System

Abstract

A system is provided.SOLUTION: A system comprising: means for a user to input information; means for a server to receive the information; means for the server to analyze the information using a generative model; means for the server to mark parts that contain exaggeration or bias; means for the server to evaluate differences from the original information; means for the server to generate a transparency report to the user; and means for the server to provide advice for verification of the true information.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] The spread of the internet and social media has led to an increase in reporting that prioritizes viewer ratings and clicks, making exaggerated information commonplace. As a result, it is difficult for the public, businesses, and policymakers to verify the truth, and overreactions to exaggerated information lead to conservative decision-making and high risk management costs. There is a need to solve this problem and provide a more transparent information environment. [Means for solving the problem]

[0005] The present invention provides a system including a means for a user to input information, a means for a server to receive the information, a means for the server to analyze the information using a generative model, a means for the server to mark parts that include exaggeration or bias, a means for the server to evaluate differences from the original information, a means for the server to generate a transparency report for the user, and a means for the server to provide advice for verifying the truth of the information. With this configuration, the user can obtain highly transparent information that is free from exaggeration and bias.

[0006] A "user" is an individual or entity that operates the system.

[0007] "Information" refers to data and media content that users input into the system.

[0008] A "server" is a central system or computer for receiving and processing information.

[0009] A "generative model" refers to a function that analyzes information using artificial intelligence or machine learning algorithms that run on the server.

[0010] "Analyzing" is the process of analyzing the content of information and identifying parts that contain exaggeration or bias.

[0011] "Exaggeration or bias" refers to a situation in which some information is given disproportionate emphasis or a particular point of view is favored.

[0012] "Marking" refers to indicating to the user in a manner that is easy to understand any parts that have been identified as containing exaggeration or bias.

[0013] "Difference" refers to the difference between the original information and the analyzed information.

[0014] A "Transparency Report" is a report that shows users the parts that contain exaggeration or bias and the evaluation results.

[0015] "Advice" refers to advice or recommendations for users to verify truthful information. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0017] 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.

[0018] First, the terms used in the following description will be explained.

[0019] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0020] 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.

[0021] 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.

[0022] 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.

[0023] 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."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 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.

[0027] 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).

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0037] The present invention is a system in which users input information, a server analyzes the information, identifies exaggerations and biases, and provides users with transparent reports and advice. The following describes the program processing of this system in detail.

[0038] Get news articles

[0039] The user opens the news app, enters the URL or text of the news article they want to read, and when the user selects an article within the news app, the device sends it to the server.

[0040] News article analysis

[0041] The server passes the received news article to an internal generative model, which uses artificial intelligence to analyze the article's content by evaluating each sentence in context and identifying any parts that may contain exaggeration or bias.

[0042] Examples:

[0043] News article text:

[0044] "The company's stock price has plummeted. Experts say this is the worst event in the last 20 years."

[0045] The AI ​​model marks the "worst case scenario" as having a high exaggeration score.

[0046] Extracting exaggerated and biased sentences

[0047] The server extracts exaggerated or highly biased parts identified by the generative model and marks them, explicitly indicating them to the user with a specific tag (e.g., <exaggerated>).

[0048] Evaluating differences and providing transparency

[0049] The server evaluates the differences between the original news article and the marked text and generates a transparency report that clearly shows which parts contain exaggerations or bias, in a format that is easy for users to understand.

[0050] Examples:

[0051] Original news article: The company's stock price has plummeted, in what experts say is the worst it's seen in 20 years.

[0052] Transparency Report: "The worst thing that could happen" is an exaggeration. Please check with another source.

[0053] Providing advice

[0054] The server generates specific advice to help users verify the truth of the information, for example, recommending reference to other reliable sources or encouraging users to check official statistics.

[0055] Examples:

[0056] This news story contains exaggerated statements. Please verify with the following independent sources:

[0057] 1. Official statistical data

[0058] 2. Other trusted news sites

[0059] 3. In-depth analysis by experts

[0060] Integration into news apps

[0061] The server will integrate the "Insight Detector" function into the news app interface, allowing users to click a specific button while browsing the news to receive analysis results and advice in real time.

[0062] In this way, the present invention provides users with highly transparent information and supports honest and unbiased decision making.

[0063] The processing flow will be explained below.

[0064] Step 1:

[0065] A user opens a news app and enters the URL or text of a news article, or selects an article within the news app.

[0066] Step 2:

[0067] The device sends the URL or text entered by the user to the server.

[0068] Step 3:

[0069] The server passes the received news articles to an internal generative model, which uses artificial intelligence algorithms to analyze the article's content.

[0070] Step 4:

[0071] A generative model in the server evaluates each sentence in the article based on its context, identifying any parts that may contain exaggeration or bias.

[0072] Step 5:

[0073] The server extracts the exaggerated or biased parts identified by the generative model and marks them, for example, with specific tags (<exaggerated>, <biased>).

[0074] Step 6:

[0075] The server evaluates the differences between the original news article and the marked text and generates the results as a transparency report.

[0076] Step 7:

[0077] The server generates a transparency report in text format and displays it to the user.

[0078] Step 8:

[0079] The server generates and provides textual advice to the user on how to verify the truth of the information, such as by recommending a list of other reliable sources or official statistical data.

[0080] Step 9:

[0081] The server integrates the "Insight Detector" function into the news app interface, allowing users to view analysis results and advice in real time as they browse the news.

[0082] Example 1

[0083] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0084] In today's information society, amidst the vast amount of news articles and information being distributed, much of the information contains exaggeration and bias. This makes it difficult for users to obtain accurate and objective information, which can lead to incorrect decision-making. There is a need to solve this problem and provide users with highly transparent and reliable information.

[0085] 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.

