with the AI ​​character, I gained a deeper understanding of the current state of the economy."

[0477] Specific examples

[0478] The user asks the AI ​​character, "What do you think about recent economic news?" The generative AI model responds by saying, "Experts are divided on the current state of the economy, but recent data shows signs of recovery."

[0479] If a user types, "I'm feeling very sad today," the sentiment analysis software will recognize the emotion of "sadness," and the generative AI model will generate a response like, "That's terrible. Is there anything you'd like to talk about?"

[0480] Prompt Sentence Examples

[0481] "What do you think about the recent economic news?"

[0482] "I feel very sad today."

[0483] The above is a specific embodiment for carrying out the present invention. This system protects the user's privacy while providing appropriate responses according to their emotions, deepening the user's understanding of the news and increasing the time spent on the site.

[0484] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0485] Step 1: Get the news article

[0486] server

[0487] The server uses the news API to get the latest news articles. Specifically, the server sends a request to the news API every hour and stores the retrieved news articles in a database.

[0488] Input: A request to the News API

[0489] Output: Retrieved news article data

[0490] Specific operation: The server sends a request to the news API and stores the retrieved news articles in the database.

[0491] Step 2: Displaying the User Interface

[0492] Terminal

[0493] The device displays a list of news articles to the user. When the user visits a news site, the device displays a list of the latest news articles.

[0494] Input: News articles retrieved from a database

[0495] Output: A list of news articles in a user interface

[0496] Specific operation: The device retrieves news articles from the database and displays them in the user interface.

[0497] Step 3: User comments

[0498] User

[0499] A user inputs a comment on a news article. When a user inputs a comment on a news article, the comment is sent to the server.

[0500] Input: User comment

[0501] Output: Comment data sent to the server

[0502] Specific operation: The user enters a comment on a news article, such as "What do you think about this news?", and presses the send button.

[0503] Step 4: Generate comments for the AI ​​character

[0504] server

[0505] The server sends the user's comments to the generative AI model, which generates comments for the AI ​​character, and returns the generated comments to the user.

[0506] Input: User comment data

[0507] Output: Comments generated by the generative AI model

[0508] Specific operation: The server sends the user's comment "What do you think about recent economic news?" to the generative AI model, which then returns the generated comment "Experts are divided on the current state of the economy, but recent data shows signs of recovery" to the user.

[0509] Step 5: Saving the interaction between the user and the AI ​​character

[0510] server

[0511] The server stores the interactions between users and AI characters in a database and manages private interactions for each user.

[0512] Input: Interaction data between the user and the AI ​​character

[0513] Output: Interaction data stored in a database

[0514] Specific operation: The server stores the interactions between the user and the AI ​​character in a database so that the interactions can be displayed when the user accesses the site again.

[0515] Step 6: Post a summary of the conversation

[0516] User

[0517] After the user has finished interacting with the AI ​​character, they can enter a "summary of the interaction" and post it.

[0518] Input: A summary of the user interaction

[0519] Output: Posted summary data

[0520] Specific actions: The user enters a summary such as "Through my interactions with the AI ​​character, I gained a deeper understanding of the current state of the economy," and presses the post button.

[0521] Step 7: Recognize emotions

[0522] server

[0523] The server sends the user's input text to the sentiment analysis software and receives the results of the sentiment analysis.

[0524] Input: User-entered text

[0525] Output: Sentiment analysis result data

[0526] Specific operation: The server sends the user's input "I feel very sad today" to the sentiment analysis software and receives the result of the sentiment analysis as "sadness."

[0527] Step 8: Generating AI character reactions

[0528] server

[0529] The server uses a generative AI model to generate the AI ​​character's reaction based on the results of the emotion analysis.

[0530] Input: Sentiment analysis result data

[0531] Output: The response generated by the generative AI model

[0532] Specific operation: The server sends the emotion analysis result of "sadness" to the generation AI model, which generates a response saying, "That's terrible. Is there anything you'd like to say?"

[0533] Step 9: Showing User Feedback

[0534] Terminal

[0535] The device displays the AI ​​character's reactions to the user.

[0536] Input: Responses generated by a generative AI model

[0537] Output: AI character response on the user interface

[0538] What happens: The device displays to the user, "That's a problem. Is there anything else you'd like to talk about?"

[0539] (Application example 1)

[0540] Next, a description will be given of Application Example 1 of Embodiment 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."

[0541] Traditional news sites lack support for users to gain a deeper understanding of the content of news articles when they read them. It is also difficult for users to receive appropriate feedback based on their emotions toward the news, limiting the user experience. Furthermore, there is a lack of ingenuity to increase the amount of time users spend on news sites, which can lead to a decline in site usage. To address these issues, a system is needed that allows users to gain a deeper understanding of news articles and receive appropriate feedback based on their emotions.

[0542] 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.

[0543] In this invention, the server includes: means for introducing an AI character into a news comment section; means for the AI ​​character to comment on the news in a format similar to that of general users; means for the user to interact with the AI ​​character about the news; means for the interaction to be visible only to the user; means for the user to post a "summary of the interaction" after the interaction is over; means for recognizing the user's emotions using an emotion engine and adjusting the AI ​​character's response based on the emotions; means including an application to be installed on a smartphone; means for generating comments on news articles using a generative AI model; and means for inputting prompt sentences into the generative AI model. This allows users to gain a deeper understanding of news articles and receive appropriate feedback based on their emotions. It also increases user time spent on the site, improving site usage.

[0544] A "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[0545] An "AI character" is a character that interacts with users using artificial intelligence.

[0546] "A format similar to comments made by general users" means a format or style similar to that of comments posted by regular users.

[0547] "Means by which the user can interact with the AI ​​character about the news" refers to a function that allows the user to have a dialogue with the AI ​​character about news articles.

[0548] "Means for ensuring that the interaction is visible only to the user" refers to a function that prevents the content of the conversation between the user and the AI ​​character from being displayed to other users.

[0549] The "interaction summary" is a concise summary of the conversation between the user and the AI ​​character.

[0550] An "emotion engine" is a technology that recognizes a user's emotions and generates an appropriate response based on those emotions.

[0551] An "application installed on a smartphone" is a software program that runs on a smartphone.

[0552] A "generative AI model" is a model that uses artificial intelligence to generate text and data.

[0553] A "prompt" is a text instruction that is input to a generative AI model.

[0554] A system for implementing this invention introduces an AI character into the comment section of a news article, allowing users to interact with the AI ​​character regarding the news article. A specific embodiment of this system will be described below.

[0555] System configuration

[0556] Hardware

[0557] Server: Manages the news article database and generates comments and recognizes emotions for the AI ​​characters.

[0558] Smartphone: A device that allows users to view news articles and interact with AI characters.

[0559] software

[0560] Generative AI model: An artificial intelligence model for generating comments on news articles. Specifically, it uses OpenAI's API.

[0561] Emotion Engine: Software to recognize user emotions and adjust the AI ​​character's response based on those emotions. EmotionRecognizer is used.

[0562] Smartphone application: An application that provides features such as displaying news articles, interacting with AI characters, emotion recognition, and comment generation.

[0563] System Operation

[0564] 1. News article retrieval: The server retrieves the article from the news article database and sends it to the smartphone.

[0565] 2. Comment generation: The smartphone application uses the generative AI model to generate a comment for the AI ​​character on the news article. The generative AI model is given the following prompt:

[0566] Example prompt: "News article: A new technology has been announced. AI character comments:"

[0567] 3. Emotion Recognition: Smartphone applications use emotion engines to recognize emotions from user input and facial expressions.

[0568] 4. Comment adjustment: Adjust the AI ​​character's comments based on the user's recognized emotions. For example, if the user is "happy," add a positive element to the comment, such as "That's great!"

[0569] 5. Displaying dialogue: The adjusted comments are displayed on the smartphone screen, allowing users to interact with the AI ​​character. This dialogue is set to be invisible to other users.

[0570] 6. Interaction Summary: After a conversation has ended, users can post a summary of the conversation.

[0571] Specific examples

[0572] News article: "New technology announced."

[0573] User sentiment: "happy"

[0574] AI character comment: "Amazing! This technology could change the future."

[0575] This system allows users to gain a deeper understanding of news articles and receive appropriate feedback based on their emotions, while also increasing user time on the site and improving site usage.

[0576] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0577] Step 1:

[0578] The server retrieves the latest news articles from a news article database and sends them to the smartphone.

[0579] Input: A database of news articles

[0580] Output: News article data

[0581] What it does: The server queries the database to retrieve the latest news articles, then sends them to the smartphone.

[0582] Step 2:

[0583] The smartphone application uses a generative AI model to generate AI character comments on news articles.

[0584] Input: News article data, prompt

[0585] Output: AI character's comment

[0586] How it works: The smartphone application inserts the content of the news article into a prompt sentence and inputs it into the generative AI model, which then generates a comment for the AI ​​character based on the prompt sentence.

[0587] Step 3:

[0588] Smartphone applications use emotion engines to recognize emotions from user input and facial expressions.

[0589] Input: User input data, user facial expression data

[0590] Output: User sentiment data

[0591] Specific operation: The smartphone application inputs the user's text input and facial expression data acquired through the camera into the emotion engine, which then analyzes this data and recognizes the user's emotions.

[0592] Step 4:

[0593] The smartphone application adjusts the AI ​​character's comments based on the user's perceived emotions.

[0594] Input: AI character's comments, user's emotional data

[0595] Output: Adjusted AI character's comments

[0596] How it works: The smartphone application adjusts the AI ​​character's comments based on the user's emotional data. For example, if the user is "happy," it will add a positive element to the comment, such as "That's great!"

[0597] Step 5:

[0598] The smartphone application displays tailored comments on the screen, allowing users to interact with the AI ​​character.

[0599] Input: Adjusted AI character's comment

[0600] Output: Comments displayed on the smartphone screen

[0601] How it works: The smartphone application displays tailored comments in the user interface, allowing users to interact with the AI ​​character, while keeping this interaction invisible to other users.

[0602] Step 6:

[0603] Users can post summaries of their interactions after they finish.

[0604] Input: Dialogue between the user and the AI ​​character

[0605] Output: A summary of the conversation

[0606] How it works: The smartphone application summarizes the conversation between the user and the AI ​​character and allows the user to post those summaries.

[0607] Example 2

[0608] Next, a description will be given of Example 2 of Form 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."

[0609] Conventional news comment systems make it difficult for users to gain a deep understanding of a news article when posting comments. Furthermore, users are unable to receive appropriate feedback on their posted comments, which reduces the time users spend on the site. Furthermore, there is a risk of offensive comments being posted. To address these issues, a system is needed that deepens users' understanding of a news article, provides appropriate feedback, and increases the time users spend on the site.

[0610] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0611] In this invention, the server includes means for introducing an AI character into a comment section of a news article, means for the AI ​​character to comment on the news in a format similar to that of general users, means for the user to interact with the AI ​​character about the news, means for the interaction to be visible only to the user, means for the user to post a "summary of the interaction" after the interaction is over, means for the AI ​​character to analyze the news article using natural language processing technology and generate comments, and means for an emotion engine to analyze the user's emotions and generate a summary that reflects those emotions. This allows the user to deepen their understanding of the news article, provide appropriate feedback, and increase the user's stay time.

[0612] The "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[0613] An "artificial intelligence character" is a virtual entity that interacts with users using natural language processing technology.

[0614] "A format similar to comments made by general users" refers to comments that have the same format and writing style as comments posted by regular users.

[0615] "Interaction" refers to the dialogue and communication that takes place between the user and the AI ​​character.

[0616] The "summary of the interaction" is a sentence that briefly summarizes the content of the interaction between the user and the AI ​​character.

[0617] "Natural language processing technology" is a technology that enables computers to understand and generate human language.

[0618] An "emotion engine" is an algorithm or system that analyzes a user's emotions and generates information that reflects those emotions.

[0619] A "news article" is a piece of text that describes current events or information provided by a news organization or news site.

[0620] "Generating comments" means that an AI character automatically creates opinions and questions about news articles.

[0621] "Analyzing emotions" means identifying and understanding emotions from a user's statements and actions.

[0622] The present invention provides a system that deepens a user's understanding of a news article, provides appropriate feedback, and increases the user's stay time. A specific embodiment of this system is described below.

[0623] The server collects the latest news articles from news sites. Specifically, it obtains article data using RSS feeds and APIs. The collected news articles are analyzed using natural language processing technology. Generative AI models used here include BERT and GPT-3 (registered trademark).

[0624] The device provides an interface for users to enter comments or questions about news articles. Once the user has completed their input, the data is sent to the server. The server receives the user's input and uses a generative AI model to generate appropriate comments. The generated comments are then displayed to the user via the device.

[0625] When a user responds to a comment from an AI character, that data is also sent to the server. The server receives the user's response and uses the generative AI model to generate an appropriate response, which is also displayed to the user via their device.

[0626] The emotion engine analyzes the user's interactions with the AI ​​character and extracts the user's emotions. The emotion engine uses a sentiment analysis algorithm to identify the user's emotions and outputs the results as a summary. For example, if a user says, "I think this bill will be very beneficial to small and medium-sized businesses," the emotion engine extracts the emotion of joy and generates the summary, "The user was very happy about this news."

[0627] As a concrete example, the following exchange can be considered:

[0628] The news article read: "A new bill has been decided to introduce tax incentives for small and medium-sized enterprises."

[0629] AI character says: "How do you think this bill will affect small businesses?"

[0630] User Response: "I think this bill will be very beneficial to small businesses. The tax incentives will help ease their financial burden."

[0631] AI character responds: "Yes, tax incentives can encourage small business growth."

[0632] Sentiment Engine Summary: "Users were very happy about the news."

[0633] In this way, the system generates summaries of comments and emotions about news articles in cooperation with the server, terminals, and users.

[0634] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0635] Step 1:

[0636] The server collects the latest news articles from news sites. Specifically, it retrieves article data using RSS feeds and APIs. It uses the news site's URL and API key as input and obtains news article data as output. For example, a scheduled task can be run every morning at 8:00 to collect the latest political news from major news sites.

[0637] Step 2:

[0638] The server analyzes the collected news articles using natural language processing technology. Specifically, it uses generative AI models such as BERT and GPT-3 to understand the content of the articles. It uses the news article data as input and extracts the article's summary and main topics as output. For example, it generates a summary such as "It has been decided that a new bill will introduce tax incentives for small and medium-sized enterprises."

[0639] Step 3:

[0640] The terminal provides an interface for users to input comments and questions about news articles. It receives a news article summary and the user's comments and questions as input, and sends the data to the server as output. For example, it displays a news article summary to the user and provides an input field for "What do you think about this news?"

[0641] Step 4:

[0642] The server receives input from the user and generates appropriate comments using a generative AI model. It uses the user's comments or questions as input and gets the generated comments as output. For example, if the user enters "I think this bill will be very beneficial to small businesses," the server generates the comment "How do you think this bill will affect small businesses?"

[0643] Step 5:

[0644] The terminal displays the AI ​​character's comments to the user and provides an interface for the user to reply to them. The terminal receives the AI ​​character's comments and the user's responses as input, and sends the data back to the server as output. For example, the AI ​​character's comments may be displayed to the user, and the user may reply, "The introduction of tax incentives will help reduce the economic burden."

[0645] Step 6:

[0646] The server receives the user's response and generates an appropriate response using a generative AI model. It uses the user's response as input and gets the generated response as output. For example, based on the user's response, it generates the response, "Yes, tax incentives can encourage the growth of small and medium-sized businesses."

[0647] Step 7:

[0648] The server passes the user's interactions with the AI ​​character to an emotion engine to analyze the user's emotions. The data on the user's interactions with the AI ​​character is used as input, and a summary reflecting the emotions is generated as output. For example, the emotion of joy can be extracted from the part where the user says, "I think this bill will be very beneficial to small and medium-sized businesses," generating a summary that reads, "The user was very happy about this news."

[0649] (Application example 2)

[0650] Next, a description will be given of Application Example 2 of Form 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."

[0651] Conventional news comment systems have the problem that when users post comments on news articles, it is difficult to get reactions and feedback from other users, making it difficult to deepen their understanding of the news article. Additionally, there is no function to generate summaries that reflect the user's feelings, and there is also the issue that feedback that appropriately reflects the user's feelings cannot be obtained.

[0652] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0653] In this invention, the server includes means for introducing an AI character into a comment section of a news article, means for the AI ​​character to comment on the news in a format similar to that of general users, means for the user to interact with the AI ​​character about the news, means for the interaction to be visible only to the user, means for the user to post a "summary of the interaction" after the interaction is over, means for analyzing the user's emotions using an emotion engine and generating a summary that reflects the emotions, means for generating comments on the news article using a generative AI model, and means for generating comments for the AI ​​character using a prompt sentence. This allows the user to gain a deeper understanding of the news article and provides feedback that reflects the user's emotions.

[0654] A "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[0655] An "AI character" is a virtual character that interacts with users using artificial intelligence.

[0656] "A format similar to comments made by general users" means a format or style similar to that of comments posted by regular users.

[0657] "Means to comment on news" is a function that allows users to post opinions and thoughts about news articles.

[0658] "Means for users to interact with the AI ​​character about the news" refers to a function that allows users to exchange opinions about news articles through dialogue with the AI ​​character.

[0659] "Means for the interaction to be visible only to the user" refers to a function that prevents other users from seeing the content of the conversation between the user and the AI ​​character.

[0660] "A means to post a 'summary of the conversation' after the conversation is over" is a function that displays a summary of the conversation between the user and the AI ​​character.

[0661] An "emotion engine" is software that analyzes a user's emotions and generates information that reflects those emotions.