[0086] In this invention, the server includes: a means for a user to input information; a means for a terminal to send the input information to the server; a means for the server to receive information; a means for the server to analyze the information using a generative AI model; a means for the server to evaluate the content of an article and identify exaggerations or bias; a means for the server to mark portions containing exaggerations or bias; a means for the server to evaluate differences from the original information and generate a transparency report; a means for the server to provide advice for verifying the truth of the information; and a means for the server to display the analysis results and advice on a user interface. This makes it easier for users to obtain reliable information, enabling them to make accurate and unbiased decisions.

[0087] "User" means a human or end-user who uses the system to input information and receive analytical results and transparency reports.

[0088] A "terminal" is an electronic device that a user uses to input information and that has the function of transmitting information to a server.

[0089] A "server" is a computer system that processes information received from a terminal and analyzes it using a generative AI model, and is a device that generates and provides various reports.

[0090] A "generative AI model" is a model that uses artificial intelligence technology to analyze information such as news articles, and is an algorithm used in particular to identify exaggeration or bias in information.

[0091] "Information" is a general term for news articles, text data, etc. entered by users, and is the data to be analyzed.

[0092] "Exaggeration" refers to information that overly exaggerates reality and lacks accuracy.

[0093] "Bias" refers to information that is biased toward a particular position or perspective, and is an unfair representation.

[0094] A "Transparency Report" is a document generated by the server that clearly shows the difference between the original information and the analysis results, and aims to improve the reliability of the information for users.

[0095] "Advice" is a recommendation or instruction provided by the server to help users verify truthful information, pointing to other reliable sources or verification methods.

[0096] "User interface" refers to the screen or operating means that allows the user to visually check the analysis results and advice from the server.

[0097] The present invention is a system in which a user inputs news information, a server analyzes the information, identifies exaggerations and bias, and provides the user with a highly transparent report and specific advice. Specific embodiments of this system are described below.

[0098] This system begins when a user inputs information through a news app and sends it to a server. When a user inputs the URL or text of a news article, the device sends this information to the server. The server passes the received news article to an internal generative AI model, which then analyzes the content of the article. For example, a model using natural language processing (NLP) is used as the generative AI model.

[0099] When analyzing a news article, the server evaluates each sentence in the article based on its context to identify exaggerations or bias. For example, consider the following news article:

[0100] "The company's stock price has plummeted. Experts say this is the worst thing that has happened in the last 20 years."

[0101] The generated AI model marks the expression "the worst thing" with a high exaggeration score. After the exaggerated and biased parts are identified, the server marks these parts using specific tags (e.g., <exaggeration>). At this time, a specific example is as follows:

[0102] "The company's stock price has plummeted. Experts say this is the <exaggeration>worst thing in the past 20 years< / exaggeration>."

[0103] Subsequently, the server evaluates the difference between the original news article and the marked text, and generates the result as a transparency report. The transparency report clearly shows which parts contain exaggeration or bias, and is a document to improve the reliability of information for users. A specific example is as follows:

[0104] Original news article: "The company's stock price has plummeted. Experts say this is the worst thing in the past 20 years."

[0105] Transparency report: "The expression 'the worst thing' is exaggerated. Please check from other information sources."

[0106] Furthermore, the server generates specific advice for users to verify the true information. For example, it recommends referring to other reliable information sources or encourages checking public statistical data. Examples of advice provided to users include the following content:

[0107] "This news contains exaggerated expressions. Please check with the following independent information sources:

[0108] 1. Public statistical data

[0109] 2. Other reliable news sites

[0110] 3. Detailed analysis articles by experts"

[0111] Finally, the server will integrate the "Insight Detector" feature into the news app interface, allowing users to click a specific button while browsing the news to receive real-time analysis and advice.

[0112] In this way, the present invention provides users with highly transparent information and supports honest and unbiased decision making.

[0113] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0114] Step 1:

[0115] The user opens a news app, which is ready to input information. The user enters the URL or text of the news article they want to read. The news app passes the information to the device.

[0116] Input: News article URL or text

[0117] Output: News article information sent to device

[0118] Step 2:

[0119] The terminal sends the entered news article information to the server. Specifically, the terminal sends the information to the server via an HTTP request.

[0120] Input: News article URL or text (on device)

[0121] Output: News article information sent to the server

[0122] Step 3:

[0123] A server receives news article information, which it prepares for input into an analytical model.

[0124] Input: News article information sent from the device

[0125] Output: News article information formatted for analysis

[0126] Step 4:

[0127] The server inputs the news article into the generative AI model, which then makes an API request to the model to begin analysis.

[0128] Input: News article information formatted for analysis

[0129] Output: News article information fed into the generative AI model

[0130] Step 5:

[0131] The generative AI model evaluates each sentence in a news article based on its context and identifies parts that may contain exaggeration or bias. Specifically, the model analyzes the text using natural language processing techniques.

[0132] Input: News article information

[0133] Output: Identification of exaggeration or bias

[0134] Step 6:

[0135] The server extracts exaggerated or highly biased parts from the output of the generative AI model, and marks the identified parts with a corresponding tag (e.g., <exaggerated>).

[0136] Input: Results identifying exaggeration or bias

[0137] Output: Tagged news article information

[0138] Step 7:

[0139] The server evaluates the differences between the original news article and the marked sentences, making it clear which parts contain exaggeration or bias.

[0140] Input: tagged news article information

[0141] Output: The result of the difference evaluation.

[0142] Step 8:

[0143] The server generates a transparency report based on the results of the difference evaluation, which is output in a format that is easy for users to understand.

[0144] Input: Result of the difference evaluation

[0145] Output: Transparency Report

[0146] Step 9:

[0147] The server generates specific advice to provide to the user, including a recommendation to refer to a trusted source.

[0148] Input: Difference Assessment Results and Transparency Report

[0149] Output: Specific advice

[0150] Step 10:

[0151] The server displays the analysis results and advice on the news app interface. Specifically, when the user clicks a specific button, the analysis results and advice are displayed in real time.

[0152] Input: Transparency reports and specific advice

[0153] Output: Analysis results and advice displayed in the user interface

[0154] Through this series of processing steps, users can obtain highly transparent information, which can support honest and unbiased decision-making.

[0155] (Application example 1)

[0156] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0157] In today's world, there is an increasing risk that users will be misled by exaggerations and bias in advertisements and news articles. As a result, users may make decisions based on inaccurate information. In particular, in an information-overloaded environment, unreliable information is easily mixed in, making it difficult for users to discern accurate information. The objective of this invention is to automatically identify exaggerations and biases in advertisement content and news articles, and provide users with transparent reports and specific advice to support honest and unbiased decision-making.

[0158] 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.

[0159] In this invention, the server includes: a means for a user to input information; a means for receiving information; a means for analyzing the information using a generative model; a means for marking parts containing exaggeration or bias; a means for evaluating differences from the original information; a means for generating a transparency report for the user; a means for providing advice for verifying the truth of the information; and a means for acquiring and analyzing advertising content, identifying exaggeration or bias, and providing a transparency report and specific advice. This allows the reliability of advertisements and news articles to be automatically evaluated, enabling users to make honest and unbiased decisions.

[0160] A "user" is someone who uses the system to input or receive information.

[0161] "Information" refers to news articles, advertising content, and other text data and content in general that is the subject of analysis.

[0162] A "server" is a computer system that receives, analyzes, and evaluates information sent by users and returns the results.

[0163] A "generative model" refers to an algorithm or data model that uses artificial intelligence to analyze data.

[0164] "Exaggeration" refers to parts that are exaggerated in a way that is greater than the actual content.

[0165] "Bias" refers to a part that contains a particular perspective or preconception.

[0166] "Marking" refers to the act of identifying exaggerations or biases and explicitly labeling them.

[0167] "Differential evaluation" is the process of evaluating the differences between the original information and the analyzed information.

[0168] A "Transparency Report" is a report that identifies areas that contain exaggeration or bias and presents them in a user-friendly format.

[0169] "Advice" means specific instructions or suggestions provided to a user to help them make a good faith decision.

[0170] "Advertising Content" refers to promotional text and multimedia content relating to products and services.

[0171] "Analysis" is the process of evaluating and examining user-provided information for a specific purpose.

[0172] The system embodying this invention analyzes input information from users, identifies exaggeration and bias, generates transparency reports, and provides specific advice. Specific steps for implementing this invention are described below.

[0173] Configuration and operation explanation

[0174] Hardware configuration:

[0175] User devices: Mobile devices such as smartphones and tablets

[0176] Server: A high-performance computer or cloud server

[0177] Network: Internet

[0178] Software configuration:

[0179] Generative AI models: Transformer-based AI model libraries (e.g., Hugging Face Transformers)

[0180] Analysis program: Python script

[0181] News app: an interface where users enter information

[0182] A user accesses a news app or ad-checking app and enters the information about the ad or news article they want to analyze. The entered information is sent from the user's device to a server. The server then supplies the received information to a generative AI model, which analyzes it for exaggeration and bias.

[0183] The server uses a generative AI model to analyze each piece of an article or ad, taking into account the context and assessing whether a particular phrase or expression is exaggerated or biased.

[0184] If exaggeration or bias is identified, the server will flag these and evaluate the deviation from the original information. For example, if a news article mentions "the best product on the market," it will be deemed exaggerated.

[0185] Additionally, the server generates a transparency report that explicitly shows users where there is exaggeration or bias, detailing the alleged exaggeration or bias and explaining why it is problematic.

[0186] Finally, the server provides specific advice to the user, such as recommending verification from a trusted source.

[0187] Specific examples

[0188] Example: A user enters the following ad text in a news app: "The miracle diet! Lose 10kg in just 10 days."

[0189] The server analyzes this text using a generative AI model and identifies phrases like "miracle" and "lose 10kg in just 10 days" as exaggerations. It then generates a transparency report explaining to the user how these phrases are exaggerated. It also provides the user with specific advice: "This information is exaggerated. Please check other reliable sources."

[0190] Example prompt sentence:

[0191] "Below is the content of the advertisement. Please analyze this content for exaggeration or bias and provide a transparent report.

[0192] Advertisement: Miracle diet method! Lose 10kg in just 10 days.

[0193] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0194] Program processing flow and detailed explanation of each step

[0195] Step 1:

[0196] The user enters the information.

[0197] The user opens a news app or ad check app and enters the text of the ad or news article to be analyzed, or specifies a URL. This is the input information. The input information is sent from the user's device to the server.

[0198] Step 2:

[0199] The server receives the information.

[0200] The server receives the information sent by the user. The received information is then directly supplied to the generative AI model for analysis. Through this receiving operation, the server obtains input data (news articles and advertising text).

[0201] Step 3:

[0202] The server analyzes the information using a generative AI model.

[0203] The server inputs the received information into a generative AI model (e.g., Hugging Face Transformers) for analysis. Specifically, the generative AI model evaluates each piece of text and determines whether it contains exaggeration or bias. The input data is processed and an analysis result is generated.

[0204] Step 4:

[0205] The server marks the parts that contain exaggeration or bias.

[0206] The server marks parts of the original text that are identified as exaggerated or biased based on the analysis results obtained from the generative AI model. The marking uses tags (e.g., <exaggeration>) to clearly indicate which parts of the original text are exaggerated. The marked text is then output.

[0207] Step 5:

[0208] The server evaluates the difference from the original information.

[0209] The server evaluates the differences between the original information and the marked information, and sorts out which parts have been exaggerated and how. At this stage, the difference evaluation results are generated.

[0210] Step 6:

[0211] The server generates a transparency report.

[0212] The server generates a transparency report based on the difference evaluation results, detailing any exaggerations or biases and explaining why. The report is then output as feedback to the user.

[0213] Step 7:

[0214] The server provides advice to verify the truth of the information.

[0215] Based on the transparency report, the server generates specific advice for users to verify accurate information, including recommendations for reliable sources, and provides this advice to users along with the report.