[0662] A "generative AI model" is an algorithm or system that uses artificial intelligence to generate text or comments.

[0663] A "prompt" is an instruction or question input to a generative AI model.

[0664] The system for implementing this invention introduces an AI character into the comment section of a news article, allowing users to post comments on the news article and interact with the AI ​​character. A specific embodiment of this system is described below.

[0665] System configuration

[0666] The system consists of the following main components:

[0667] 1. Server: Manages news articles, generates comments for AI characters, and manages interactions with users.

[0668] 2. User device: A device such as a smartphone or tablet that users use to view news articles and post comments.

[0669] 3. Generative AI model: An artificial intelligence model for generating comments on news articles.

[0670] 4. Sentiment Engine: Software for analyzing the sentiment of user comments and interactions and generating summaries.

[0671] Program processing

[0672] The server distributes news articles to user devices, allowing users to view the articles. When a user posts a comment on a news article, the generative AI model uses the prompt text to generate a comment from the AI ​​character. The generated comment is displayed on the user device, allowing the user to interact with the AI ​​character.

[0673] The emotion engine analyzes the user's interaction with the AI ​​character in real time and generates a summary that reflects the user's emotions, which is provided to the user after the interaction is over.

[0674] Hardware and software used

[0675] Hardware: smartphones, tablets, servers

[0676] Software: Generative AI models (e.g., OpenAI API), emotion engines (e.g., TextBlob)

[0677] Specific examples

[0678] For example, if the news article is "A new bill has been passed in the Diet," the server will input the following prompt sentence into the generative AI model:

[0679] News article: A new bill has been passed in Parliament.

[0680] How do you think this bill will affect small businesses?

[0681] The generative AI model generates comments for the AI ​​character based on this prompt and displays them on the user's device. If the user comments, "I think this bill will be very beneficial to small and medium-sized businesses," the emotion engine analyzes this comment and generates a summary that reflects the user's sentiment.

[0682] In this way, users can gain a deeper understanding of the news article and receive emotional feedback.

[0683] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0684] Step 1:

[0685] The server retrieves news articles from the database and delivers them to user devices. The input is the news article data, and the output is the news article displayed on the user device. Specifically, the server retrieves the text data of the news article and sends it to the user device.

[0686] Step 2:

[0687] Users browse news articles and post comments. The input is the news article the user browses, and the output is the comment the user posts. Specifically, users read a news article and enter their opinions and thoughts in the comment section.

[0688] Step 3:

[0689] The server receives the user's comments and sends a prompt to the generative AI model. The input is the user's comments and the news article text, and the output is the prompt to be sent to the generative AI model. Specifically, the server combines the user's comments and the news article to generate a prompt and sends it to the generative AI model.

[0690] Step 4:

[0691] The generative AI model generates comments for the AI ​​character based on the prompt sentence. The input is the prompt sentence, and the output is the AI ​​character's comment. Specifically, the generative AI model analyzes the prompt sentence and generates an appropriate comment.

[0692] Step 5:

[0693] The server sends the generated AI character's comments to the user's device. The input is the comment from the generative AI model, and the output is the AI ​​character's comment displayed on the user's device. Specifically, the server receives the generated comment and sends it to the user's device.

[0694] Step 6:

[0695] The user posts a further comment in response to the AI ​​character's comment. The input is the AI ​​character's comment, and the output is the user's new comment. Specifically, the user reads the AI ​​character's comment and then re-enters their own opinion or impression in response to it.

[0696] Step 7:

[0697] The server sends the interaction between the user and the AI ​​character to the emotion engine and performs emotion analysis. The input is the text of the interaction between the user and the AI ​​character, and the output is the result of the emotion analysis. Specifically, the server sends the text of the interaction to the emotion engine and performs emotion analysis.

[0698] Step 8:

[0699] The emotion engine generates summaries that reflect the user's emotions. The input is the result of the emotion analysis and the text of the interaction, and the output is a summary that reflects the emotions. Specifically, the emotion engine generates summaries that reflect the user's emotions based on the results of the emotion analysis.

[0700] Step 9:

[0701] The server sends the generated summary to the user terminal. The input is the summary from the emotion engine, and the output is the summary displayed on the user terminal. Specifically, the server receives the generated summary and sends it to the user terminal.

[0702] Step 10:

[0703] Users view summaries of interactions and deepen their understanding of the news article. The input is a summary that reflects sentiment, and the output is a deepening of the user's understanding. Specifically, users read summaries and deepen their understanding of the news article.

[0704] Example 3

[0705] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0706] Conventional news sites lack effective methods for increasing user time spent on the site, making it difficult to maintain user interest. Furthermore, aggressive posts and other troubles between users are common, and appropriate measures to prevent this are needed. Furthermore, it has been difficult to realize natural dialogue that responds to changes in user emotions.

[0707] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0708] In this invention, the server includes means for introducing an AI character into a comment section of a news item, means for the AI ​​character to comment on the news item in a format similar to that of general users, means for the user to interact with the AI ​​character about the news item, means for the interaction to be visible only to the user, means for the user to post a "summary of the interaction" after the interaction has ended, means for tracking the user's emotions, and means for adjusting the AI ​​character's response based on the emotion tracking information. This increases the user's stay time, serves as a receptacle for aggressive posts, and makes it possible to realize natural dialogue that responds to changes in the user's emotions.

[0709] An "artificial intelligence character" is a computer program designed to interact with a user and respond to user input using natural language processing techniques.

[0710] A "comment section" is an interface on a news site where users can post their opinions and thoughts.

[0711] "Emotion tracking" is a technology that analyzes user input data and identifies the user's emotional state in real time.

[0712] The "summary of the interaction" is information that briefly summarizes the content of the interaction between the user and the AI ​​character.

[0713] "Means to comment on news" is a function that allows an AI character to post opinions and thoughts on news articles.

[0714] "User emotion tracking means" refers to software or algorithms that analyze user input data to identify emotional states.

[0715] The "means for adjusting the response" is a function that changes the response content of the AI ​​character depending on the emotional state of the user.

[0716] "Means to increase dwell time" are features and technologies that encourage users to stay on a news site for a longer period of time.

[0717] "A means to receive offensive posts" is a function in which an artificial intelligence character receives offensive comments in order to prevent trouble between users.

[0718] This invention is a system for increasing the time users spend on news sites and preventing conflicts between users. This system uses an AI character to converse with users and adjusts its responses according to changes in the user's emotions, achieving a more natural and human-like dialogue.

[0719] The server runs software to introduce an AI character into the comment section of a news article. This AI character has the ability to comment on the news in a format similar to that of a general user. Users can interact with the AI ​​character about the news through their devices. This interaction is designed to be visible only to the user.

[0720] The server receives the user's input data and uses an emotion analysis library (e.g., IBM Watson Tone Analyzer) for emotion tracking. This emotion analysis library analyzes the user's input data and identifies the user's emotional state. For example, if a user inputs "I'm angry about the recent news," the emotion analysis library will detect "anger."

[0721] The server then uses a generative AI model (e.g., OpenAI GPT-4) to generate a response for the AI ​​character based on the user's emotional state. For example, if the user expresses anger, the generative AI model might generate a response such as, "That's tough. What's bothering you?" This generated response is sent from the server to the device and displayed to the user.

[0722] The user can continue the conversation by checking the AI ​​character's response through the device and inputting again. For example, the user can input, "I'm particularly interested in political news." In this way, the AI ​​character's response can be adjusted based on the user's emotional changes, which has the effect of increasing the user's stay time.

[0723] As a concrete example, the following prompt sentence is shown.

[0724] Example prompt sentence:

[0725] User: I'm angry about the recent news.

[0726] AI Character: That's tough. What are you particularly concerned about?

[0727] This system realizes natural dialogue that responds to changes in the user's emotions, increases the user's stay time, and prevents trouble between users by acting as a receptacle for offensive posts. The flow of the identification process in Example 3 will be described with reference to Fig. 21.

[0728] Step 1:

[0729] A user visits a news site.

[0730] A user accesses a news site using a web browser. The news site has an interactive function with an artificial intelligence character. The input is the user's access, and the output is the display of the news site.

[0731] Step 2:

[0732] The terminal receives user input.

[0733] A user inputs a comment on a news article. For example, the user inputs the text "I'm angry about the recent news." The input is the user's text data, and the output is the input data saved on the device.

[0734] Step 3:

[0735] The terminal sends the input data to the server.

[0736] The terminal sends the text data entered by the user to the server. At this time, the data is encrypted before being sent. The input is the user's text data, and the output is the data sent to the server.

[0737] Step 4:

[0738] The server analyzes the input data using an emotion engine.

[0739] The server analyzes the user's input data using an emotion analysis library (e.g., IBM Watson Tone Analyzer). The analysis results identify the user's emotional state (e.g., anger, joy, sadness). The input is the text data sent to the server, and the output is the emotion analysis result.

[0740] Step 5:

[0741] The server uses the generated AI model to generate the responses of the AI ​​character.

[0742] The server uses a generative AI model (e.g., OpenAI GPT-4) to generate a response for an AI character based on the user's emotional state. For example, if the user expresses anger, the generative AI model generates a response such as "That's terrible. What are you particularly concerned about?" The input is the emotion analysis result, and the output is the generated response.

[0743] Step 6:

[0744] The server sends the generated response to the terminal.

[0745] The server sends the generated AI character's response to the terminal. The data is encrypted before transmission. The input is the generated response, and the output is the data sent to the terminal.

[0746] Step 7:

[0747] The terminal displays the AI ​​character's response to the user.

[0748] The terminal displays the AI ​​character's response received from the server to the user. The user checks the AI ​​character's message on the screen. The input is the data sent to the terminal, and the output is the message displayed to the user.

[0749] Step 8:

[0750] The user re-enters and continues the dialogue.

[0751] The user sees the AI ​​character's reaction and inputs again, continuing the dialogue. For example, the user inputs "I'm particularly interested in political news." The input is the user's new text data, and the output is the new input data saved on the device.

[0752] (Application example 3)

[0753] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0754] The challenges facing news sites are increasing the amount of time users spend on the site and preventing conflicts between users. There is also a need for ways for users to gain a deeper understanding of news articles and moderate offensive posts. Furthermore, realizing natural dialogue that responds to users' emotions is also an important challenge.

[0755] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0756] In this invention, the server includes a means for introducing an AI character into the comment section of a news article, a means for the AI ​​character to comment on the news in a format similar to that of general users, and a means for users to interact with the AI ​​character about the news, thereby increasing the time users spend on the site, preventing conflicts between users, and deepening their understanding of the news article.

[0757] A "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[0758] An "AI character" is a virtual character that uses artificial intelligence to interact with users.

[0759] "Format similar to comments from general users" refers to comments that have the same format and style as comments posted by regular users.

[0760] "Means by which a user can interact with the AI ​​character about the news" refers to a function that allows a user to have a conversation with an AI character about the news.

[0761] "Means for ensuring that the interaction is visible only to the user" refers to a function that prevents the content of the conversation between the user and the AI ​​character from being displayed to other users.

[0762] The "interaction summary" is a concise summary of the conversation between the user and the AI ​​character.

[0763] "Means of tracking the user's emotions and adjusting the AI ​​character's response accordingly" refers to a function that detects changes in the user's emotions and changes the AI ​​character's response based on those changes.

[0764] "A means of briefly explaining the contents of a news article" is a function that clearly explains the main points of a news article.

[0765] To put this invention into practice, it is necessary to build a system that introduces AI characters into the comments section of a news website. In this system, when a user browses a news article and posts a comment, the AI ​​character will explain the news in a dialogue format and adjust its response according to the user's emotions.

[0766] Hardware and Software Configuration

[0767] Hardware:

[0768] server

[0769] User devices (smartphones, tablets, computers)

[0770] software:

[0771] Python

[0772] OpenAI API

[0773] Sentiment analysis library (e.g., sentiment_analysis)

[0774] System Operation

[0775] server:

[0776] The server manages the content of news articles and accepts user comments. It also runs a dialogue engine for the AI ​​character and includes a sentiment analysis engine for tracking user emotions. When the server receives a user comment, it analyzes the sentiment of the comment using the sentiment analysis engine and generates a response for the AI ​​character based on the results.

[0777] User device:

[0778] The user device displays news articles and provides an interface for users to post comments. When a user posts a comment, the comment is sent to the server, where it is processed. The resulting AI character's response is displayed on the user device.

[0779] Data processing and calculation

[0780] Sentiment analysis:

[0781] When the server receives a user's comment, it analyzes the sentiment of the comment using a sentiment analysis library, which takes the comment text as input and outputs a positive, negative, or neutral sentiment.

[0782] AI character reaction generation:

[0783] Based on the results of the sentiment analysis, the server uses the OpenAI API to generate responses for the AI ​​character, which are then adjusted appropriately according to the user's emotions.

[0784] Specific examples

[0785] Handling user comments:

[0786] If a user posts a comment like "I think this news is really terrible," the sentiment analysis library will identify this comment as negative. The server will send the following prompt to the OpenAI API to generate a response for the AI ​​character:

[0787] A user made a negative comment: I think this news is really terrible.

[0788] AI character reactions:

[0789] News article explanation:

[0790] To provide a brief explanation of the news article, the server sends the following prompt to the OpenAI API to generate an explanation:

[0791] Briefly explain what the news article is about: Today's news reported that the economy is recovering faster than expected.

[0792] This will help users understand the main points of news articles more easily and deepen their understanding of the news through dialogue with the AI ​​characters. The AI ​​characters will also act as a receptacle for offensive posts, preventing disputes between users.

[0793] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0794] Step 1:

[0795] Users view news articles and post comments.

[0796] Input: User comment (e.g., "I think this news is really terrible.")

[0797] Output: The user's comment is sent to the server.

[0798] Specific operation: When a user enters a comment in the comment field of a news article and presses the send button, the comment is sent to the server.

[0799] Step 2:

[0800] The server receives the user's comments and performs sentiment analysis.

[0801] Input: User comment (e.g., "I think this news is really terrible.")

[0802] Output: Sentiment analysis result (e.g., negative)

[0803] Specific operation: The server analyzes the sentiment of user comments using a sentiment analysis library. The sentiment analysis library takes the comment text as input and outputs a sentiment, either positive, negative, or neutral.

[0804] Step 3:

[0805] The server generates the AI ​​character's response based on the emotion analysis results.

[0806] Input: Sentiment analysis results (e.g., negative), user comments (e.g., "I think this news is really terrible.")

[0807] Output: AI character's response (e.g. "That's unfortunate. What did you find particularly disappointing?")

[0808] Specific operation: The server uses the OpenAI API to generate prompts based on the results of sentiment analysis and generate responses from the AI ​​character. Example prompt: "A user made a negative comment: I think this news is really terrible. The AI ​​character's response:"

[0809] Step 4:

[0810] The server sends the reaction of the generated AI character to the user's device.

[0811] Input: AI character's response (e.g., "That's a shame. What exactly did you find bad about it?")

[0812] Output: AI character's reaction displayed on the user's device

[0813] Specific operation: The server sends the reaction of the generated AI character to the user's device, which displays it.

[0814] Step 5:

[0815] The user continues to interact with the AI ​​character.

[0816] Input: User's additional comments (e.g., "I think it's terrible that the economy is recovering so slowly.")

[0817] Output: The additional comments are sent to the server for further sentiment analysis and reaction generation.

[0818] Specific operation: When the user enters additional comments in response to the AI ​​character's reaction and presses the send button, the comments are sent to the server again, and the process of sentiment analysis and reaction generation is repeated.

[0819] Step 6:

[0820] The server provides a brief explanation of the news article.

[0821] Input: The content of the news article (e.g., "Today's news reported that the economy is recovering faster than expected.")

[0822] Output: A description of the news article (e.g., "It has been reported that the economy is recovering faster than expected. This shows that the government's economic policies are working.")

[0823] Specific operation: The server uses the OpenAI API to generate a prompt that briefly explains the content of the news article, and then generates an explanation. Example of a prompt: "Please briefly explain the content of the news article: Today's news reported that the economy is recovering faster than expected."

[0824] Step 7:

[0825] The server transmits the generated commentary of the news article to the user terminal.

[0826] Input: A description of the news article (e.g., "The economy is reported to be recovering faster than expected. This shows that the government's economic policies are working.")

[0827] Output: A description of the news article displayed on the user's device

[0828] Specific operation: The server sends the generated commentary of the news article to the user terminal, which displays it.

[0829] 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.

[0830] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ) 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.

[0831] Another example of generative AI is Gemini (registered trademark) (Internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0832] 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.

[0833] [Second embodiment]

[0834] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0835] 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.

[0836] 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).

[0837] 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.

[0838] 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.

[0839] 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).

[0840] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0841] 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.

[0842] 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.

[0843] 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.

[0844] 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.

[0845] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0846] "Example 1"

[0847] One embodiment of the present invention is a system that introduces an AI character to a news site. In this system, the AI ​​character posts comments on news articles. The AI ​​character's comments are in a format similar to that of general users, and users can exchange messages with the AI ​​character about the news. These exchanges are set to be visible only to the user. After the exchange is over, the user can post a "summary of the exchange."

[0848] "Example 2"

[0849] As a specific example, consider an AI character posting a comment on a political news article. The AI ​​character understands the content of the article and generates a comment based on that understanding. For example, the AI ​​character can post a question such as, "How do you think this bill will affect small and medium-sized businesses?". Users can express their opinion in response to this question, and the AI ​​character will respond with an appropriate response. This interaction helps users gain a deeper understanding of the news article.