[0216] For example, if a user enters the ad copy "Miracle Diet! Lose 10kg in just 10 days," the server receives it and analyzes it using a generative AI model. It identifies and marks the exaggerated parts "miracle" and "Lose 10kg in just 10 days." It then evaluates the difference between the original text and the marked text, and generates a transparency report with specific advice, such as "This text contains exaggerated parts. Please check other reliable sources," and provides it to the user.

[0217] 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.

[0218] The present invention is a system that identifies exaggeration and bias in the information entered by a user and provides a transparency report and advice that takes into account the user's feelings. The program processing of the system will be described in detail below.

[0219] Get news articles

[0220] When a user opens a news app and enters the URL or text of a news article, or selects an article within the news app, the device sends this to the server.

[0221] News article analysis

[0222] The server receives news articles and passes them to an internal generative model, which uses artificial intelligence algorithms to analyze the content of the article, evaluate the context of the sentences, and identify any parts that may contain exaggeration or bias.

[0223] Examples:

[0224] News article text:

[0225] "The company's stock price has plummeted. Experts say this is the worst event in the last 20 years."

[0226] The AI ​​model marks the "worst case scenario" as having a high exaggeration score.

[0227] Extracting exaggerated and biased sentences

[0228] The server extracts the exaggerated or biased parts identified by the generative model and marks them, explicitly indicating them to the user with specific tags (<exaggerated>, <biased>).

[0229] Emotion Analysis

[0230] The server uses an emotion engine to analyze emotions based on the text entered by the user and the selections made by the user, and this emotion analysis identifies the emotional state (e.g., anger, anxiety, joy, etc.) that the user is in when reading a news article.

[0231] Examples:

[0232] When a user reads a news article, the emotion engine detects "anxiety" from the expression in the text input.

[0233] Evaluating differences and providing transparency

[0234] The server evaluates the differences between the original news article and the marked text and generates a transparency report that clearly shows which parts contain exaggeration or bias, in a user-friendly format.

[0235] Examples:

[0236] Original news article: The company's stock price has plummeted, in what experts say is the worst it's seen in 20 years.

[0237] Transparency Report: "The worst thing that could happen" is an exaggeration. Please check with another source.

[0238] Providing advice

[0239] The server generates specific advice for the user to confirm the truth of the information. This advice reflects the analysis results of the emotion engine and provides appropriate instructions according to the user's emotional state.

[0240] Examples:

[0241] If users are feeling "uneasy," they should be advised that "This news contains exaggerated statements. Please check the following reliable sources and do not worry too much:

[0242] 1. Official statistical data

[0243] 2. Other trusted news sites

[0244] 3. In-depth analysis by experts

[0245] Integration into news apps

[0246] Sarva will integrate the "Insight Detector" function into the news app interface. This will allow users to click a specific button when viewing news to receive analysis results and advice in real time. By providing highly transparent information that takes user emotions into consideration, Sarva will support honest and unbiased decision-making.

[0247] Thus, the present invention is a system that takes into account the user's emotional state while providing highly transparent information without exaggeration or bias.

[0248] The processing flow will be explained below.

[0249] Step 1:

[0250] A user opens a news app and enters the URL or text of a news article, or selects an article within the news app.

[0251] Step 2:

[0252] The device sends the URL or text entered by the user to the server.

[0253] Step 3:

[0254] The server passes the received news articles to an internal generative model, which uses artificial intelligence algorithms to analyze the article's content.

[0255] Step 4:

[0256] A generative model in the server evaluates each sentence in the article based on its context, identifying any parts that may contain exaggeration or bias.

[0257] Step 5:

[0258] The server extracts the exaggerated or biased parts identified by the generative model and marks them, for example, with specific tags (<exaggerated>, <biased>).

[0259] Step 6:

[0260] The server activates an emotion engine to recognize the user's emotions while the user is reading the article.

[0261] Step 7:

[0262] The emotion engine in the server analyzes the user's current emotional state (anger, anxiety, joy, etc.) based on the user's text input and selection operations.

[0263] Examples:

[0264] When a user reads a news article, the emotion engine recognizes that they are feeling "anxiety."

[0265] Step 8:

[0266] The server evaluates the differences between the original news article and the marked text and generates a transparency report that clearly shows which parts contain exaggerations or bias.

[0267] Examples:

[0268] Original news article: The company's stock price has plummeted, in what experts say is the worst it's seen in 20 years.

[0269] Transparency Report: "The worst thing that could happen" is an exaggeration. Please check with another source.

[0270] Step 9:

[0271] The server generates appropriate advice for the user based on the results of the analysis by the emotion engine. For example, if the user is feeling anxious, the server will provide advice to calm the user.

[0272] Examples:

[0273] "This news contains exaggerated statements. Please check the following reliable sources and don't worry too much:

[0274] 1. Official statistical data

[0275] 2. Other trusted news sites

[0276] 3. Detailed analysis by experts

[0277] Step 10:

[0278] The server integrates the "Insight Detector" function into the news app interface, allowing users to view analysis results and advice in real time as they browse the news.

[0279] In this way, the server takes into account the user's emotional state and provides transparent information without exaggeration or bias, thereby supporting honest and unbiased decision-making.

[0280] Example 2

[0281] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0282] In recent years, the amount of information on the Internet has increased dramatically, and the news and articles users encounter often contain exaggerations and bias. As a result, users are at greater risk of receiving incorrect information or holding biased viewpoints. Furthermore, when reading news and articles, users' emotions can significantly influence how they perceive the information, making it difficult for users to objectively evaluate it. In such a situation, there is a need for accurate and transparent information provision.

[0283] 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.

[0284] In this invention, the server includes a means for a user to input information, a means for a terminal to send information to the server, a means for the server to receive information, a means for the server to analyze the information using a generative model, a means for the server to mark parts containing exaggeration or bias, a means for the server to analyze the user's emotions, a means for the server to evaluate differences from the original information, a means for the server to generate a transparency report for the user, and a means for the server to provide advice for verifying the truth of information. This allows users to recognize exaggeration and bias when reading news or articles, understand their own emotional state, and obtain more accurate and transparent information.

[0285] "User" refers to a person who uses this system to input information and view news and articles.

[0286] "Terminal" refers to a device used by a user that provides a means for transmitting information to a server.

[0287] "Server" refers to the main computing system that receives and analyzes information sent from users and terminals.

[0288] A "generative model" refers to an algorithm or system that uses artificial intelligence to analyze and evaluate documents.

[0289] "Exaggeration or bias" refers to exaggerated or biased viewpoints in texts or news articles.

[0290] "Marking" refers to the act of explicitly indicating exaggerated or biased parts using specific tags or highlighting.

[0291] "Emotion analysis" refers to the process of identifying a user's emotional state (anger, anxiety, joy, etc.) based on the user's input text and actions.

[0292] "Differential assessment" refers to assessing the difference between the original information and the information that has been marked with exaggeration or bias.

[0293] A "transparency report" refers to a report that clearly shows users which parts contain exaggeration or bias.

[0294] "Advice" refers to the act of providing specific instructions or recommendations for users to verify truthful information.

[0295] The present invention provides a system that identifies exaggerations and biases in information entered by a user and provides transparency reports and advice that take the user's emotions into account. This system is configured by combining a user, a terminal, a server, and a generative model.

[0296] Get news articles

[0297] The user opens the news app and enters the URL or text of a news article, or can select an article within the news app, and the device then sends the information to the server.

[0298] News article analysis

[0299] The server passes the received news article to an internal generative model (e.g., GPT-4), which uses natural language processing techniques to analyze the article's content, specifically evaluating the context of each sentence and identifying parts that may contain exaggeration or bias.

[0300] Examples:

[0301] News article text:

[0302] "The company's stock price has plummeted. Experts say this is the worst event in the last 20 years."

[0303] The AI ​​model marks the "worst case scenario" as having a high exaggeration score.

[0304] Extracting exaggerated and biased sentences

[0305] The server extracts the exaggerated or biased parts identified by the generative model and marks them explicitly, using tags such as "<exaggerated>" and "<biased>".

[0306] Emotion Analysis

[0307] The server uses an emotion engine to analyze emotions based on the text entered by the user and the selections made, thereby identifying the emotional state (e.g., anger, anxiety, joy) the user is in when reading a news article.

[0308] Examples:

[0309] When a user reads a news article, the emotion engine detects "anxiety" from the expression in the text input.

[0310] Evaluating differences and providing transparency

[0311] The server evaluates the differences between the original news article and the marked text and generates a transparency report that clearly shows users which parts contain exaggerations or bias, in an easy-to-understand format.

[0312] Examples:

[0313] Original news article: "The company's stock price has plummeted, with experts saying it's the worst in 20 years."

[0314] Transparency Report: "The worst thing that could happen" is an exaggeration. Please check with another source.

[0315] Providing advice

[0316] The server generates specific advice for the user to confirm the truth of the information. This advice reflects the analysis results of the emotion engine and provides appropriate instructions according to the user's emotional state.

[0317] Examples:

[0318] If users are feeling "uneasy," they should be advised that "This news contains exaggerated statements. Please check the following reliable sources and do not worry too much:

[0319] 1. Official statistical data

[0320] 2. Other trusted news sites

[0321] 3. In-depth analysis by experts

[0322] Integration into news apps

[0323] Sarva will integrate the "Insight Detector" feature into the news app interface. This will allow users to click a specific button while browsing the news to receive analysis results and advice in real time. This feature will enable users to make honest and unbiased decisions while receiving transparent information that takes emotions into account.

[0324] This system takes into account the user's emotions and provides highly transparent information without exaggeration or bias.

[0325] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0326] Step 1:

[0327] The user launches a news app and enters the URL or text of a news article, or selects an article within the news app. The device then sends this to the server.

[0328] Input: News article URL or text entered by the user.

[0329] Output: The URL or text of the news article entered is sent to the server.

[0330] Specific behavior:

[0331] When a user pastes a URL into the search bar of a news app or taps on a specific article from the "Recent News" section, the device instantly sends it to the server.

[0332] Step 2:

[0333] The server passes the received news article to a generative model, which uses natural language processing techniques to analyze the article's content.

[0334] Input: News article text data.

[0335] Output: Contextual data analyzed using natural language processing techniques.

[0336] Specific behavior:

[0337] The server feeds the text of a news article into a generative model, which evaluates the context of each sentence to identify potential exaggerations or biases.

[0338] Step 3:

[0339] The server extracts the exaggerated or biased parts identified by the generative model and marks them to indicate this explicitly.

[0340] Input: Analyzed contextual data, identifying exaggerations and biases.

[0341] Output: Marked text data.

[0342] Specific behavior:

[0343] The server tags the identified phrases with tags such as "<exaggeration>" or "<bias>." For example, the phrase "the worst thing that could happen" in a sentence might be tagged with "<exaggeration>."

[0344] Step 4:

[0345] The server uses an emotion engine to analyze emotions based on the user's text input and selections.

[0346] Input: User input text and operation data.

[0347] Output: Sentiment analysis results (e.g., anxiety, anger, joy, etc.).

[0348] Specific behavior:

[0349] The emotion engine analyzes comments and reactions when users read news articles and identifies emotions such as anxiety and vigilance.

[0350] Step 5:

[0351] The server evaluates the differences between the original news article and the marked text and generates the results as a transparency report.

[0352] Input: Original news article, marked text data.

[0353] Output: Transparency report.

[0354] Specific behavior:

[0355] The server compares the original sentence with the marked sentence and compiles a report that the expression "this is the worst thing that could happen" is an exaggeration.

[0356] Step 6:

[0357] The server generates advice to help users confirm the truth of information, reflecting the analysis results of the emotion engine and providing appropriate instructions according to the user's emotional state.