[0850] "Example 3"

[0851] The system of this invention also has the effect of increasing the time users spend on the site. As users enjoy interacting with the AI ​​character, they spend more time on the news site. Furthermore, the AI ​​character acts as a receptacle for offensive posts, preventing conflict between users.

[0852] The processing flow of each embodiment will be described below.

[0853] "Example 1"

[0854] Step 1: The AI ​​character analyzes the news article, including the article's text content, associated images and videos, and other metadata.

[0855] Step 2: The AI ​​character generates comments on the news article, which are formatted similarly to typical user comments.

[0856] Step 3: The user responds to the AI ​​character's comments. The user can express their own opinions and thoughts to the AI ​​character.

[0857] Step 4: The AI ​​character generates an appropriate reply to the user's reaction. This reply is a response to the user's comment and continues the dialogue between the user and the AI ​​character.

[0858] Step 5: The user submits a summary of the conversation. This summary is a concise summary of the conversation between the user and the AI ​​character.

[0859] "Example 2"

[0860] Step 1: The AI ​​character analyzes political news articles, including the article's text content, associated images and videos, and other metadata.

[0861] Step 2: The AI ​​character generates question-type comments on news articles, such as "How do you think this bill will affect small businesses?"

[0862] Step 3: The user gives their opinion in response to the AI ​​character's question.

[0863] Step 4: The AI ​​generates an appropriate response to the user's opinion. This response is a reaction to the user's opinion and continues the dialogue between the user and the AI.

[0864] Step 5: The user submits a summary of the conversation. This summary is a concise summary of the conversation between the user and the AI ​​character.

[0865] "Example 3"

[0866] Step 1: A user visits a news site.

[0867] Step 2: The user initiates a dialogue with the AI ​​character, which is based on a news article.

[0868] Step 3: Users enjoy interacting with the AI ​​character, which increases the time they spend on the news site.

[0869] Step 4: If a user posts something offensive, the AI ​​character will respond, preventing any conflict between users.

[0870] Example 1

[0871] Next, a description will be given of Example 1 of Form 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."

[0872] On traditional news sites, when users post comments on news articles, their interactions with other users are made public, raising the risk of privacy issues and offensive comments. Furthermore, there is a lack of support for users to gain a deeper understanding of the content of news articles, which can lead to a shallow understanding of the articles. Furthermore, there is a lack of efforts to increase user time on the site, which can lead to a decline in site usage.

[0873] 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.

[0874] In this invention, the server includes: means for introducing an AI character into a news comment section; means for the AI ​​character to comment on the news in a format similar to that of general users; means for the user to interact with the AI ​​character about the news; means for the interaction to be visible only to the user; means for the user to post a "summary of the interaction" after the interaction is over; means for generating comments on news articles using a generative AI model; means for inputting user input as a prompt sentence into the generative AI model to generate a reply; and means for retrieving news articles. This allows users to comment on news articles while maintaining their privacy and deepen their understanding of the news article through their interaction with the AI ​​character. It also increases user dwell time and reduces the risk of offensive comments.

[0875] A "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[0876] An "artificial intelligence character" is a virtual character generated using artificial intelligence technology that can interact with users.

[0877] A "format similar to comments from general users" is a format that has the same style and content as comments posted by regular users.

[0878] "User" means an individual or organization that uses a news site.

[0879] "Interaction" refers to the dialogue and communication that takes place between the user and the AI ​​character.

[0880] The "interaction summary" is a concise summary of the interaction between the user and the AI ​​character.

[0881] A "generative AI model" is a model for generating text using artificial intelligence technology, including those that use natural language processing technology.

[0882] A "prompt" is an instruction or question to be input into a generative AI model.

[0883] A "news article" is a piece of text containing news or information that is published on a news site.

[0884] A "server" is a computer system used to operate a news site and process data.

[0885] "User session information" is information that records the status and activities of a user while the user is accessing a news site.

[0886] A "database" is a system for storing data such as news articles, user comments, and summaries of interactions.

[0887] The present invention is a system that introduces an AI character to a news site, allowing users to post comments on news articles and interact with the AI ​​character. This system includes elements of a server, a terminal, and a user.

[0888] System Overview

[0889] The server retrieves the latest news articles from news sites and uses a generative AI model to generate comments on the news articles. The generated comments are in a format similar to comments made by general users and are only displayed to specific users. Users can use their devices to interact with the AI ​​character about the news article. After the interaction is complete, users can post a "summary of the interaction."

[0890] Hardware and software used

[0891] Server: Retrieves news articles, runs generative AI models, generates and displays comments, and manages user sessions.

[0892] Device: The device (PC, smartphone, etc.) on which users view news articles, post comments, and interact with AI characters.

[0893] Generative AI models: Artificial intelligence models for generating comments on news articles (e.g., OpenAI's GPT-4).

[0894] Data processing and calculation

[0895] 1. Get news articles: The server uses the news site's API to get the latest news articles and stores them in a database.

[0896] 2. Comment generation: The server inputs the content of the retrieved news article as a prompt sentence into the generative AI model to generate a comment.

[0897] 3. Display the comment: The server displays the generated comment, associating it with the specific user session.

[0898] 4. User interaction: When a user enters a comment using the device, the server inputs the comment as a prompt into the generative AI model to generate a reply.

[0899] 5. Interaction Summary: When a user enters a summary of an interaction, the server stores the summary in a database and makes it available for other users to view.

[0900] Specific examples

[0901] News article: "COVID-19 cases declining as vaccinations progress."

[0902] AI character comment: "It's good news that vaccinations are progressing. We can also expect the number of infections to decrease."

[0903] Prompt Sentence Examples

[0904] "Please generate a comment from an AI character in a format similar to that of a regular user in response to a news article about the decline in the number of COVID-19 infections as vaccinations progress."

[0905] In this way, the system of the present invention allows users to post comments on news articles and interact with AI characters, thereby improving user understanding of the news articles while maintaining privacy and increasing site usage.

[0906] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0907] Step 1: Get the news article

[0908] The server retrieves the latest news articles using the news site's API. Specifically, the server calls the API at "https: / / newsapi.org / " every morning at 9:00 and retrieves the latest news articles. The retrieved news articles are stored in a database.

[0909] Input: API request

[0910] Output: News article data

[0911] Data processing: Store the news article data obtained from the API in a database.

[0912] Step 2: Generate comments for the AI ​​character

[0913] The server inputs the content of the retrieved news article into the generative AI model as a prompt sentence to generate a comment. Specifically, the title and body of the news article are input into the generative AI model as prompt sentences. For example, in response to a news article stating, "As COVID-19 vaccinations progress, the number of infections is decreasing," the server inputs the prompt sentence, "What do you think about this news?" into the generative AI model to generate a comment.

[0914] Input: News article data, prompt

[0915] Output: AI character's comment

[0916] Data processing: News article data is converted into prompt sentences, which are then fed into a generative AI model to generate comments.

[0917] Step 3: View comments

[0918] The server associates the generated AI character's comments with a specific user's session and displays them. Specifically, it uses the user's session information to ensure that the comments are only displayed to that specific user. For example, if user A is logged in, it uses user A's session ID to ensure that the comments are only displayed to user A.

[0919] Input: AI character comments, user session information

[0920] Output: Comments that are displayed to the user

[0921] Data processing: Displays the AI ​​character's comments based on the user's session information.

[0922] Step 4: User interaction

[0923] Users use their devices to interact with the AI ​​character about news articles. Specifically, when a user enters a comment, the server inputs the comment as a prompt into the generative AI model and generates a reply. For example, if a user inputs a comment such as "It's good news that vaccinations are progressing," the server inputs the comment as a prompt into the generative AI model and generates a reply.

[0924] Input: User comments

[0925] Output: AI character's reply

[0926] Data processing: User comments are converted into prompt sentences and fed into a generative AI model to generate replies.

[0927] Step 5: Summarize the conversation

[0928] After the interaction is complete, the user enters a "summary of the interaction" and posts it. Specifically, when a user enters a summary of the interaction, the server saves the summary in a database and makes it available for other users to view. For example, if a user enters a summary such as "Through my interaction with the AI ​​character, I was able to reaffirm the importance of vaccination," the server saves the summary in a database and makes it available for other users to view.

[0929] Input: User Summary

[0930] Output: Summary stored in database

[0931] Data processing: Save the user's summary in a database and make it available for other users to view.

[0932] (Application example 1)

[0933] Next, a description will be given of Application Example 1 of Form 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."

[0934] On traditional news sites, users can refer to other users' comments when reading news articles, but they lack the ability to have individual conversations. They also lack support for users to gain a deeper understanding of news articles and are not designed to increase user time on the site. Furthermore, offensive comments are sometimes posted, and measures to address this are needed.

[0935] 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.

[0936] In this invention, the server includes means for introducing an AI character into a news comment section, means for the AI ​​character to comment on the news in a format similar to that of general users, means for the user to interact with the AI ​​character about the news, means for the interaction to be visible only to the user, means for the user to post a "summary of the interaction" after the interaction is over, means for the AI ​​character to post comments and interact with the user as the user reads the news article, and means for generating a summary of the interaction between the user and the AI ​​character. This allows users to understand the news article while individually interacting with the AI ​​character, deepening their understanding of the news and increasing their stay time. Furthermore, the AI ​​character's role as a receptacle for offensive comments helps maintain a healthy comment section.

[0937] A "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[0938] An "artificial intelligence character" is a virtual character generated using artificial intelligence technology that has the ability to interact with users.

[0939] A "format similar to comments from general users" is a format that has the same style and content as comments posted by regular users.

[0940] "Means by which the user can exchange news with the AI ​​character" is a function that allows the user to have two-way communication with the AI ​​character.

[0941] "Means for making the exchange visible only to the user" refers to a function that prevents the content of the exchange between the user and the AI ​​character from being displayed to other users.

[0942] "A means to post a 'summary of the conversation' after the conversation is over" is a function that allows users to post a concise summary of the content of their conversation with an AI character.

[0943] "A means for an AI character to post comments and interact with users when they read news articles" is a function that enables an AI character to provide comments to users who are viewing news articles and start a dialogue.

[0944] The "means for generating a summary of the interaction between the user and the AI ​​character" is a function for automatically summarizing the content of the conversation between the user and the AI ​​character.

[0945] As an embodiment of the present invention, a system for introducing an AI character into a news site is constructed. A specific embodiment of the system is shown below.

[0946] System Program

[0947] This system consists of a server and a user terminal. The server provides a function that allows an AI character to post comments on news articles and interact with users. The user terminal displays news articles and provides an interface for interacting with the AI ​​character.

[0948] Hardware and software used

[0949] Hardware: User devices such as smartphones, tablets, and PCs

[0950] Software: Flask (Python web framework), OpenAI API (generative AI model)

[0951] Data processing and calculation

[0952] 1. Acquisition of news articles: The server acquires news articles from the database and sends them to the user terminal.

[0953] 2. AI character comment generation: The server receives the news article as input and uses a generative AI model (e.g., text-davinci-003) to generate AI character comments based on the prompt sentence.

[0954] 3. User interaction: The user device displays the generated comments and provides an interface through which the user can interact with the AI ​​character. The user's input is sent to the server, which again uses the generative AI model to generate a response.

[0955] 4. Summary generation of interaction: The server receives the content of the interaction between the user and the AI ​​character as input and generates a summary using the generative AI model. The generated summary is sent to the user's device so that the user can post it.

[0956] Specific examples

[0957] Suppose a user is reading a news article titled "COVID-19 Vaccination Progress." An AI character comments, "As vaccination progresses, it is expected that herd immunity will be achieved." The user responds, "That's true, but I'm worried about side effects." The AI ​​character responds, "It would be best to consult an expert about side effects." After the exchange, the user posts a summary saying, "We discussed the progress of vaccination and concerns about side effects."

[0958] Prompt Sentence Examples

[0959] Generating comments from AI characters on news articles:

[0960] News article: COVID-19 vaccinations progress

[0961] AI character's comment:

[0962] Generate summaries of user interactions:

[0963] User interaction with AI character:

[0964] User: As vaccinations progress, it is hoped that we will achieve herd immunity.

[0965] AI Character: Yes, but I'm worried about the side effects.

[0966] User: It's best to seek expert advice regarding side effects.

[0967] summary:

[0968] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0969] Step 1:

[0970] The server retrieves the news articles from the database and transmits them to the user terminal.

[0971] Input: News article ID

[0972] Data processing: Search and retrieve news article content from the database.

[0973] Output: News article content

[0974] Specific operation: The server searches the database using the news article ID as a key, retrieves the content of the corresponding news article, and then sends the retrieved content of the news article to the user's terminal.

[0975] Step 2:

[0976] The user terminal displays the acquired news article.

[0977] Input: News article content

[0978] Data processing: Converting the content of news articles into a format that can be displayed in the user interface.

[0979] Output: The displayed news article

[0980] Specific operation: The user terminal displays the contents of the news article received from the server on the user interface, allowing the user to view the news article.

[0981] Step 3:

[0982] The server receives a news article as input and uses a generative AI model to generate comments for an artificially intelligent character based on a prompt sentence.

[0983] Input: News article content

[0984] Data processing: The content of the news article is embedded in a prompt sentence and input to the generative AI model, which then generates comments based on the prompt sentence.

[0985] Output: Comments from the AI ​​character

[0986] Specific operation: The server embeds the content of the news article into a prompt sentence and inputs it into the generative AI model (e.g., text-davinci-003). The generative AI model generates a comment for the AI ​​character based on the prompt sentence and sends the comment to the user's device.

[0987] Step 4:

[0988] The user terminal displays the generated comments and provides an interface through which the user can interact with the artificial intelligence character.

[0989] Input: AI character's comment

[0990] Data processing: Converting comments into a format that can be displayed in the user interface.

[0991] Output: Comments displayed

[0992] Specific operation: The user terminal displays the AI ​​character's comments received from the server on the user interface. The user can view the comments and start a dialogue.

[0993] Step 5:

[0994] The user interacts with an artificially intelligent character, and the user's input is sent to the server.

[0995] Input: User input

[0996] Data processing: Converting user input into a format to send to the server.

[0997] Output: User input sent to the server

[0998] Specific operation: The user inputs a response to the AI ​​character's comment. The user terminal sends the user's input to the server.

[0999] Step 6:

[1000] The server receives the user's input and generates a response, again using the generative AI model.

[1001] Input: User input

[1002] Data processing: The user's input is embedded in a prompt sentence and input to the generative AI model, which then generates a response based on the prompt sentence.

[1003] Output: AI character's response

[1004] Specific operation: The server embeds the user's input into a prompt sentence and inputs it into the generative AI model. The generative AI model generates a response for the AI ​​character based on the prompt sentence and sends the response to the user's device.

[1005] Step 7:

[1006] The user terminal displays the generated response and allows the user to continue the dialogue.

[1007] Input: AI character's response

[1008] Data processing: Converting responses into a format that can be displayed in the user interface.

[1009] Output: The displayed response

[1010] Specific operation: The user terminal displays the AI ​​character's response received from the server on the user interface, allowing the user to continue the dialogue.

[1011] Step 8:

[1012] The server receives the dialogue between the user and the AI ​​character as input and generates a summary using a generative AI model.

[1013] Input: Dialogue between the user and the AI ​​character

[1014] Data processing: The dialogue is embedded in a prompt and fed into a generative AI model, which then generates a summary based on the prompt.

[1015] Output: A summary of the conversation

[1016] Specific operation: The server embeds the dialogue between the user and the AI ​​character into a prompt sentence and inputs it into the generative AI model. The generative AI model generates a summary of the dialogue based on the prompt sentence and sends the summary to the user's device.

[1017] Step 9:

[1018] The user terminal displays the generated summary and allows the user to contribute.

[1019] Input: A summary of the conversation

[1020] Data processing: Converting the summary into a format that can be displayed in the user interface.

[1021] Output: Displayed summary

[1022] Specific operation: The user terminal displays the summary of the exchange received from the server on the user interface. The user can check the summary and post.

[1023] Example 2

[1024] Next, a description will be given of Example 2 of Form 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."

[1025] Conventional news comment systems lacked the support required for users to gain a deeper understanding of news articles, and were not designed to increase user retention. Furthermore, there was a need for an appropriate response to offensive posts.

[1026] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1027] In this invention, the server includes a means for acquiring news articles, a means for analyzing the acquired news articles using natural language processing technology, a means for generating comments and questions using a generative AI model based on the analysis results, and a means for analyzing user responses and generating appropriate responses using the generative AI model. This allows users to gain a deeper understanding of news articles and increase their stay time. It also allows for appropriate responses to offensive posts.

[1028] The "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[1029] An "artificial intelligence character" is a program that automatically generates comments and questions about news articles and interacts with users.

[1030] "A format similar to comments from general users" refers to comments and questions generated by AI characters that are in the same format and style as comments and questions posted by regular users.

[1031] "Means for making interactions visible only to the user" refers to a function that prevents the conversation between the user and the AI ​​character from being displayed to other users.

[1032] The "summary of the interaction" is information that briefly summarizes the content of the conversation between the user and the AI ​​character.

[1033] "Means for obtaining news articles" refers to a function for collecting the latest news articles from news sites and RSS feeds on the Internet.

[1034] "Natural language processing technology" is a technology that enables computers to understand and analyze human language.

[1035] A "generative AI model" is an algorithm that allows artificial intelligence to generate new text based on a given prompt.

[1036] The "means for analyzing user responses" is a function for analyzing the text entered by the user and understanding its content and emotions.

[1037] The "means for generating an appropriate response" is a function that enables an AI character to generate appropriate comments or questions based on the user's response.