[0358] Input: Sentiment analysis results.

[0359] Output: Text data of advice.

[0360] Specific behavior:

[0361] If the user feels "anxious," the server generates advice such as "This news contains exaggerated statements. Please check the reliable sources below and try not to worry too much," and provides links to reliable sources.

[0362] Step 7:

[0363] The server integrates the "Insight Detector" function into the news app interface, allowing users to receive analysis results and advice in real time by clicking a specific button.

[0364] Input: User action (clicking a button).

[0365] Output: Display data of analysis results and advice.

[0366] Specific behavior:

[0367] When a user clicks the "Insight Detector" button in a news app, the analysis results and advice received from the server are immediately displayed. This feature allows users to obtain information while being aware of exaggeration and bias when reading the news.

[0368] (Application example 2)

[0369] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0370] Current information provision systems have difficulty detecting exaggerations and biases in the information users are exposed to and providing it in a transparent manner. As a result, users often make decisions based on incorrect information and believe unreliable information. Furthermore, because systems do not take the user's emotional state into account, the advice provided may not be appropriate for the user's psychological state. A comprehensive system is needed to solve these problems.

[0371] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for the user to input information, a means for the server to receive information, a means for the server to analyze the information using a generative model, a means for the server to mark parts containing exaggeration or bias, a means for the server to evaluate differences from the original information, a means for the server to generate a transparency report for the user, a means for the server to analyze the emotional state, and a means for the server to provide advice based on the emotional state. This makes it possible to detect exaggeration or bias in the information the user comes into contact with in real time and provide it in a highly transparent manner, and also to provide appropriate advice according to the user's emotional state.

[0372] A "user" is an entity that inputs information and receives feedback from the system.

[0373] "Information" is text data such as news and advertisements entered by the user.

[0374] A "server" is a computer system that analyzes information received from a user and performs the necessary processing.

[0375] A "generative model" is software that uses artificial intelligence algorithms to analyze information.

[0376] "Analysis" is the process of using generative models to evaluate information and identify its properties.

[0377] "Exaggeration" is when a piece of information is exaggerated in a way that is greater than the actual content.

[0378] "Bias" is a state in which some information is influenced by a particular perspective or prejudice.

[0379] "Marking" is tagging to explicitly indicate exaggerated or biased parts.

[0380] "Difference" refers to the difference between the original information and the analyzed information.

[0381] A "Transparency Report" is a report that clearly shows any exaggeration or bias in the information for users.

[0382] An "emotional state" is the psychological state a user feels when viewing information.

[0383] "Advice" is a set of instructions or suggestions that help users verify information without exaggeration or bias.

[0384] The present invention is a system that detects exaggeration and bias in the information a user comes into contact with in real time, provides the information in a highly transparent manner, and provides advice that takes into account the user's emotional state. Specific embodiments for realizing this system are described below.

[0385] Program processing overview

[0386] The system is implemented as an "Ad-Transparency Detector" application installed on smartphones or smart glasses. When a user views an ad, it analyzes the ad text and marks any exaggerated or biased parts, displaying them in real time. It also analyzes the user's emotions based on their browsing behavior and comments, and provides advice based on those emotions.

[0387] Hardware and Software

[0388] Hardware:

[0389] Smartphone or smart glasses: The device through which the user views the advertisement.

[0390] Server: A central management system that analyzes information and runs generative models.

[0391] software:

[0392] Generative models: Artificial intelligence algorithms that analyze information and identify exaggeration and bias.

[0393] Natural Language Processing Library (nltk): A library for analyzing the emotional state of the user.

[0394] Data Acquisition Library (requests): A library for acquiring text data for advertisements.

[0395] Specific examples

[0396] A user launches the "Ad-Transparency Detector" application and enters the URL of an ad. The ad includes a claim that "This product has achieved record sales!" The system's server receives the ad text and analyzes it using a generative model. The generative model determines that the "record sales" part is exaggerated and marks it.

[0397] The server then analyzes the user's input comments and determines that the user is feeling "anxious." Based on the user's emotional state, the server provides advice such as, "This ad contains exaggerated statements. Please also check the following reliable sources: official website, third-party review sites, etc. Don't worry too much."

[0398] Example of a prompt sentence to be entered

[0399] There is an advertisement that claims, "This product has achieved record sales!" Please evaluate whether this is an exaggeration and explain why. Furthermore, if a user viewing this advertisement feels "anxious," what advice would you give them?

[0400] In this way, users can easily assess the trustworthiness of the advertisements they come into contact with and make appropriate decisions.

[0401] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0402] Step 1:

[0403] The user enters the URL of an ad. The user launches the "Ad-Transparency Detector" application and enters or pastes the URL of the ad. At this time, the URL entered by the user is sent from the device to the server. The input data is the URL of the ad, and the output data is the URL sent to the server.

[0404] Step 2:

[0405] The server retrieves the ad text. Using the received URL, the server uses a data retrieval library (requests) to retrieve the HTML content of the corresponding ad page. The input data is the ad URL, and the output data is the retrieved ad text. Specifically, an HTTP request is sent and the server downloads the HTML source of the ad page.

[0406] Step 3:

[0407] The server analyzes the ad text. The acquired ad text is passed to a generative model, which analyzes the information. The generative model uses an artificial intelligence algorithm to evaluate the ad text and identify parts that contain exaggeration or bias. The input data is the ad text, and the output data is the analysis results. Specifically, natural language processing technology is used to evaluate the context.

[0408] Step 4:

[0409] The server marks exaggerated or biased parts. Parts with high levels of exaggeration or bias identified by the generative model are tagged with specific tags (e.g., <exaggerated>, <biased>) to make them easy for users to understand. The input data is the analysis results, and the output data is the marked advertising text. Specifically, tags are inserted into the relevant parts of the text.