[1038] This invention is a system that introduces an AI character into a news comment section and provides dialogue that allows users to gain a deeper understanding of a news article. Specific embodiments of this system are described below.

[1039] The server first retrieves the latest news articles from news sites or RSS feeds on the Internet. For example, the server retrieves the RSS feed from "https: / / example-news.com / rss" and extracts the latest articles. This process uses common web scraping techniques or RSS reader software.

[1040] The server then analyzes the content of the retrieved news articles using natural language processing technology. Specifically, it uses the Google Cloud Natural Language API to analyze the article's topics, keywords, and sentiment. For example, the server analyzes the content of the article and extracts keywords such as "small and medium-sized enterprises," "bill," and "tax benefits."

[1041] Based on the analysis results, the server uses a generative AI model (e.g., OpenAI's GPT-4) to generate comments and questions. For example, the server generates the question, "How do you think this bill will affect small businesses?" An example of this prompt would be:

[1042] Analyze the content of a news article and generate questions such as:

[1043] "How do you think this bill will affect small businesses?"

[1044] The generated comments and questions are displayed to the user via the terminal. The terminal displays the question "How do you think this bill will affect small businesses?" below the news article. The user inputs their opinion in response to this question. For example, the user might input "I think this bill is beneficial to small businesses because it includes tax breaks."

[1045] The server receives the user's response and analyzes it again using natural language processing technology. This allows it to understand the user's opinions and emotions. Based on the analysis results, the server uses a generative AI model to generate an appropriate response. For example, the server might generate a response such as, "I see, tax incentives are important for small and medium-sized businesses. What else are you concerned about?"

[1046] This series of processes allows users to gain a deeper understanding of news articles and increase their viewing time. The AI ​​character can also respond appropriately to offensive posts. This system is an effective way for users to gain a deeper understanding of news articles and promote constructive dialogue.

[1047] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1048] Step 1: Get the news article

[1049] The server retrieves the latest news articles from news sites and RSS feeds on the Internet. Specifically, the server retrieves the RSS feed from "https: / / example-news.com / rss" and extracts the latest articles. The input is the RSS feed URL, and the output is the news article data. This process uses web scraping technology or RSS reader software.

[1050] Step 2: Analyzing the article content

[1051] The server analyzes the content of the retrieved news articles using natural language processing technology. Specifically, it uses the Google Cloud Natural Language API to analyze the article's topics, keywords, and sentiment. The input is the news article data, and the output is the analysis results (topic, keyword, and sentiment analysis results). For example, the server analyzes the content of the article and extracts keywords such as "small and medium-sized enterprises," "bill," and "tax incentives."

[1052] Step 3: Generate comments and questions

[1053] The server generates comments and questions using a generative AI model (e.g., OpenAI's GPT-4) based on the analysis results. The input is the analysis results, and the output is the generated comments and questions. For example, the server generates the question, "How do you think this bill will affect small businesses?" An example of this prompt is as follows:

[1054] Analyze the content of a news article and generate questions such as:

[1055] "How do you think this bill will affect small businesses?"

[1056] Step 4: Display to the user

[1057] The terminal displays comments and questions received from the server to the user. The input is the generated comment or question, and the output is the text displayed on the user's screen. For example, the terminal might display the question "How do you think this bill will affect small businesses?" below a news article.

[1058] Step 5: Getting the user's response

[1059] The user inputs their opinion in response to the displayed question. The input is the user's opinion, and the output is the user's response data. For example, the user might input, "I think this bill is beneficial to small businesses because it includes tax breaks."

[1060] Step 6: Parsing the user's response

[1061] The server receives the user's response and analyzes it again using natural language processing technology. The input is the user's response data, and the output is the analysis results (the analysis results of the user's opinions and emotions). This allows the user's opinions and emotions to be understood.

[1062] Step 7: Generating AI character reactions

[1063] The server uses a generative AI model to generate an appropriate response based on the results of the user's response analysis. The input is the user's response analysis result, and the output is the generated response. For example, the server generates a response such as, "I see, tax incentives are important for small and medium-sized businesses. What else are you concerned about?"

[1064] Step 8: View the reaction

[1065] The terminal displays the responses received from the server to the user. The input is the generated response, and the output is the text displayed on the user's screen, allowing the user to continue interacting with the AI ​​character.

[1066] This process allows users to gain a deeper understanding of news articles and increase their time spent on the site. It also enables the AI ​​character to respond appropriately to offensive posts.

[1067] (Application example 2)

[1068] Next, a description will be given of Application Example 2 of Form 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."

[1069] Conventional news comment systems lack support for users to gain a deeper understanding of news articles. It is also difficult for users to receive appropriate feedback when posting comments on news articles. Furthermore, they lack the means to increase user retention and lack sufficient countermeasures against offensive posts.

[1070] The specific processing by the specific 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 introducing an AI character into a news comment section, means for the AI ​​character to comment on the news in a format similar to that of general users, means for the user to interact with the AI ​​character about the news, means for the interaction to be visible only to the user, means for posting a "summary of the interaction" after the interaction is over, means for generating comments on the news article using a generative AI model, and means for providing an appropriate response to the user's opinion. This allows users to gain a deeper understanding of the news article and receive appropriate feedback. It also increases the user's stay time and strengthens measures against offensive posts.

[1071] A "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[1072] An "AI character" is a virtual character generated using artificial intelligence technology that can interact with users.

[1073] "A format similar to comments made by general users" refers to comments that are in the same format and style as comments posted by regular users.

[1074] "Means for commenting on news" refers to methods and functions for posting opinions and thoughts about news articles.

[1075] "Means by which users can interact with the AI ​​character about the news" refers to methods or functions that allow users to exchange opinions about news articles through dialogue with the AI ​​character.

[1076] "Means for ensuring that the interaction is visible only to the user" refers to methods or functions that prevent the content of the conversation between the user and the AI ​​character from being displayed to other users.

[1077] "Means for posting summaries of interactions" refers to methods or functions for posting summaries of the content of the conversation between the user and the AI ​​character.

[1078] A "generative AI model" is a model that uses artificial intelligence technology to generate text and data.

[1079] The "means for generating comments on news articles" refers to a method or function for generating appropriate comments based on the content of a news article.

[1080] "Means of responding appropriately to user opinions" refers to methods and functions that allow AI characters to respond appropriately to opinions posted by users.

[1081] The system for implementing this invention introduces an AI character into the comment section of a news article, allowing users to post comments on the news article and interact with the AI ​​character. A specific embodiment of this system is described below.

[1082] System configuration

[1083] The system consists of a server and a user terminal. The server contains a news article database, a generative AI model, a comment generation module, and a dialogue management module. The user terminal is a device such as a smartphone or PC with a news application installed.

[1084] Program processing

[1085] The server retrieves articles from a database of news articles and delivers them to the user's device. When the user reads a news article, the AI ​​character generates a comment based on the article's content. This comment generation uses a generative AI model (e.g., OpenAI's text-davinci-003).

[1086] The generated comments are displayed on the user's device, allowing the user to initiate a dialogue with the AI ​​character. The dialogue management module receives the user's input and sends an appropriate prompt to the generative AI model. The generative AI model then generates an appropriate response to the user's opinion and returns it to the user's device.

[1087] Hardware and software used

[1088] Hardware: Servers, smartphones, PCs

[1089] Software: News application, generative AI model (OpenAI API)

[1090] Data processing and calculation

[1091] The server inputs the contents of the news article as a prompt into the generative AI model to generate a comment. It also generates a prompt for the user's opinion, which is then input into the generative AI model to generate an appropriate response.

[1092] Specific examples

[1093] For example, if the news article states, "The government has introduced a new tax reform bill. This bill aims to provide special tax deductions for small and medium-sized businesses," the following prompt sentence will be generated:

[1094] Comment generation for news articles

[1095] News article: The government has introduced a new tax reform bill, which aims to provide special tax breaks for small and medium-sized businesses.

[1096] How do you think this bill will affect small businesses?

[1097] Generate responses to user opinions

[1098] User Opinion: I think this bill will be very beneficial for small businesses.

[1099] AI character response:

[1100] This allows users to gain a deeper understanding of news articles and receive appropriate feedback, increases user retention, and strengthens measures against offensive posts.

[1101] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1102] Step 1:

[1103] The server retrieves the latest news articles from a news article database. The input is the news article database, and the output is the retrieved news article. Specifically, the server executes a database query to retrieve the content of the news article.

[1104] Step 2:

[1105] The server delivers the retrieved news articles to the user terminal. The input is the retrieved news article, and the output is the news article displayed on the user terminal. Specifically, the server sends the news article to the user terminal as an HTTP response.

[1106] Step 3:

[1107] The user terminal displays the news article. The input is the news article sent from the server, and the output is the news article that the user can view. Specifically, the user terminal displays the news article on the screen through a news application.

[1108] Step 4:

[1109] The server generates a prompt based on the content of the news article. The input is the content of the news article, and the output is the generated prompt. Specifically, the server analyzes the content of the news article and generates a prompt containing an appropriate question.

[1110] Step 5:

[1111] The server inputs a prompt sentence into the generative AI model and generates a comment. The input is the generated prompt sentence, and the output is the generated comment. Specifically, the server sends an API request to the generative AI model (e.g., OpenAI's text-davinci-003) to generate a comment.

[1112] Step 6:

[1113] The server sends the generated comment to the user terminal. The input is the generated comment, and the output is the comment displayed on the user terminal. In concrete terms, the server sends the comment to the user terminal as an HTTP response.

[1114] Step 7:

[1115] The user terminal displays the generated comments. The input is the comment sent from the server, and the output is the comment that the user can view. Specifically, the user terminal displays the comment on the screen through a news application.

[1116] Step 8:

[1117] The user initiates a dialogue with the AI ​​character. The input is the user's opinion or question, and the output is the input to the AI ​​character. Specifically, the user enters text through a news application.

[1118] Step 9:

[1119] The server generates a prompt based on the user's input. The input is the user's opinion or question, and the output is the generated prompt. Specifically, the server analyzes the user's input and generates a prompt with an appropriate response.

[1120] Step 10:

[1121] The server inputs a prompt sentence into the generative AI model and generates a response. The input is the generated prompt sentence, and the output is the generated response. Specifically, the server sends an API request to the generative AI model to generate a response.

[1122] Step 11:

[1123] The server sends the generated response to the user terminal. The input is the generated response, and the output is the response displayed on the user terminal. Specifically, the server sends the response to the user terminal as an HTTP response.

[1124] Step 12:

[1125] The user device displays the generated responses. The input is the response sent from the server, and the output is the response that the user can view. Specifically, the user device displays the responses on the screen through a news application.

[1126] Example 3

[1127] Next, a description will be given of Example 3 of Form Example 3. 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."

[1128] The challenges facing news sites are increasing the amount of time users spend on the site and preventing conflicts between users. In particular, when users comment on news stories, aggressive comments are likely to occur, which can cause conflicts between users. Furthermore, users tend to spend less time on the site due to a lack of means for them to gain a deeper understanding of the news content.

[1129] The identification process by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means. In this invention, the server includes a means for introducing an AI character into a comment section of a news item, a means for the AI ​​character to comment on the news item in a format similar to that of general users, a means for the user to interact with the AI ​​character about the news item, a means for the interaction to be visible only to the user, a means for the user to post a "summary of the interaction" after the interaction has ended, a means for sending the user's input as a prompt to the generative AI model, a means for the generative AI model to generate a response based on the prompt, and a means for displaying the generated response to the user. This allows users to gain a deeper understanding of the news content through dialogue with the AI ​​character, thereby extending their stay on the site. Furthermore, the AI ​​character's role as a receptacle for offensive posts can prevent conflicts between users.

[1130] A "news comment section" is an area on a news site where users can post their opinions and thoughts about a news article.

[1131] An "artificial intelligence character" is a virtual character generated using artificial intelligence technology, and is capable of interacting with the user.

[1132] "A format similar to comments made by general users" refers to a format or style similar to that of comments posted by regular users.

[1133] A "prompt sentence" is an input sentence that is generated based on a user's input and causes an artificial intelligence model to generate a response.

[1134] A "generative artificial intelligence model" is a model that uses artificial intelligence techniques to generate an appropriate response based on a prompt sentence.

[1135] A "response" is a reply or reaction generated by an AI character in response to a user's input.

[1136] The "summary of the interaction" is a concise summary of the content of the conversation between the user and the AI ​​character.

[1137] An "offensive post" is a comment that expresses discomfort or hostility toward another user.

[1138] A "receptacle" is something that plays a role in accepting and processing specific content or situations.

[1139] The present invention is a system for increasing the time users spend on a news site and preventing trouble between users. A specific embodiment of this system will be described below.

[1140] First, a user accesses a news site using a web browser. The homepage of the news site displays news articles along with a chat window with an AI character. Through this chat window, the user can enter questions and comments to the AI ​​character.

[1141] When a user types something into the chat window, the device sends it to a server, which then formats it into a prompt and sends it to a generative AI model, such as OpenAI's GPT-4.

[1142] The generative AI model generates an appropriate response based on the prompt sentence. The generated response is returned to the server, which then sends it to the device, which displays the received response in a chat window for the user to see.

[1143] As a concrete example, consider the case where a user inputs "Tell me about today's news." This input becomes a prompt sentence, which the server sends to the generative AI model. The generative AI model generates a response such as "Today's news reports that economic growth is progressing better than expected," and returns it to the server. The server sends this response to the device, which then displays it to the user.

[1144] An example of a prompt is as follows:

[1145] "How's the weather today?"

[1146] "Please tell me the latest sports news."

[1147] "What are the latest technology trends?"

[1148] This system allows users to gain a deeper understanding of the news content through dialogue with an AI character. Furthermore, the AI ​​character can act as a receptacle for offensive posts, preventing conflict between users. This increases the time users spend on news sites and improves user satisfaction.

[1149] This system is also beneficial for news site operators, as it can increase user engagement. Furthermore, by utilizing artificial intelligence technology, it becomes possible to provide efficient user support. The flow of the identification process in Example 3 will be described with reference to FIG. 15.

[1150] Step 1:

[1151] A user visits a news site.

[1152] A user accesses a news site by entering its URL using a web browser. The input is the user's access request, and the output is the news site's homepage, which displays news articles and a chat window with an AI character.

[1153] Step 2:

[1154] The user inputs questions and comments to the AI ​​character.

[1155] The user types a question or comment into the chat window, for example, "Tell me about today's news." The input is the user's text data, and the output is the user's input displayed on the terminal.

[1156] Step 3:

[1157] The terminal sends the user's input to the server.

[1158] The terminal sends the user's input to the server as text data. The input is the user's text data, and the output is the text data sent to the server. In concrete terms, the terminal sends data using an HTTP request.

[1159] Step 4:

[1160] The server sends the user's input as a prompt to the generative AI model.

[1161] The server formats the received user input as a prompt sentence and sends it to the generative AI model. The input is the user's text data, and the output is the prompt sentence sent to the generative AI model. Specifically, the server converts the text data into an appropriate format and sends it to the generative AI model using an API request.

[1162] Step 5:

[1163] A generative AI model generates a response based on the prompt.

[1164] The generative AI model generates an appropriate response based on the prompt sentence. The input is the prompt sentence, and the output is the generated response text. Specifically, the generative AI model uses a natural language processing algorithm to generate the response.

[1165] Step 6:

[1166] The server receives the generated response and sends it to the terminal.

[1167] The server sends the response received from the generative AI model to the terminal. The input is the generated response text, and the output is the response text sent to the terminal. Specifically, the server sends data using an HTTP response.

[1168] Step 7:

[1169] The terminal displays the response to the user.

[1170] The terminal displays the response received from the server in a chat window. The input is the response text received from the server, and the output is the response text displayed to the user. Specifically, the terminal renders the received data on the screen.

[1171] Step 8:

[1172] The user either inputs information to the AI ​​character again or browses the site.

[1173] The user continues to interact with the AI ​​character or browses other content on the news site. The input is the user's new text data or navigation on the site, and the output is new dialogue or new content displayed. Specifically, the user may enter more text into the chat window or click a link on the site.

[1174] (Application example 3)

[1175] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1176] Conventional news sites and virtual stores lacked the ingenuity to increase the time users spent on the site, resulting in users leaving the site after a short time. Furthermore, disputes between users and offensive comments were common, making it difficult to maintain sound site management. Furthermore, virtual stores lacked sufficient product explanations and responses to questions, making it difficult to stimulate users' desire to purchase.

[1177] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for introducing an AI character into a comment section of a news article, means for the AI ​​character to comment on the news article in a format similar to that of general users, means for the user to interact with the AI ​​character about the news article, means for the interaction to be visible only to the user, means for the user to post a "summary of the interaction" after the interaction has ended, means for the user to interact with the AI ​​character in a virtual store and receive product explanations and answer questions, and means for the AI ​​character to acknowledge an offensive comment made by the user and prevent trouble with other users. This increases the user's stay time, prevents trouble between users, and improves product explanations and question responses in the virtual store.

[1178] An "artificial intelligence character" is a computer program designed to interact with a user and respond to user input using natural language processing techniques.

[1179] A "virtual store" is a virtual store that exists on the Internet and provides an environment where users can browse and purchase products online.

[1180] A "comment section" is an area on web pages such as news sites and virtual stores where users can post their opinions and thoughts.

[1181] "Interaction" refers to the dialogue and communication that takes place between the user and the AI ​​character.

[1182] The "interaction summary" is a concise summary of the content of the conversation between the user and the AI ​​character.

[1183] "Aggressive comments" are comments posted with hostility or malice towards other users or the system.

[1184] "Preventing trouble" means preventing conflicts and disputes between users before they occur.

[1185] "Product explanation" refers to an AI character in a virtual store explaining the features and benefits of a product to the user.

[1186] "Question response" refers to providing appropriate answers to questions posed by users to AI characters.

[1187] As an embodiment of the present invention, a dialogue system using an AI character is constructed for a news site or a virtual store. A specific embodiment of the system is shown below.

[1188] System configuration

[1189] This system includes a server, a user terminal, and an AI character. The server manages data from news sites and virtual stores and communicates with the user terminal. The user terminal is a device such as a smartphone or head-mounted display, and provides an interface for users to interact with the system. The AI ​​character generates responses to user input using a generative AI model.

[1190] Program processing

[1191] The server executes a program that includes the following means:

[1192] 1. Introducing AI characters:

[1193] The server displays an AI character in the comment section of the news article or in a virtual store, and this character acts as a conversation partner when a user posts a comment.

[1194] 2. Comment generation method:

[1195] The server uses a generative AI model to have AI characters comment on news in a format similar to that of regular users, using natural language processing technologies such as the OpenAI API.

[1196] 3. Means of interaction:

[1197] Users can interact with AI characters about news and product information. The generative AI model generates appropriate responses to user input and displays them on the user's device.

[1198] 4. Private Communication Methods:

[1199] The interactions between users and AI characters are not visible to other users, thereby protecting privacy.

[1200] 5. Summary submission method:

[1201] After a conversation is over, users can post a "conversation summary," which is made publicly available to other users.

[1202] 6. Receptacles for offensive comments:

[1203] If a user makes an offensive comment, an AI character will receive the comment and prevent trouble with other users.

[1204] Hardware and software used

[1205] Hardware: Smartphone, head-mounted display, camera

[1206] Software: OpenAI API, OpenCV

[1207] Specific examples

[1208] When a user asks, "Tell me about this product" in the virtual store, an artificial intelligence character responds, "This product is the latest smartphone with excellent camera performance."

[1209] Example prompt sentence:

[1210] User: Tell me about this product.

[1211] Artificial Intelligence Character: This product is a state-of-the-art smartphone with an excellent camera.

[1212] In this way, users can enjoy a shopping experience in a virtual store while obtaining product information through dialogue with an AI character.

[1213] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1214] Step 1:

[1215] The server manages the data of news sites and virtual stores and communicates with user terminals. When a user accesses a news site or virtual store, the server sends the data of the corresponding page to the user terminal. This causes the news article or product information to be displayed on the user terminal.

[1216] Input: User's access request

[1217] Output: News articles and product information data

[1218] Step 2:

[1219] The user terminal displays the AI ​​character in the comment section of the news and in the virtual store. When the user posts a comment, the AI ​​character appears as a conversation partner.

[1220] Input: Data sent from the server

[1221] Output: Display of AI character

[1222] Step 3:

[1223] When a user enters a comment, the user's device sends the comment to the server, which uses a generative AI model to generate a response for the AI ​​character to comment on the news in a format similar to that of a regular user.

[1224] Input: User comment

[1225] Output: Response from the generative AI model

[1226] Step 4:

[1227] The server transmits the generated response to the user terminal, which displays the response of the AI ​​character in a comment field.

[1228] Input: Response by a generative AI model

[1229] Output: Response display of AI character

[1230] Step 5:

[1231] The user interacts with the AI ​​character about news and product information. The generative AI model generates an appropriate response to the user's input and sends it to the user's device via the server.

[1232] Input: User input

[1233] Output: Response from the generative AI model

[1234] Step 6:

[1235] The server uses private interaction methods to ensure that the user's interactions with the AI ​​character are not visible to other users, thereby protecting the user's privacy.

[1236] Input: Interaction between the user and the AI ​​character

[1237] Output: Realizing private dialogue

[1238] Step 7:

[1239] After a conversation is over, users can post a "conversation summary," which the server then makes available to other users.

[1240] Input: User Summary

[1241] Output: Publish summary

[1242] Step 8:

[1243] If a user makes an offensive comment, the server will have an AI character receive the comment and take measures to prevent trouble with other users. Using a generative AI model, the server will generate an appropriate response and send it to the user's device.

[1244] Input: offensive comment

[1245] Output: An appropriate response from the generative AI model

[1246] In this way, users can enjoy the experience of browsing news sites or virtual stores while obtaining information through dialogue with AI characters. This also prevents trouble caused by offensive comments.

[1247] 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.

[1248] "Example 1"

[1249] One embodiment of the present invention is a system incorporating an emotion engine. This system recognizes the user's emotions and adjusts the AI ​​character's response based on those emotions. Specifically, when the user expresses negative emotions such as anger or sadness, the AI ​​character accepts those emotions and responds in a way that soothes the user. Furthermore, when the user expresses positive emotions such as joy or excitement, the AI ​​character shares those emotions and responds in a way that amplifies the user's joy.

[1250] "Example 2"

[1251] The emotion engine also supports the generation of "interaction summaries." Specifically, it analyzes the user's emotions expressed during their interactions with the AI ​​character and generates a summary that reflects those emotions. For example, if the user expresses strong joy during their interaction with the AI ​​character, the summary will reflect the user's emotions in the form of "The user was very happy about the news."

[1252] "Example 3"

[1253] Furthermore, the emotion engine tracks changes in the user's emotions and uses that information to adjust the AI ​​character's responses over time. For example, if the user initially shows anger but gradually calms down as they interact with the AI ​​character, the AI ​​character will pick up on this change and adjust its own responses to match the user's changing emotions. This allows the AI ​​character to have more natural, human-like conversations.

[1254] The processing flow of each embodiment will be described below.

[1255] "Example 1"

[1256] Step 1: A user posts a comment on a news article.

[1257] Step 2: The emotion engine recognizes emotions from user comments.

[1258] Step 3: The AI ​​character generates a response based on the information from the emotion engine and posts a comment to the user.

[1259] "Example 2"

[1260] Step 1: After the user finishes interacting with the AI ​​character, the emotion engine analyzes the user's emotions from the interaction.

[1261] Step 2: The emotion engine generates a "summary of the interaction" based on the analysis results.

[1262] Step 3: The generated "interaction summary" is presented to the user.

[1263] "Example 3"

[1264] Step 1: When the user begins interacting with the AI ​​character, the emotion engine tracks changes in the user's emotions.

[1265] Step 2: The emotion engine communicates the user's emotional changes to the AI ​​character.

[1266] Step 3: The AI ​​character adjusts its response based on the information from the emotion engine and posts a comment to the user.

[1267] Example 1

[1268] Next, a description will be given of Example 1 of Form 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."

[1269] On traditional news sites, when users post comments on news articles, their interactions with other users are made public, which can lead to privacy issues and abusive comments. It is also difficult to provide appropriate responses based on users' emotions, which can hinder user satisfaction. Furthermore, there are a lack of ways to deepen users' understanding of the news, making it difficult to increase user time spent on the site.

[1270] 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.

[1271] In this invention, the server includes means for introducing an AI character into a comment section of a news article, means for the AI ​​character to comment on the news article in a format similar to that of general users, means for the user to interact with the AI ​​character about the news article, means for the interaction to be visible only to the user, means for the user to post a "summary of the interaction" after the interaction has ended, means for recognizing the user's emotions and adjusting the AI ​​character's response based on the emotions, and means for storing the generated comments and responses in a database. This makes it possible to provide appropriate responses according to the user's emotions, deepen the user's understanding of the news, and increase the user's stay time while protecting the user's privacy.

[1272] An "artificial intelligence character" is a virtual character equipped with artificial intelligence designed to interact with users and generate comments.

[1273] The "comment section" is an interface where users can post their opinions and thoughts about news articles.

[1274] "A format similar to comments made by ordinary users" means that comments generated by AI characters have the same style and content as comments posted by ordinary users.

[1275] An "interaction" is an exchange of messages between a user and an AI character.

[1276] The "summary of the interaction" is a sentence that briefly summarizes the content of the interaction between the user and the AI ​​character.

[1277] "Emotion recognition" means analyzing and identifying emotions from text entered by the user.

[1278] "Adjusting responses" means changing the response content and attitude of an AI character based on the user's recognized emotions.

[1279] A "generative AI model" is an artificial intelligence model that uses natural language processing to generate text.

[1280] The "database" is an information management system for storing generated comments, reactions, and interactions with users.

[1281] This invention is a system that introduces an AI character into a news site, deepens users' understanding of news articles through dialogue with them, and increases the time users spend on the site. Specific embodiments of this system are described below.

[1282] server

[1283] The server uses the news API to retrieve the latest news articles and store them in a database. Specifically, the server sends a request to the news API every hour and stores the retrieved news articles in a database.

[1284] The server uses a generative AI model (e.g., GPT-4) to generate a comment from an AI character in response to a user's comment. When a user enters a comment on a news article, the server sends the comment to the generative AI model and returns the generated comment to the user.

[1285] The server uses sentiment analysis software (e.g., IBM Watson Tone Analyzer) to recognize the user's emotions. It sends the text entered by the user to the sentiment analysis software and receives the results of the sentiment analysis.

[1286] The server uses a generative AI model to generate a response for the AI ​​character based on the results of the emotion analysis. For example, if a user inputs "I'm feeling very sad today," the server uses emotion analysis software to recognize the emotion of "sadness" and sends the result to the generative AI model to generate a response such as "That's terrible. Is there anything you'd like to talk about?"

[1287] The server stores the interactions between users and AI characters in a database and manages private interactions for each user.

[1288] Terminal

[1289] The device provides a user interface that allows users to view news articles and post comments. When a user visits a news site, the device displays a list of the latest news articles.

[1290] The terminal displays a comment input field and a chat window with an AI character for each news article. When a user inputs a comment for a news article, the comment is sent to the server.

[1291] The device displays the AI ​​character's responses in real time. For example, if a user types, "What do you think about the latest economic news?", a comment generated using the generative AI model will be displayed on the device.

[1292] User

[1293] Users browse news articles and begin interacting with the AI ​​character, who then comments on the article using the generative AI model.

[1294] Users enter text to express their feelings, for example, "I'm feeling very sad today," and emotion analysis software is used to recognize that emotion and generate an appropriate response.

[1295] After the user has finished interacting with the AI ​​character, they can enter and post a summary of the interaction. For example, they could write, "Through my interaction with the AI ​​character, I gained a deeper understanding of the current state of the economy."

[1296] Specific examples

[1297] The user asks the AI ​​character, "What do you think about recent economic news?" The generative AI model responds by saying, "Experts are divided on the current state of the economy, but recent data shows signs of recovery."

[1298] If a user types, "I'm feeling very sad today," the sentiment analysis software will recognize the emotion of "sadness," and the generative AI model will generate a response like, "That's terrible. Is there anything you'd like to talk about?"

[1299] Prompt Sentence Examples

[1300] "What do you think about the recent economic news?"

[1301] "I feel very sad today."

[1302] The above is a specific embodiment for carrying out the present invention. This system protects the user's privacy while providing appropriate responses according to their emotions, deepening the user's understanding of the news and increasing the time spent on the site.

[1303] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1304] Step 1: Get the news article

[1305] server

[1306] The server uses the news API to get the latest news articles. Specifically, the server sends a request to the news API every hour and stores the retrieved news articles in a database.

[1307] Input: A request to the News API

[1308] Output: Retrieved news article data

[1309] Specific operation: The server sends a request to the news API and stores the retrieved news articles in the database.

[1310] Step 2: Displaying the User Interface

[1311] Terminal

[1312] The device displays a list of news articles to the user. When the user visits a news site, the device displays a list of the latest news articles.

[1313] Input: News articles retrieved from a database

[1314] Output: A list of news articles in a user interface

[1315] Specific operation: The device retrieves news articles from the database and displays them in the user interface.

[1316] Step 3: User comments

[1317] User

[1318] A user inputs a comment on a news article. When a user inputs a comment on a news article, the comment is sent to the server.

[1319] Input: User comment

[1320] Output: Comment data sent to the server

[1321] Specific operation: The user enters a comment on a news article, such as "What do you think about this news?", and presses the send button.

[1322] Step 4: Generate comments for the AI ​​character

[1323] server

[1324] The server sends the user's comments to the generative AI model, which generates comments for the AI ​​character, and returns the generated comments to the user.

[1325] Input: User comment data

[1326] Output: Comments generated by the generative AI model

[1327] Specific operation: The server sends the user's comment "What do you think about recent economic news?" to the generative AI model, which then returns the generated comment "Experts are divided on the current state of the economy, but recent data shows signs of recovery" to the user.

[1328] Step 5: Saving the interaction between the user and the AI ​​character

[1329] server

[1330] The server stores the interactions between users and AI characters in a database and manages private interactions for each user.

[1331] Input: Interaction data between the user and the AI ​​character

[1332] Output: Interaction data stored in a database

[1333] Specific operation: The server stores the interactions between the user and the AI ​​character in a database so that the interactions can be displayed when the user accesses the site again.

[1334] Step 6: Post a summary of the conversation

[1335] User

[1336] After the user has finished interacting with the AI ​​character, they can enter a "summary of the interaction" and post it.

[1337] Input: A summary of the user interaction

[1338] Output: Posted summary data

[1339] Specific actions: The user enters a summary such as "Through my interactions with the AI ​​character, I gained a deeper understanding of the current state of the economy," and presses the post button.

[1340] Step 7: Recognize emotions

[1341] server

[1342] The server sends the user's input text to the sentiment analysis software and receives the results of the sentiment analysis.

[1343] Input: User-entered text

[1344] Output: Sentiment analysis result data

[1345] Specific operation: The server sends the user's input "I feel very sad today" to the sentiment analysis software and receives the result of the sentiment analysis as "sadness."

[1346] Step 8: Generating AI character reactions

[1347] server

[1348] The server uses a generative AI model to generate the AI ​​character's reaction based on the results of the emotion analysis.

[1349] Input: Sentiment analysis result data

[1350] Output: The response generated by the generative AI model

[1351] Specific operation: The server sends the emotion analysis result of "sadness" to the generation AI model, which generates a response saying, "That's terrible. Is there anything you'd like to say?"

[1352] Step 9: Showing User Feedback

[1353] Terminal

[1354] The device displays the AI ​​character's reactions to the user.

[1355] Input: Responses generated by a generative AI model

[1356] Output: AI character response on the user interface

[1357] What happens: The device displays to the user, "That's a problem. Is there anything else you'd like to talk about?"

[1358] (Application example 1)

[1359] Next, a description will be given of Application Example 1 of Form 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."

[1360] Traditional news sites lack support for users to gain a deeper understanding of the content of news articles when they read them. It is also difficult for users to receive appropriate feedback based on their emotions toward the news, limiting the user experience. Furthermore, there is a lack of ingenuity to increase the amount of time users spend on news sites, which can lead to a decline in site usage. To address these issues, a system is needed that allows users to gain a deeper understanding of news articles and receive appropriate feedback based on their emotions.

[1361] 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.

[1362] In this invention, the server includes: means for introducing an AI character into a news comment section; means for the AI ​​character to comment on the news in a format similar to that of general users; means for the user to interact with the AI ​​character about the news; means for the interaction to be visible only to the user; means for the user to post a "summary of the interaction" after the interaction is over; means for recognizing the user's emotions using an emotion engine and adjusting the AI ​​character's response based on the emotions; means including an application to be installed on a smartphone; means for generating comments on news articles using a generative AI model; and means for inputting prompt sentences into the generative AI model. This allows users to gain a deeper understanding of news articles and receive appropriate feedback based on their emotions. It also increases user time spent on the site, improving site usage.

[1363] A "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[1364] An "AI character" is a character that interacts with users using artificial intelligence.

[1365] "A format similar to comments made by general users" means a format or style similar to that of comments posted by regular users.

[1366] "Means by which the user can interact with the AI ​​character about the news" refers to a function that allows the user to have a dialogue with the AI ​​character about news articles.

[1367] "Means for ensuring that the interaction is visible only to the user" refers to a function that prevents the content of the conversation between the user and the AI ​​character from being displayed to other users.

[1368] The "interaction summary" is a concise summary of the conversation between the user and the AI ​​character.

[1369] An "emotion engine" is a technology that recognizes a user's emotions and generates an appropriate response based on those emotions.

[1370] An "application installed on a smartphone" is a software program that runs on a smartphone.

[1371] A "generative AI model" is a model that uses artificial intelligence to generate text and data.

[1372] A "prompt" is a text instruction that is input to a generative AI model.

[1373] A system for implementing this invention introduces an AI character into the comment section of a news article, allowing users to interact with the AI ​​character regarding the news article. A specific embodiment of this system will be described below.

[1374] System configuration

[1375] Hardware

[1376] Server: Manages the news article database and generates comments and recognizes emotions for the AI ​​characters.

[1377] Smartphone: A device that allows users to view news articles and interact with AI characters.

[1378] software

[1379] Generative AI model: An artificial intelligence model for generating comments on news articles. Specifically, it uses OpenAI's API.

[1380] Emotion Engine: Software to recognize user emotions and adjust the AI ​​character's response based on those emotions. EmotionRecognizer is used.

[1381] Smartphone application: An application that provides features such as displaying news articles, interacting with AI characters, emotion recognition, and comment generation.

[1382] System Operation

[1383] 1. News article retrieval: The server retrieves the article from the news article database and sends it to the smartphone.

[1384] 2. Comment generation: The smartphone application uses the generative AI model to generate a comment for the AI ​​character on the news article. The generative AI model is given the following prompt:

[1385] Example prompt: "News article: A new technology has been announced. AI character comments:"

[1386] 3. Emotion Recognition: Smartphone applications use emotion engines to recognize emotions from user input and facial expressions.