[0410] Step 5:

[0411] The server analyzes the user's emotional state. It uses a natural language processing library (nltk) to analyze the emotional state based on the comments and feedback entered by the user when viewing the ad text. The input data is the user's comments, and the output data is the emotional state (e.g., anxiety, joy). Specifically, it applies a sentiment analysis algorithm to calculate an emotional score.

[0412] Step 6:

[0413] The server evaluates the differences from the original information. It compares the original ad text with the marked ad text and generates a transparency report to indicate to the user any exaggerations or biases. The input data is the original ad text and the marked ad text, and the output data is a transparency report. Specifically, it uses a text comparison algorithm to extract the differences.

[0414] Step 7:

[0415] The server provides advice based on the emotional state. Advice reflecting the emotional state is generated and presented to the user. The input data is the emotional state and a transparency report, and the output data is advice. Specifically, an appropriate message is generated according to the emotional state and presented to the user as feedback.

[0416] Step 8:

[0417] The server sends the transparency report and advice to the terminal. The generated transparency report and advice are sent to the device used by the user (smartphone or smart glasses) and displayed in real time. The input data are the transparency report and advice, and the output data is the information displayed on the user's terminal. Specifically, the data is sent via network communication.

[0418] 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.

[0419] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0420] 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.

[0421] [Second embodiment]

[0422] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0423] 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.

[0424] 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).

[0425] 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.

[0426] 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.

[0427] 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).

[0428] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0429] 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.

[0430] 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.

[0431] 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.

[0432] In the smart glasses 214, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0433] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0434] The present invention is a system in which users input information, a server analyzes the information, identifies exaggerations and biases, and provides users with transparent reports and advice. The following describes the program processing of this system in detail.

[0435] Get news articles

[0436] The user opens the news app, enters the URL or text of the news article they want to read, and when the user selects an article within the news app, the device sends it to the server.

[0437] News article analysis

[0438] The server passes the received news article to an internal generative model, which uses artificial intelligence to analyze the article's content by evaluating each sentence in context and identifying any parts that may contain exaggeration or bias.

[0439] Examples:

[0440] News article text:

[0441] "The company's stock price has plummeted. Experts say this is the worst event in the last 20 years."

[0442] The AI ​​model marks the "worst case scenario" as having a high exaggeration score.

[0443] Extracting exaggerated and biased sentences

[0444] The server extracts exaggerated or highly biased parts identified by the generative model and marks them, explicitly indicating them to the user with a specific tag (e.g., <exaggerated>).

[0445] Evaluating differences and providing transparency

[0446] The server evaluates the differences between the original news article and the marked text and generates a transparency report that clearly shows which parts contain exaggerations or bias, in a format that is easy for users to understand.

[0447] Examples:

[0448] Original news article: The company's stock price has plummeted, in what experts say is the worst it's seen in 20 years.

[0449] Transparency Report: "The worst thing that could happen" is an exaggeration. Please check with another source.

[0450] Providing advice

[0451] The server generates specific advice to help users verify the truth of the information, for example, recommending reference to other reliable sources or encouraging users to check official statistics.

[0452] Examples:

[0453] This news story contains exaggerated statements. Please verify with the following independent sources:

[0454] 1. Official statistical data

[0455] 2. Other trusted news sites

[0456] 3. In-depth analysis by experts

[0457] Integration into news apps

[0458] The server will integrate the "Insight Detector" function into the news app interface, allowing users to click a specific button while browsing the news to receive analysis results and advice in real time.

[0459] In this way, the present invention provides users with highly transparent information and supports honest and unbiased decision making.

[0460] The processing flow will be explained below.

[0461] Step 1:

[0462] A user opens a news app and enters the URL or text of a news article, or selects an article within the news app.

[0463] Step 2:

[0464] The device sends the URL or text entered by the user to the server.

[0465] Step 3:

[0466] The server passes the received news articles to an internal generative model, which uses artificial intelligence algorithms to analyze the article's content.

[0467] Step 4:

[0468] A generative model in the server evaluates each sentence in the article based on its context, identifying any parts that may contain exaggeration or bias.

[0469] Step 5:

[0470] The server extracts the exaggerated or biased parts identified by the generative model and marks them, for example, with specific tags (<exaggerated>, <biased>).

[0471] Step 6:

[0472] The server evaluates the differences between the original news article and the marked text and generates the results as a transparency report.

[0473] Step 7:

[0474] The server generates a transparency report in text format and displays it to the user.

[0475] Step 8:

[0476] The server generates and provides textual advice to the user on how to verify the truth of the information, such as by recommending a list of other reliable sources or official statistical data.

[0477] Step 9:

[0478] The server integrates the "Insight Detector" function into the news app interface, allowing users to view analysis results and advice in real time as they browse the news.

[0479] Example 1

[0480] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0481] In today's information society, amidst the vast amount of news articles and information being distributed, much of the information contains exaggeration and bias. This makes it difficult for users to obtain accurate and objective information, which can lead to incorrect decision-making. There is a need to solve this problem and provide users with highly transparent and reliable information.

[0482] 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.

[0483] In this invention, the server includes: a means for a user to input information; a means for a terminal to send the input information to the server; a means for the server to receive information; a means for the server to analyze the information using a generative AI model; a means for the server to evaluate the content of an article and identify exaggerations or biases; a means for the server to mark parts containing exaggerations or biases; a means for the server to evaluate differences from the original information and generate a transparency report; a means for the server to provide advice for verifying the truth of the information; and a means for the server to display the analysis results and advice on a user interface. This makes it easier for users to obtain reliable information and enables them to make accurate and unbiased decisions.

[0484] "User" means a human or end-user who uses the system to input information and receive analytical results and transparency reports.

[0485] A "terminal" is an electronic device that a user uses to input information and that has the function of transmitting information to a server.

[0486] A "server" is a computer system that processes information received from a terminal and analyzes it using a generative AI model, and is a device that generates and provides various reports.