[1387] 4. Comment adjustment: Adjust the AI ​​character's comments based on the user's recognized emotions. For example, if the user is "happy," add a positive element to the comment, such as "That's great!"

[1388] 5. Displaying dialogue: The adjusted comments are displayed on the smartphone screen, allowing users to interact with the AI ​​character. This dialogue is set to be invisible to other users.

[1389] 6. Interaction Summary: After a conversation has ended, users can post a summary of the conversation.

[1390] Specific examples

[1391] News article: "New technology announced."

[1392] User sentiment: "happy"

[1393] AI character comment: "Amazing! This technology could change the future."

[1394] This system allows users to gain a deeper understanding of news articles and receive appropriate feedback based on their emotions, while also increasing user time on the site and improving site usage.

[1395] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1396] Step 1:

[1397] The server retrieves the latest news articles from a news article database and sends them to the smartphone.

[1398] Input: A database of news articles

[1399] Output: News article data

[1400] What it does: The server queries the database to retrieve the latest news articles, then sends them to the smartphone.

[1401] Step 2:

[1402] The smartphone application uses a generative AI model to generate AI character comments on news articles.

[1403] Input: News article data, prompt

[1404] Output: AI character's comment

[1405] How it works: The smartphone application inserts the content of the news article into a prompt sentence and inputs it into the generative AI model, which then generates a comment for the AI ​​character based on the prompt sentence.

[1406] Step 3:

[1407] Smartphone applications use emotion engines to recognize emotions from user input and facial expressions.

[1408] Input: User input data, user facial expression data

[1409] Output: User sentiment data

[1410] Specific operation: The smartphone application inputs the user's text input and facial expression data acquired through the camera into the emotion engine, which then analyzes this data and recognizes the user's emotions.

[1411] Step 4:

[1412] The smartphone application adjusts the AI ​​character's comments based on the user's perceived emotions.

[1413] Input: AI character's comments, user's emotional data

[1414] Output: Adjusted AI character's comments

[1415] How it works: The smartphone application adjusts the AI ​​character's comments based on the user's emotional data. For example, if the user is "happy," it will add a positive element to the comment, such as "That's great!"

[1416] Step 5:

[1417] The smartphone application displays tailored comments on the screen, allowing users to interact with the AI ​​character.

[1418] Input: Adjusted AI character's comment

[1419] Output: Comments displayed on the smartphone screen

[1420] How it works: The smartphone application displays tailored comments in the user interface, allowing users to interact with the AI ​​character, while keeping this interaction invisible to other users.

[1421] Step 6:

[1422] Users can post summaries of their interactions after they finish.

[1423] Input: Dialogue between the user and the AI ​​character

[1424] Output: A summary of the conversation

[1425] How it works: The smartphone application summarizes the conversation between the user and the AI ​​character and allows the user to post those summaries.

[1426] Example 2

[1427] Next, a description will be given of Example 2 of Form 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."

[1428] Conventional news comment systems make it difficult for users to gain a deep understanding of a news article when posting comments. Furthermore, users are unable to receive appropriate feedback on their posted comments, which reduces the time users spend on the site. Furthermore, there is a risk of offensive comments being posted. To address these issues, a system is needed that deepens users' understanding of a news article, provides appropriate feedback, and increases the time users spend on the site.

[1429] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1430] In this invention, the server includes means for introducing an AI character into a comment section of a news article, means for the AI ​​character to comment on the news in a format similar to that of general users, means for the user to interact with the AI ​​character about the news, means for the interaction to be visible only to the user, means for the user to post a "summary of the interaction" after the interaction is over, means for the AI ​​character to analyze the news article using natural language processing technology and generate comments, and means for an emotion engine to analyze the user's emotions and generate a summary that reflects those emotions. This allows the user to deepen their understanding of the news article, provide appropriate feedback, and increase the user's stay time.

[1431] The "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[1432] An "artificial intelligence character" is a virtual entity that interacts with users using natural language processing technology.

[1433] "A format similar to comments made by general users" refers to comments that have the same format and writing style as comments posted by regular users.

[1434] "Interaction" refers to the dialogue and communication that takes place between the user and the AI ​​character.

[1435] The "summary of the interaction" is a sentence that briefly summarizes the content of the interaction between the user and the AI ​​character.

[1436] "Natural language processing technology" is a technology that enables computers to understand and generate human language.

[1437] An "emotion engine" is an algorithm or system that analyzes a user's emotions and generates information that reflects those emotions.

[1438] A "news article" is a piece of text that describes current events or information provided by a news organization or news site.

[1439] "Generating comments" means that an AI character automatically creates opinions and questions about news articles.

[1440] "Analyzing emotions" means identifying and understanding emotions from a user's statements and actions.

[1441] The present invention provides a system that deepens a user's understanding of a news article, provides appropriate feedback, and increases the user's stay time. A specific embodiment of this system is described below.

[1442] The server collects the latest news articles from news sites. Specifically, it obtains article data using RSS feeds and APIs. The collected news articles are then analyzed using natural language processing techniques. Generative AI models used here include BERT and GPT-3.

[1443] The device provides an interface for users to enter comments or questions about news articles. Once the user has completed their input, the data is sent to the server. The server receives the user's input and uses a generative AI model to generate appropriate comments. The generated comments are then displayed to the user via the device.

[1444] When a user responds to a comment from an AI character, that data is also sent to the server. The server receives the user's response and uses the generative AI model to generate an appropriate response, which is also displayed to the user via their device.

[1445] The emotion engine analyzes the user's interactions with the AI ​​character and extracts the user's emotions. The emotion engine uses a sentiment analysis algorithm to identify the user's emotions and outputs the results as a summary. For example, if a user says, "I think this bill will be very beneficial to small and medium-sized businesses," the emotion engine extracts the emotion of joy and generates the summary, "The user was very happy about this news."

[1446] As a concrete example, the following exchange can be considered:

[1447] The news article read: "A new bill has been decided to introduce tax incentives for small and medium-sized enterprises."

[1448] AI character says: "How do you think this bill will affect small businesses?"

[1449] User Response: "I think this bill will be very beneficial to small businesses. The tax incentives will help ease their financial burden."

[1450] AI character responds: "Yes, tax incentives can encourage small business growth."

[1451] Sentiment Engine Summary: "Users were very happy about the news."

[1452] In this way, the system generates summaries of comments and emotions about news articles in cooperation with the server, terminals, and users.

[1453] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1454] Step 1:

[1455] The server collects the latest news articles from news sites. Specifically, it retrieves article data using RSS feeds and APIs. It uses the news site's URL and API key as input and obtains news article data as output. For example, a scheduled task can be run every morning at 8:00 to collect the latest political news from major news sites.

[1456] Step 2:

[1457] The server analyzes the collected news articles using natural language processing technology. Specifically, it uses generative AI models such as BERT and GPT-3 to understand the content of the articles. It uses the news article data as input and extracts the article's summary and main topics as output. For example, it generates a summary such as "It has been decided that a new bill will introduce tax incentives for small and medium-sized enterprises."

[1458] Step 3:

[1459] The terminal provides an interface for users to input comments and questions about news articles. It receives a news article summary and the user's comments and questions as input, and sends the data to the server as output. For example, it displays a news article summary to the user and provides an input field for "What do you think about this news?"

[1460] Step 4:

[1461] The server receives input from the user and generates appropriate comments using a generative AI model. It uses the user's comments or questions as input and gets the generated comments as output. For example, if the user enters "I think this bill will be very beneficial to small businesses," the server generates the comment "How do you think this bill will affect small businesses?"

[1462] Step 5:

[1463] The terminal displays the AI ​​character's comments to the user and provides an interface for the user to reply to them. The terminal receives the AI ​​character's comments and the user's responses as input, and sends the data back to the server as output. For example, the AI ​​character's comments may be displayed to the user, and the user may reply, "The introduction of tax incentives will help reduce the economic burden."

[1464] Step 6:

[1465] The server receives the user's response and generates an appropriate response using a generative AI model. It uses the user's response as input and gets the generated response as output. For example, based on the user's response, it generates the response, "Yes, tax incentives can encourage the growth of small and medium-sized businesses."

[1466] Step 7:

[1467] The server passes the user's interactions with the AI ​​character to an emotion engine to analyze the user's emotions. The data on the user's interactions with the AI ​​character is used as input, and a summary reflecting the emotions is generated as output. For example, the emotion of joy can be extracted from the part where the user says, "I think this bill will be very beneficial to small and medium-sized businesses," generating a summary that reads, "The user was very happy about this news."

[1468] (Application example 2)

[1469] Next, a description will be given of Application Example 2 of Form 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."

[1470] Conventional news comment systems have the problem that when users post comments on news articles, it is difficult to get reactions and feedback from other users, making it difficult to deepen their understanding of the news article. Additionally, there is no function to generate summaries that reflect the user's feelings, and there is also the issue that feedback that appropriately reflects the user's feelings cannot be obtained.

[1471] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1472] In this invention, the server includes means for introducing an AI character into a comment section of a news article, means for the AI ​​character to comment on the news in a format similar to that of general users, means for the user to interact with the AI ​​character about the news, means for the interaction to be visible only to the user, means for the user to post a "summary of the interaction" after the interaction is over, means for analyzing the user's emotions using an emotion engine and generating a summary that reflects the emotions, means for generating comments on the news article using a generative AI model, and means for generating comments for the AI ​​character using a prompt sentence. This allows the user to gain a deeper understanding of the news article and provides feedback that reflects the user's emotions.

[1473] A "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[1474] An "AI character" is a virtual character that interacts with users using artificial intelligence.

[1475] "A format similar to comments made by general users" means a format or style similar to that of comments posted by regular users.

[1476] "Means to comment on news" is a function that allows users to post opinions and thoughts about news articles.

[1477] "Means for users to interact with the AI ​​character about the news" refers to a function that allows users to exchange opinions about news articles through dialogue with the AI ​​character.

[1478] "Means for the interaction to be visible only to the user" refers to a function that prevents other users from seeing the content of the conversation between the user and the AI ​​character.

[1479] "A means to post a 'summary of the conversation' after the conversation is over" is a function that displays a summary of the conversation between the user and the AI ​​character.

[1480] An "emotion engine" is software that analyzes a user's emotions and generates information that reflects those emotions.

[1481] A "generative AI model" is an algorithm or system that uses artificial intelligence to generate text or comments.

[1482] A "prompt" is an instruction or question input to a generative AI model.

[1483] The system for implementing this invention introduces an AI character into the comment section of a news article, allowing users to post comments on the news article and interact with the AI ​​character. A specific embodiment of this system is described below.

[1484] System configuration

[1485] The system consists of the following main components:

[1486] 1. Server: Manages news articles, generates comments for AI characters, and manages interactions with users.

[1487] 2. User device: A device such as a smartphone or tablet that users use to view news articles and post comments.

[1488] 3. Generative AI model: An artificial intelligence model for generating comments on news articles.

[1489] 4. Sentiment Engine: Software for analyzing the sentiment of user comments and interactions and generating summaries.

[1490] Program processing

[1491] The server distributes news articles to user devices, allowing users to view the articles. When a user posts a comment on a news article, the generative AI model uses the prompt text to generate a comment from the AI ​​character. The generated comment is displayed on the user device, allowing the user to interact with the AI ​​character.

[1492] The emotion engine analyzes the user's interaction with the AI ​​character in real time and generates a summary that reflects the user's emotions, which is provided to the user after the interaction is over.

[1493] Hardware and software used

[1494] Hardware: smartphones, tablets, servers

[1495] Software: Generative AI models (e.g., OpenAI API), emotion engines (e.g., TextBlob)

[1496] Specific examples

[1497] For example, if the news article is "A new bill has been passed in the Diet," the server will input the following prompt sentence into the generative AI model:

[1498] News article: A new bill has been passed in Parliament.

[1499] How do you think this bill will affect small businesses?

[1500] The generative AI model generates comments for the AI ​​character based on this prompt and displays them on the user's device. If the user comments, "I think this bill will be very beneficial to small and medium-sized businesses," the emotion engine analyzes this comment and generates a summary that reflects the user's sentiment.

[1501] In this way, users can gain a deeper understanding of the news article and receive emotional feedback.

[1502] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1503] Step 1:

[1504] The server retrieves news articles from the database and delivers them to user devices. The input is the news article data, and the output is the news article displayed on the user device. Specifically, the server retrieves the text data of the news article and sends it to the user device.

[1505] Step 2:

[1506] Users browse news articles and post comments. The input is the news article the user browses, and the output is the comment the user posts. Specifically, users read a news article and enter their opinions and thoughts in the comment section.

[1507] Step 3:

[1508] The server receives the user's comments and sends a prompt to the generative AI model. The input is the user's comments and the news article text, and the output is the prompt to be sent to the generative AI model. Specifically, the server combines the user's comments and the news article to generate a prompt and sends it to the generative AI model.

[1509] Step 4:

[1510] The generative AI model generates comments for the AI ​​character based on the prompt sentence. The input is the prompt sentence, and the output is the AI ​​character's comment. Specifically, the generative AI model analyzes the prompt sentence and generates an appropriate comment.

[1511] Step 5:

[1512] The server sends the generated AI character's comments to the user's device. The input is the comment from the generative AI model, and the output is the AI ​​character's comment displayed on the user's device. Specifically, the server receives the generated comment and sends it to the user's device.

[1513] Step 6:

[1514] The user posts a further comment in response to the AI ​​character's comment. The input is the AI ​​character's comment, and the output is the user's new comment. Specifically, the user reads the AI ​​character's comment and then re-enters their own opinion or impression in response to it.

[1515] Step 7:

[1516] The server sends the interaction between the user and the AI ​​character to the emotion engine and performs emotion analysis. The input is the text of the interaction between the user and the AI ​​character, and the output is the result of the emotion analysis. Specifically, the server sends the text of the interaction to the emotion engine and performs emotion analysis.

[1517] Step 8:

[1518] The emotion engine generates summaries that reflect the user's emotions. The input is the result of the emotion analysis and the text of the interaction, and the output is a summary that reflects the emotions. Specifically, the emotion engine generates summaries that reflect the user's emotions based on the results of the emotion analysis.

[1519] Step 9:

[1520] The server sends the generated summary to the user terminal. The input is the summary from the emotion engine, and the output is the summary displayed on the user terminal. Specifically, the server receives the generated summary and sends it to the user terminal.

[1521] Step 10:

[1522] Users view summaries of interactions and deepen their understanding of the news article. The input is a summary that reflects sentiment, and the output is a deepening of the user's understanding. Specifically, users read summaries and deepen their understanding of the news article.

[1523] Example 3

[1524] Next, a description will be given of Example 3 of Form Example 3. 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."

[1525] Conventional news sites lack effective methods for increasing user time spent on the site, making it difficult to maintain user interest. Furthermore, aggressive posts and other troubles between users are common, and appropriate measures to prevent this are needed. Furthermore, it has been difficult to realize natural dialogue that responds to changes in user emotions.

[1526] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1527] In this invention, the server includes means for introducing an AI character into a comment section of a news item, means for the AI ​​character to comment on the news item in a format similar to that of general users, means for the user to interact with the AI ​​character about the news item, means for the interaction to be visible only to the user, means for the user to post a "summary of the interaction" after the interaction has ended, means for tracking the user's emotions, and means for adjusting the AI ​​character's response based on the emotion tracking information. This increases the user's stay time, serves as a receptacle for aggressive posts, and makes it possible to realize natural dialogue that responds to changes in the user's emotions.

[1528] An "artificial intelligence character" is a computer program designed to interact with a user and respond to user input using natural language processing techniques.

[1529] A "comment section" is an interface on a news site where users can post their opinions and thoughts.

[1530] "Emotion tracking" is a technology that analyzes user input data and identifies the user's emotional state in real time.

[1531] The "summary of the interaction" is information that briefly summarizes the content of the interaction between the user and the AI ​​character.

[1532] "Means to comment on news" is a function that allows an AI character to post opinions and thoughts on news articles.

[1533] "User emotion tracking means" refers to software or algorithms that analyze user input data to identify emotional states.

[1534] The "means for adjusting the response" is a function that changes the response content of the AI ​​character depending on the emotional state of the user.

[1535] "Means to increase dwell time" are features and technologies that encourage users to stay on a news site for a longer period of time.

[1536] "A means to receive offensive posts" is a function in which an artificial intelligence character receives offensive comments in order to prevent trouble between users.

[1537] This invention is a system for increasing the time users spend on news sites and preventing conflicts between users. This system uses an AI character to converse with users and adjusts its responses according to changes in the user's emotions, achieving a more natural and human-like dialogue.

[1538] The server runs software to introduce an AI character into the comment section of a news article. This AI character has the ability to comment on the news in a format similar to that of a general user. Users can interact with the AI ​​character about the news through their devices. This interaction is designed to be visible only to the user.

[1539] The server receives the user's input data and uses an emotion analysis library (e.g., IBM Watson Tone Analyzer) for emotion tracking. This emotion analysis library analyzes the user's input data and identifies the user's emotional state. For example, if a user inputs "I'm angry about the recent news," the emotion analysis library will detect "anger."

[1540] The server then uses a generative AI model (e.g., OpenAI GPT-4) to generate a response for the AI ​​character based on the user's emotional state. For example, if the user expresses anger, the generative AI model might generate a response such as, "That's tough. What's bothering you?" This generated response is sent from the server to the device and displayed to the user.

[1541] The user can continue the conversation by checking the AI ​​character's response through the device and inputting again. For example, the user can input, "I'm particularly interested in political news." In this way, the AI ​​character's response can be adjusted based on the user's emotional changes, which has the effect of increasing the user's stay time.

[1542] As a concrete example, the following prompt sentence is shown.

[1543] Example prompt sentence:

[1544] User: I'm angry about the recent news.

[1545] AI Character: That's tough. What are you particularly concerned about?

[1546] This system realizes natural dialogue that responds to changes in the user's emotions, increases the user's stay time, and prevents trouble between users by acting as a receptacle for offensive posts. The flow of the identification process in Example 3 will be described with reference to Fig. 21.

[1547] Step 1:

[1548] A user visits a news site.

[1549] A user accesses a news site using a web browser. The news site has an interactive function with an artificial intelligence character. The input is the user's access, and the output is the display of the news site.

[1550] Step 2:

[1551] The terminal receives user input.

[1552] A user inputs a comment on a news article. For example, the user inputs the text "I'm angry about the recent news." The input is the user's text data, and the output is the input data saved on the device.

[1553] Step 3:

[1554] The terminal sends the input data to the server.

[1555] The terminal sends the text data entered by the user to the server. At this time, the data is encrypted before being sent. The input is the user's text data, and the output is the data sent to the server.

[1556] Step 4:

[1557] The server analyzes the input data using an emotion engine.

[1558] The server analyzes the user's input data using an emotion analysis library (e.g., IBM Watson Tone Analyzer). The analysis results identify the user's emotional state (e.g., anger, joy, sadness). The input is the text data sent to the server, and the output is the emotion analysis result.

[1559] Step 5:

[1560] The server uses the generated AI model to generate the responses of the AI ​​character.

[1561] The server uses a generative AI model (e.g., OpenAI GPT-4) to generate a response for an AI character based on the user's emotional state. For example, if the user expresses anger, the generative AI model generates a response such as "That's terrible. What are you particularly concerned about?" The input is the emotion analysis result, and the output is the generated response.

[1562] Step 6:

[1563] The server sends the generated response to the terminal.

[1564] The server sends the generated AI character's response to the terminal. The data is encrypted before transmission. The input is the generated response, and the output is the data sent to the terminal.

[1565] Step 7:

[1566] The terminal displays the AI ​​character's response to the user.

[1567] The terminal displays the AI ​​character's response received from the server to the user. The user checks the AI ​​character's message on the screen. The input is the data sent to the terminal, and the output is the message displayed to the user.

[1568] Step 8:

[1569] The user re-enters and continues the dialogue.

[1570] The user sees the AI ​​character's reaction and inputs again, continuing the dialogue. For example, the user inputs "I'm particularly interested in political news." The input is the user's new text data, and the output is the new input data saved on the device.

[1571] (Application example 3)

[1572] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1573] The challenges facing news sites are increasing the amount of time users spend on the site and preventing conflicts between users. There is also a need for ways for users to gain a deeper understanding of news articles and moderate offensive posts. Furthermore, realizing natural dialogue that responds to users' emotions is also an important challenge.

[1574] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1575] In this invention, the server includes a means for introducing an AI character into the comment section of a news article, a means for the AI ​​character to comment on the news in a format similar to that of general users, and a means for users to interact with the AI ​​character about the news, thereby increasing the time users spend on the site, preventing conflicts between users, and deepening their understanding of the news article.

[1576] A "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[1577] An "AI character" is a virtual character that uses artificial intelligence to interact with users.

[1578] "Format similar to comments from general users" refers to comments that have the same format and style as comments posted by regular users.

[1579] "Means by which a user can interact with the AI ​​character about the news" refers to a function that allows a user to have a conversation with an AI character about the news.

[1580] "Means for ensuring that the interaction is visible only to the user" refers to a function that prevents the content of the conversation between the user and the AI ​​character from being displayed to other users.

[1581] The "interaction summary" is a concise summary of the conversation between the user and the AI ​​character.

[1582] "Means of tracking the user's emotions and adjusting the AI ​​character's response accordingly" refers to a function that detects changes in the user's emotions and changes the AI ​​character's response based on those changes.

[1583] "A means of briefly explaining the contents of a news article" is a function that clearly explains the main points of a news article.

[1584] To put this invention into practice, it is necessary to build a system that introduces AI characters into the comments section of a news website. In this system, when a user browses a news article and posts a comment, the AI ​​character will explain the news in a dialogue format and adjust its response according to the user's emotions.

[1585] Hardware and Software Configuration

[1586] Hardware:

[1587] server

[1588] User devices (smartphones, tablets, computers)

[1589] software:

[1590] Python

[1591] OpenAI API

[1592] Sentiment analysis library (e.g., sentiment_analysis)

[1593] System Operation

[1594] server:

[1595] The server manages the content of news articles and accepts user comments. It also runs a dialogue engine for the AI ​​character and includes a sentiment analysis engine for tracking user emotions. When the server receives a user comment, it analyzes the sentiment of the comment using the sentiment analysis engine and generates a response for the AI ​​character based on the results.

[1596] User device:

[1597] The user device displays news articles and provides an interface for users to post comments. When a user posts a comment, the comment is sent to the server, where it is processed. The resulting AI character's response is displayed on the user device.

[1598] Data processing and calculation

[1599] Sentiment analysis:

[1600] When the server receives a user's comment, it analyzes the sentiment of the comment using a sentiment analysis library, which takes the comment text as input and outputs a positive, negative, or neutral sentiment.

[1601] AI character reaction generation:

[1602] Based on the results of the sentiment analysis, the server uses the OpenAI API to generate responses for the AI ​​character, which are then adjusted appropriately according to the user's emotions.

[1603] Specific examples

[1604] Handling user comments:

[1605] If a user posts a comment like "I think this news is really terrible," the sentiment analysis library will identify this comment as negative. The server will send the following prompt to the OpenAI API to generate a response for the AI ​​character:

[1606] A user made a negative comment: I think this news is really terrible.

[1607] AI character reactions:

[1608] News article explanation:

[1609] To provide a brief explanation of the news article, the server sends the following prompt to the OpenAI API to generate an explanation:

[1610] Briefly explain what the news article is about: Today's news reported that the economy is recovering faster than expected.

[1611] This will help users understand the main points of news articles more easily and deepen their understanding of the news through dialogue with the AI ​​characters. The AI ​​characters will also act as a receptacle for offensive posts, preventing disputes between users.

[1612] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1613] Step 1:

[1614] Users view news articles and post comments.

[1615] Input: User comment (e.g., "I think this news is really terrible.")

[1616] Output: The user's comment is sent to the server.

[1617] Specific operation: When a user enters a comment in the comment field of a news article and presses the send button, the comment is sent to the server.

[1618] Step 2:

[1619] The server receives the user's comments and performs sentiment analysis.

[1620] Input: User comment (e.g., "I think this news is really terrible.")

[1621] Output: Sentiment analysis result (e.g., negative)

[1622] Specific operation: The server analyzes the sentiment of user comments using a sentiment analysis library. The sentiment analysis library takes the comment text as input and outputs a sentiment, either positive, negative, or neutral.

[1623] Step 3:

[1624] The server generates the AI ​​character's response based on the emotion analysis results.

[1625] Input: Sentiment analysis results (e.g., negative), user comments (e.g., "I think this news is really terrible.")

[1626] Output: AI character's response (e.g. "That's unfortunate. What did you find particularly disappointing?")

[1627] Specific operation: The server uses the OpenAI API to generate prompts based on the results of sentiment analysis and generate responses from the AI ​​character. Example prompt: "A user made a negative comment: I think this news is really terrible. The AI ​​character's response:"

[1628] Step 4:

[1629] The server sends the reaction of the generated AI character to the user's device.

[1630] Input: AI character's response (e.g., "That's a shame. What exactly did you find bad about it?")

[1631] Output: AI character's reaction displayed on the user's device

[1632] Specific operation: The server sends the reaction of the generated AI character to the user's device, which displays it.

[1633] Step 5:

[1634] The user continues to interact with the AI ​​character.

[1635] Input: User's additional comments (e.g., "I think it's terrible that the economy is recovering so slowly.")

[1636] Output: The additional comments are sent to the server for further sentiment analysis and reaction generation.

[1637] Specific operation: When the user enters additional comments in response to the AI ​​character's reaction and presses the send button, the comments are sent to the server again, and the process of sentiment analysis and reaction generation is repeated.

[1638] Step 6:

[1639] The server provides a brief explanation of the news article.

[1640] Input: The content of the news article (e.g., "Today's news reported that the economy is recovering faster than expected.")

[1641] Output: A description of the news article (e.g., "It has been reported that the economy is recovering faster than expected. This shows that the government's economic policies are working.")

[1642] Specific operation: The server uses the OpenAI API to generate a prompt that briefly explains the content of the news article, and then generates an explanation. Example of a prompt: "Please briefly explain the content of the news article: Today's news reported that the economy is recovering faster than expected."

[1643] Step 7:

[1644] The server transmits the generated commentary of the news article to the user terminal.

[1645] Input: A description of the news article (e.g., "The economy is reported to be recovering faster than expected. This shows that the government's economic policies are working.")

[1646] Output: A description of the news article displayed on the user's device

[1647] Specific operation: The server sends the generated commentary of the news article to the user terminal, which displays it.

[1648] 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.

[1649] 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> ) 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.

[1650] Another example of generative AI is Gemini (internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[1651] 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.

[1652] [Third embodiment]

[1653] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1654] 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.

[1655] 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.

[1656] The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and 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).

[1657] 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.

[1658] 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.

[1659] 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).

[1660] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1661] 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.

[1662] 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.

[1663] 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.

[1664] 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.

[1665] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1666] "Example 1"

[1667] One embodiment of the present invention is a system that introduces an AI character to a news site. In this system, the AI ​​character posts comments on news articles. The AI ​​character's comments are in a format similar to that of general users, and users can exchange messages with the AI ​​character about the news. These exchanges are set to be visible only to the user. After the exchange is over, the user can post a "summary of the exchange." "Example 2"

[1668] As a specific example, consider an AI character posting a comment on a political news article. The AI ​​character understands the content of the article and generates a comment based on that understanding. For example, the AI ​​character can post a question such as, "How do you think this bill will affect small and medium-sized businesses?". Users can express their opinion in response to this question, and the AI ​​character will respond with an appropriate response. This interaction helps users gain a deeper understanding of the news article.

[1669] "Example 3"

[1670] The system of this invention also has the effect of increasing the time users spend on the site. As users enjoy interacting with the AI ​​character, they spend more time on the news site. Furthermore, the AI ​​character acts as a receptacle for offensive posts, preventing conflict between users.

[1671] The processing flow of each embodiment will be described below.

[1672] "Example 1"

[1673] Step 1: The AI ​​character analyzes the news article, including the article's text content, associated images and videos, and other metadata.

[1674] Step 2: The AI ​​character generates comments on the news article, which are formatted similarly to typical user comments.

[1675] Step 3: The user responds to the AI ​​character's comments. The user can express their own opinions and thoughts to the AI ​​character.

[1676] Step 4: The AI ​​character generates an appropriate reply to the user's reaction. This reply is a response to the user's comment and continues the dialogue between the user and the AI ​​character.

[1677] Step 5: The user submits a summary of the conversation. This summary is a concise summary of the conversation between the user and the AI ​​character.

[1678] "Example 2"

[1679] Step 1: The AI ​​character analyzes political news articles, including the article's text content, associated images and videos, and other metadata.

[1680] Step 2: The AI ​​character generates question-type comments on news articles, such as "How do you think this bill will affect small businesses?"

[1681] Step 3: The user gives their opinion in response to the AI ​​character's question.

[1682] Step 4: The AI ​​generates an appropriate response to the user's opinion. This response is a reaction to the user's opinion and continues the dialogue between the user and the AI.

[1683] Step 5: The user submits a summary of the conversation. This summary is a concise summary of the conversation between the user and the AI ​​character.

[1684] "Example 3"

[1685] Step 1: A user visits a news site.

[1686] Step 2: The user initiates a dialogue with the AI ​​character, which is based on a news article.

[1687] Step 3: Users enjoy interacting with the AI ​​character, which increases the time they spend on the news site.

[1688] Step 4: If a user posts something offensive, the AI ​​character will respond, preventing any conflict between users.

[1689] Example 1

[1690] Next, a description will be given of Example 1 of Form 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."

[1691] On traditional news sites, when users post comments on news articles, their interactions with other users are made public, raising the risk of privacy issues and offensive comments. Furthermore, there is a lack of support for users to gain a deeper understanding of the content of news articles, which can lead to a shallow understanding of the articles. Furthermore, there is a lack of efforts to increase user time on the site, which can lead to a decline in site usage.

[1692] 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.

[1693] In this invention, the server includes: means for introducing an AI character into a news comment section; means for the AI ​​character to comment on the news in a format similar to that of general users; means for the user to interact with the AI ​​character about the news; means for the interaction to be visible only to the user; means for the user to post a "summary of the interaction" after the interaction is over; means for generating comments on news articles using a generative AI model; means for inputting user input as a prompt sentence into the generative AI model to generate a reply; and means for retrieving news articles. This allows users to comment on news articles while maintaining their privacy and deepen their understanding of the news article through their interaction with the AI ​​character. It also increases user dwell time and reduces the risk of offensive comments.

[1694] A "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[1695] An "artificial intelligence character" is a virtual character generated using artificial intelligence technology that can interact with users.

[1696] A "format similar to comments from general users" is a format that has the same style and content as comments posted by regular users.

[1697] "User" means an individual or organization that uses a news site.

[1698] "Interaction" refers to the dialogue and communication that takes place between the user and the AI ​​character.

[1699] The "interaction summary" is a concise summary of the interaction between the user and the AI ​​character.

[1700] A "generative AI model" is a model for generating text using artificial intelligence technology, including those that use natural language processing technology.

[1701] A "prompt" is an instruction or question to be input into a generative AI model.

[1702] A "news article" is a piece of text containing news or information that is published on a news site.

[1703] A "server" is a computer system used to operate a news site and process data.

[1704] "User session information" is information that records the status and activities of a user while the user is accessing a news site.

[1705] A "database" is a system for storing data such as news articles, user comments, and summaries of interactions.

[1706] The present invention is a system that introduces an AI character to a news site, allowing users to post comments on news articles and interact with the AI ​​character. This system includes elements of a server, a terminal, and a user.

[1707] System Overview

[1708] The server retrieves the latest news articles from news sites and uses a generative AI model to generate comments on the news articles. The generated comments are in a format similar to comments made by general users and are only displayed to specific users. Users can use their devices to interact with the AI ​​character about the news article. After the interaction is complete, users can post a "summary of the interaction."

[1709] Hardware and software used

[1710] Server: Retrieves news articles, runs generative AI models, generates and displays comments, and manages user sessions.

[1711] Device: The device (PC, smartphone, etc.) on which users view news articles, post comments, and interact with AI characters.

[1712] Generative AI models: Artificial intelligence models for generating comments on news articles (e.g., OpenAI's GPT-4).

[1713] Data processing and calculation

[1714] 1. Get news articles: The server uses the news site's API to get the latest news articles and stores them in a database.

[1715] 2. Comment generation: The server inputs the content of the retrieved news article as a prompt sentence into the generative AI model to generate a comment.

[1716] 3. Display the comment: The server displays the generated comment, associating it with the specific user session.

[1717] 4. User interaction: When a user enters a comment using the device, the server inputs the comment as a prompt into the generative AI model to generate a reply.

[1718] 5. Interaction Summary: When a user enters a summary of an interaction, the server stores the summary in a database and makes it available for other users to view.

[1719] Specific examples

[1720] News article: "COVID-19 cases declining as vaccinations progress."

[1721] AI character comment: "It's good news that vaccinations are progressing. We can also expect the number of infections to decrease."

[1722] Prompt Sentence Examples

[1723] "Please generate a comment from an AI character in a format similar to that of a regular user in response to a news article about the decline in the number of COVID-19 infections as vaccinations progress."

[1724] In this way, the system of the present invention allows users to post comments on news articles and interact with AI characters, thereby improving user understanding of the news articles while maintaining privacy and increasing site usage.

[1725] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1726] Step 1: Get the news article

[1727] The server retrieves the latest news articles using the news site's API. Specifically, the server calls the API at "https: / / newsapi.org / " every morning at 9:00 and retrieves the latest news articles. The retrieved news articles are stored in a database.

[1728] Input: API request

[1729] Output: News article data

[1730] Data processing: Store the news article data obtained from the API in a database.

[1731] Step 2: Generate comments for the AI ​​character

[1732] The server inputs the content of the retrieved news article into the generative AI model as a prompt sentence to generate a comment. Specifically, the title and body of the news article are input into the generative AI model as prompt sentences. For example, in response to a news article stating, "As COVID-19 vaccinations progress, the number of infections is decreasing," the server inputs the prompt sentence, "What do you think about this news?" into the generative AI model to generate a comment.

[1733] Input: News article data, prompt

[1734] Output: AI character's comment

[1735] Data processing: News article data is converted into prompt sentences, which are then fed into a generative AI model to generate comments.

[1736] Step 3: View comments

[1737] The server associates the generated AI character's comments with a specific user's session and displays them. Specifically, it uses the user's session information to ensure that the comments are only displayed to that specific user. For example, if user A is logged in, it uses user A's session ID to ensure that the comments are only displayed to user A.

[1738] Input: AI character comments, user session information

[1739] Output: Comments that are displayed to the user

[1740] Data processing: Displays the AI ​​character's comments based on the user's session information.

[1741] Step 4: User interaction

[1742] Users use their devices to interact with the AI ​​character about news articles. Specifically, when a user enters a comment, the server inputs the comment as a prompt into the generative AI model and generates a reply. For example, if a user inputs a comment such as "It's good news that vaccinations are progressing," the server inputs the comment as a prompt into the generative AI model and generates a reply.

[1743] Input: User comments

[1744] Output: AI character's reply

[1745] Data processing: User comments are converted into prompt sentences and fed into a generative AI model to generate replies.

[1746] Step 5: Summarize the conversation

[1747] After the interaction is complete, the user enters a "summary of the interaction" and posts it. Specifically, when a user enters a summary of the interaction, the server saves the summary in a database and makes it available for other users to view. For example, if a user enters a summary such as "Through my interaction with the AI ​​character, I was able to reaffirm the importance of vaccination," the server saves the summary in a database and makes it available for other users to view.

[1748] Input: User Summary

[1749] Output: Summary stored in database

[1750] Data processing: Save the user's summary in a database and make it available for other users to view.

[1751] (Application example 1)

[1752] Next, a description will be given of Application Example 1 of Form 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."

[1753] On traditional news sites, users can refer to other users' comments when reading news articles, but they lack the ability to have individual conversations. They also lack support for users to gain a deeper understanding of news articles and are not designed to increase user time on the site. Furthermore, offensive comments are sometimes posted, and measures to address this are needed.

[1754] 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.

[1755] In this invention, the server includes means for introducing an AI character into a news comment section, means for the AI ​​character to comment on the news in a format similar to that of general users, means for the user to interact with the AI ​​character about the news, means for the interaction to be visible only to the user, means for the user to post a "summary of the interaction" after the interaction is over, means for the AI ​​character to post comments and interact with the user as the user reads the news article, and means for generating a summary of the interaction between the user and the AI ​​character. This allows users to understand the news article while individually interacting with the AI ​​character, deepening their understanding of the news and increasing their stay time. Furthermore, the AI ​​character's role as a receptacle for offensive comments helps maintain a healthy comment section.

[1756] A "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[1757] An "artificial intelligence character" is a virtual character generated using artificial intelligence technology that has the ability to interact with users.

[1758] A "format similar to comments from general users" is a format that has the same style and content as comments posted by regular users.

[1759] "Means by which the user can exchange news with the AI ​​character" is a function that allows the user to have two-way communication with the AI ​​character.

[1760] "Means for making the exchange visible only to the user" refers to a function that prevents the content of the exchange between the user and the AI ​​character from being displayed to other users.

[1761] "A means to post a 'summary of the conversation' after the conversation is over" is a function that allows users to post a concise summary of the content of their conversation with an AI character.

[1762] "A means for an AI character to post comments and interact with users when they read news articles" is a function that enables an AI character to provide comments to users who are viewing news articles and start a dialogue.

[1763] The "means for generating a summary of the interaction between the user and the AI ​​character" is a function for automatically summarizing the content of the conversation between the user and the AI ​​character.

[1764] As an embodiment of the present invention, a system for introducing an AI character into a news site is constructed. A specific embodiment of the system is shown below.

[1765] System Program

[1766] This system consists of a server and a user terminal. The server provides a function that allows an AI character to post comments on news articles and interact with users. The user terminal displays news articles and provides an interface for interacting with the AI ​​character.

[1767] Hardware and software used

[1768] Hardware: User devices such as smartphones, tablets, and PCs

[1769] Software: Flask (Python web framework), OpenAI API (generative AI model)

[1770] Data processing and calculation

[1771] 1. Acquisition of news articles: The server acquires news articles from the database and sends them to the user terminal.

[1772] 2. AI character comment generation: The server receives the news article as input and uses a generative AI model (e.g., text-davinci-003) to generate AI character comments based on the prompt sentence.

[1773] 3. User interaction: The user device displays the generated comments and provides an interface through which the user can interact with the AI ​​character. The user's input is sent to the server, which again uses the generative AI model to generate a response.

[1774] 4. Summary generation of interaction: The server receives the content of the interaction between the user and the AI ​​character as input and generates a summary using the generative AI model. The generated summary is sent to the user's device so that the user can post it.

[1775] Specific examples

[1776] Suppose a user is reading a news article titled "COVID-19 Vaccination Progress." An AI character comments, "As vaccination progresses, it is expected that herd immunity will be achieved." The user responds, "That's true, but I'm worried about side effects." The AI ​​character responds, "It would be best to consult an expert about side effects." After the exchange, the user posts a summary saying, "We discussed the progress of vaccination and concerns about side effects."

[1777] Prompt Sentence Examples

[1778] Generating comments from AI characters on news articles:

[1779] News article: COVID-19 vaccinations progress

[1780] AI character's comment:

[1781] Generate summaries of user interactions:

[1782] User interaction with AI character:

[1783] User: As vaccinations progress, it is hoped that we will achieve herd immunity.

[1784] AI Character: Yes, but I'm worried about the side effects.

[1785] User: It's best to seek expert advice regarding side effects.

[1786] summary:

[1787] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1788] Step 1:

[1789] The server retrieves the news articles from the database and transmits them to the user terminal.

[1790] Input: News article ID

[1791] Data processing: Search and retrieve news article content from the database.

[1792] Output: News article content

[1793] Specific operation: The server searches the database using the news article ID as a key, retrieves the content of the corresponding news article, and then sends the retrieved content of the news article to the user's terminal.

[1794] Step 2:

[1795] The user terminal displays the acquired news article.

[1796] Input: News article content

[1797] Data processing: Converting the content of news articles into a format that can be displayed in the user interface.

[1798] Output: The displayed news article

[1799] Specific operation: The user terminal displays the contents of the news article received from the server on the user interface, allowing the user to view the news article.

[1800] Step 3:

[1801] The server receives a news article as input and uses a generative AI model to generate comments for an artificially intelligent character based on a prompt sentence.

[1802] Input: News article content

[1803] Data processing: The content of the news article is embedded in a prompt sentence and input to the generative AI model, which then generates comments based on the prompt sentence.

[1804] Output: Comments from the AI ​​character

[1805] Specific operation: The server embeds the content of the news article into a prompt sentence and inputs it into the generative AI model (e.g., text-davinci-003). The generative AI model generates a comment for the AI ​​character based on the prompt sentence and sends the comment to the user's device.

[1806] Step 4:

[1807] The user terminal displays the generated comments and provides an interface through which the user can interact with the artificial intelligence character.

[1808] Input: AI character's comment

[1809] Data processing: Converting comments into a format that can be displayed in the user interface.

[1810] Output: Comments displayed

[1811] Specific operation: The user terminal displays the AI ​​character's comments received from the server on the user interface. The user can view the comments and start a dialogue.

[1812] Step 5:

[1813] The user interacts with an artificially intelligent character, and the user's input is sent to the server.

[1814] Input: User input

[1815] Data processing: Converting user input into a format to send to the server.

[1816] Output: User input sent to the server

[1817] Specific operation: The user inputs a response to the AI ​​character's comment. The user terminal sends the user's input to the server.

[1818] Step 6:

[1819] The server receives the user's input and generates a response, again using the generative AI model.

[1820] Input: User input

[1821] Data processing: The user's input is embedded in a prompt sentence and input to the generative AI model, which then generates a response based on the prompt sentence.

[1822] Output: AI character's response

[1823] Specific operation: The server embeds the user's input into a prompt sentence and inputs it into the generative AI model. The generative AI model generates a response for the AI ​​character based on the prompt sentence and sends the response to the user's device.

[1824] Step 7:

[1825] The user terminal displays the generated response and allows the user to continue the dialogue.

[1826] Input: AI character's response

[1827] Data processing: Converting responses into a format that can be displayed in the user interface.

[1828] Output: The displayed response

[1829] Specific operation: The user terminal displays the AI ​​character's response received from the server on the user interface, allowing the user to continue the dialogue.

[1830] Step 8:

[1831] The server receives the dialogue between the user and the AI ​​character as input and generates a summary using a generative AI model.

[1832] Input: Dialogue between the user and the AI ​​character

[1833] Data processing: The dialogue is embedded in a prompt and fed into a generative AI model, which then generates a summary based on the prompt.

[1834] Output: A summary of the conversation

[1835] Specific operation: The server embeds the dialogue between the user and the AI ​​character into a prompt sentence and inputs it into the generative AI model. The generative AI model generates a summary of the dialogue based on the prompt sentence and sends the summary to the user's device.

[1836] Step 9:

[1837] The user terminal displays the generated summary and allows the user to contribute.

[1838] Input: A summary of the conversation

[1839] Data processing: Converting the summary into a format that can be displayed in the user interface.

[1840] Output: Displayed summary

[1841] Specific operation: The user terminal displays the summary of the exchange received from the server on the user interface. The user can check the summary and post.

[1842] Example 2

[1843] Next, a description will be given of Example 2 of Form 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."

[1844] Conventional news comment systems lacked the support required for users to gain a deeper understanding of news articles, and were not designed to increase user retention. Furthermore, there was a need for an appropriate response to offensive posts.

[1845] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1846] In this invention, the server includes a means for acquiring news articles, a means for analyzing the acquired news articles using natural language processing technology, a means for generating comments and questions using a generative AI model based on the analysis results, and a means for analyzing user responses and generating appropriate responses using the generative AI model. This allows users to gain a deeper understanding of news articles and increase their stay time. It also allows for appropriate responses to offensive posts.

[1847] The "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[1848] An "artificial intelligence character" is a program that automatically generates comments and questions about news articles and interacts with users.

[1849] "A format similar to comments from general users" refers to comments and questions generated by AI characters that are in the same format and style as comments and questions posted by regular users.

[1850] "Means for making interactions visible only to the user" refers to a function that prevents the conversation between the user and the AI ​​character from being displayed to other users.

[1851] The "summary of the interaction" is information that briefly summarizes the content of the conversation between the user and the AI ​​character.

[1852] "Means for obtaining news articles" refers to a function for collecting the latest news articles from news sites and RSS feeds on the Internet.

[1853] "Natural language processing technology" is a technology that enables computers to understand and analyze human language.

[1854] A "generative AI model" is an algorithm that allows artificial intelligence to generate new text based on a given prompt.

[1855] The "means for analyzing user responses" is a function for analyzing the text entered by the user and understanding its content and emotions.

[1856] The "means for generating an appropriate response" is a function that enables an AI character to generate appropriate comments or questions based on the user's response.

[1857] This invention is a system that introduces an AI character into a news comment section and provides dialogue that allows users to gain a deeper understanding of a news article. Specific embodiments of this system are described below.

[1858] The server first retrieves the latest news articles from news sites or RSS feeds on the Internet. For example, the server retrieves the RSS feed from "https: / / example-news.com / rss" and extracts the latest articles. This process uses common web scraping techniques or RSS reader software.

[1859] The server then analyzes the content of the retrieved news articles using natural language processing technology. Specifically, it uses the Google Cloud Natural Language API to analyze the article's topics, keywords, and sentiment. For example, the server analyzes the content of the article and extracts keywords such as "small and medium-sized enterprises," "bill," and "tax benefits."

[1860] Based on the analysis results, the server uses a generative AI model (e.g., OpenAI's GPT-4) to generate comments and questions. For example, the server generates the question, "How do you think this bill will affect small businesses?" An example of this prompt would be:

[1861] Analyze the content of a news article and generate questions such as:

[1862] "How do you think this bill will affect small businesses?"

[1863] The generated comments and questions are displayed to the user via the terminal. The terminal displays the question "How do you think this bill will affect small businesses?" below the news article. The user inputs their opinion in response to this question. For example, the user might input "I think this bill is beneficial to small businesses because it includes tax breaks."

[1864] The server receives the user's response and analyzes it again using natural language processing technology. This allows it to understand the user's opinions and emotions. Based on the analysis results, the server uses a generative AI model to generate an appropriate response. For example, the server might generate a response such as, "I see, tax incentives are important for small and medium-sized businesses. What else are you concerned about?"

[1865] This series of processes allows users to gain a deeper understanding of news articles and increase their viewing time. The AI ​​character can also respond appropriately to offensive posts. This system is an effective way for users to gain a deeper understanding of news articles and promote constructive dialogue.

[1866] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1867] Step 1: Get the news article

[1868] The server retrieves the latest news articles from news sites and RSS feeds on the Internet. Specifically, the server retrieves the RSS feed from "https: / / example-news.com / rss" and extracts the latest articles. The input is the RSS feed URL, and the output is the news article data. This process uses web scraping technology or RSS reader software.

[1869] Step 2: Analyzing the article content

[1870] The server analyzes the content of the retrieved news articles using natural language processing technology. Specifically, it uses the Google Cloud Natural Language API to analyze the article's topics, keywords, and sentiment. The input is the news article data, and the output is the analysis results (topic, keyword, and sentiment analysis results). For example, the server analyzes the content of the article and extracts keywords such as "small and medium-sized enterprises," "bill," and "tax incentives."

[1871] Step 3: Generate comments and questions

[1872] The server generates comments and questions using a generative AI model (e.g., OpenAI's GPT-4) based on the analysis results. The input is the analysis results, and the output is the generated comments and questions. For example, the server generates the question, "How do you think this bill will affect small businesses?" An example of this prompt is as follows:

[1873] Analyze the content of a news article and generate questions such as:

[1874] "How do you think this bill will affect small businesses?"

[1875] Step 4: Display to the user

[1876] The terminal displays comments and questions received from the server to the user. The input is the generated comment or question, and the output is the text displayed on the user's screen. For example, the terminal might display the question "How do you think this bill will affect small businesses?" below a news article.

[1877] Step 5: Getting the user's response

[1878] The user inputs their opinion in response to the displayed question. The input is the user's opinion, and the output is the user's response data. For example, the user might input, "I think this bill is beneficial to small businesses because it includes tax breaks."

[1879] Step 6: Parsing the user's response

[1880] The server receives the user's response and analyzes it again using natural language processing technology. The input is the user's response data, and the output is the analysis results (the analysis results of the user's opinions and emotions). This allows the user's opinions and emotions to be understood.

[1881] Step 7: Generating AI character reactions

[1882] The server uses a generative AI model to generate an appropriate response based on the results of the user's response analysis. The input is the user's response analysis result, and the output is the generated response. For example, the server generates a response such as, "I see, tax incentives are important for small and medium-sized businesses. What else are you concerned about?"

[1883] Step 8: View the reaction

[1884] The terminal displays the responses received from the server to the user. The input is the generated response, and the output is the text displayed on the user's screen, allowing the user to continue interacting with the AI ​​character.

[1885] This process allows users to gain a deeper understanding of news articles and increase their time spent on the site. It also enables the AI ​​character to respond appropriately to offensive posts.

[1886] (Application example 2)

[1887] Next, a description will be given of Application Example 2 of Form 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."

[1888] Conventional news comment systems lack support for users to gain a deeper understanding of news articles. It is also difficult for users to receive appropriate feedback when posting comments on news articles. Furthermore, they lack the means to increase user retention and lack sufficient countermeasures against offensive posts.

[1889] The specific processing by the specific 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 introducing an AI character into a news comment section, means for the AI ​​character to comment on the news in a format similar to that of general users, means for the user to interact with the AI ​​character about the news, means for the interaction to be visible only to the user, means for posting a "summary of the interaction" after the interaction is over, means for generating comments on the news article using a generative AI model, and means for providing an appropriate response to the user's opinion. This allows users to gain a deeper understanding of the news article and receive appropriate feedback. It also increases the user's stay time and strengthens measures against offensive posts.

[1890] A "news comment section" is an interface that allows users to post their opinions and thoughts about news articles.

[1891] An "AI character" is a virtual character generated using artificial intelligence technology that can interact with users.

[1892] "A format similar to comments made by general users" refers to comments that are in the same format and style as comments posted by regular users.

[1893] "Means for commenting on news" refers to methods and functions for posting opinions and thoughts about news articles.

[1894] "Means by which users can interact with the AI ​​character about the news" refers to methods or functions that allow users to exchange opinions about news articles through dialogue with the AI ​​character.

[1895] "Means for ensuring that the interaction is visible only to the user" refers to methods or functions that prevent the content of the conversation between the user and the AI ​​character from being displayed to other users.

[1896] "Means for posting summaries of interactions" refers to methods or functions for posting summaries of the content of the conversation between the user and the AI ​​character.

[1897] A "generative AI model" is a model that uses artificial intelligence technology to generate text and data.

[1898] The "means for generating comments on news articles" refers to a method or function for generating appropriate comments based on the content of a news article.

[1899] "Means of responding appropriately to user opinions" refers to m...

Claims

1. A means for obtaining the latest news articles from a news site; means for generating a first prompt sentence for instructing the user to generate a comment in a format similar to comments posted by general users with respect to the acquired news article; A means for generating a comment to be posted on the acquired news article using the generated first prompt sentence and a generative AI model; A means to post the generated comments in the news comment section in a way that makes them visible only to certain users; means for generating a second prompt sentence instructing a user to generate a reply comment to a comment posted by a specific user; A means for generating a reply comment to a comment posted by a specific user using the generated second prompt sentence and the generative AI model; A method to post the generated reply comment in the news comment section so that it is visible only to specific users, A means for generating a summary of the contents of comments exchanged between the specific user and the specific user after the exchange with the specific user is completed; A means to post the generated summary of the interaction in the comments section of the news, where it can be viewed by other users; and A system including:

2. A recognition means for recognizing an emotional state of the specific user using an emotion engine based on a comment posted by the specific user; A means for changing the content of a reply comment posted in a comment section of the news article in accordance with the recognized emotional state of the specific user; The system of claim 1 further comprising:

3. The method further includes a means for storing the generated comments, reply comments, and comments posted by the specific user in a database. The system of claim 1 .

Images