[0487] A "generative AI model" is a model that uses artificial intelligence technology to analyze information such as news articles, and is an algorithm used in particular to identify exaggeration or bias in information.

[0488] "Information" is a general term for news articles, text data, etc. entered by users, and is the data to be analyzed.

[0489] "Exaggeration" refers to information that overly exaggerates reality and lacks accuracy.

[0490] "Bias" refers to information that is biased toward a particular position or perspective, and is an unfair representation.

[0491] A "Transparency Report" is a document generated by the server that clearly shows the difference between the original information and the analysis results, and aims to improve the reliability of the information for users.

[0492] "Advice" is a recommendation or instruction provided by the server to help users verify truthful information, pointing to other reliable sources or verification methods.

[0493] "User interface" refers to the screen or operating means that allows the user to visually check the analysis results and advice from the server.

[0494] The present invention is a system in which a user inputs news information, and a server analyzes the information to identify exaggerations and biases, and provides a highly transparent report and specific advice to the user. Hereinafter, specific embodiments of this system will be described.

[0495] This system starts when a user inputs information through a news app and the information is sent to the server. When the user inputs the URL or text of a news article, the terminal sends this information to the server. The server passes the received news article to an internal generative AI model, and this generative AI model analyzes the content of the article. As the generative AI model, for example, a model using natural language processing (NLP) is used.

[0496] In the analysis of the news article, the server evaluates each sentence of the article based on the context and identifies parts containing exaggerations or biases. For example, suppose there is a news article as follows:

[0497] "The company's stock price has dropped sharply. Experts say this is the worst thing in the past 20 years."

[0498] The generative AI model marks the expression "the worst thing" with a high exaggeration score. After parts with exaggerations or biases are identified, the server marks these parts using specific tags (e.g., <exaggeration>). At this time, as a specific example, it is as follows:

[0499] "The company's stock price has dropped sharply. Experts say this is <exaggeration>the worst thing< / exaggeration> in the past 20 years."

[0500] Subsequently, the server evaluates the difference between the original news article and the marked article and generates the result as a transparency report. The transparency report clearly shows which parts contain exaggerations or biases and is a document for improving the reliability of information for the user. As a specific example, it is as follows:

[0501] Original news article: "The company's stock price has plummeted, with experts calling it the worst in 20 years."

[0502] Transparency Report: "'The worst thing that could have happened' is an exaggeration. Please verify with another source."

[0503] Additionally, the server generates specific advice to help users verify the truth of the information, such as recommending other reliable sources or checking official statistics. Examples of advice provided to users include:

[0504] "This news story contains exaggerated statements. Please check with the following independent sources:

[0505] 1. Official statistical data

[0506] 2. Other trusted news sites

[0507] 3. Detailed analysis by experts

[0508] Finally, the server will integrate the "Insight Detector" feature into the news app interface, allowing users to click a specific button while browsing the news to receive real-time analysis and advice.

[0509] In this way, the present invention provides users with highly transparent information and supports honest and unbiased decision making.

[0510] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0511] Step 1:

[0512] The user opens a news app, which is ready to input information. The user enters the URL or text of the news article they want to read. The news app passes the information to the device.

[0513] Input: News article URL or text

[0514] Output: News article information sent to device

[0515] Step 2:

[0516] The terminal sends the entered news article information to the server. Specifically, the terminal sends the information to the server via an HTTP request.

[0517] Input: News article URL or text (on device)

[0518] Output: News article information sent to the server

[0519] Step 3:

[0520] A server receives news article information, which it prepares for input into an analytical model.

[0521] Input: News article information sent from the device

[0522] Output: News article information formatted for analysis

[0523] Step 4:

[0524] The server inputs the news article into the generative AI model, which then makes an API request to the model to begin analysis.

[0525] Input: News article information formatted for analysis

[0526] Output: News article information fed into the generative AI model

[0527] Step 5:

[0528] The generative AI model evaluates each sentence in a news article based on its context and identifies parts that may contain exaggeration or bias. Specifically, the model analyzes the text using natural language processing techniques.

[0529] Input: News article information

[0530] Output: Identification of exaggeration or bias

[0531] Step 6:

[0532] The server extracts exaggerated or highly biased parts from the output of the generative AI model, and marks the identified parts with a corresponding tag (e.g., <exaggerated>).

[0533] Input: Results identifying exaggeration or bias

[0534] Output: Tagged news article information

[0535] Step 7:

[0536] The server evaluates the differences between the original news article and the marked sentences, making it clear which parts contain exaggeration or bias.

[0537] Input: tagged news article information

[0538] Output: The result of the difference evaluation.

[0539] Step 8:

[0540] The server generates a transparency report based on the results of the difference evaluation, which is output in a format that is easy for users to understand.

[0541] Input: Result of the difference evaluation

[0542] Output: Transparency Report

[0543] Step 9:

[0544] The server generates specific advice to provide to the user, including a recommendation to refer to a trusted source.

[0545] Input: Difference Assessment Results and Transparency Report

[0546] Output: Specific advice

[0547] Step 10:

[0548] The server displays the analysis results and advice on the news app interface. Specifically, when the user clicks a specific button, the analysis results and advice are displayed in real time.

[0549] Input: Transparency reports and specific advice

[0550] Output: Analysis results and advice displayed in the user interface

[0551] Through this series of processing steps, users can obtain highly transparent information, which can support honest and unbiased decision-making.

[0552] (Application example 1)

[0553] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0554] In today's world, there is an increasing risk that users will be misled by exaggerations and bias in advertisements and news articles. As a result, users may make decisions based on inaccurate information. In particular, in an information-overloaded environment, unreliable information is easily mixed in, making it difficult for users to discern accurate information. The objective of this invention is to automatically identify exaggerations and biases in advertisement content and news articles, and provide users with transparent reports and specific advice to support honest and unbiased decision-making.

[0555] 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.

[0556] In this invention, the server includes: a means for a user to input information; a means for receiving information; a means for analyzing the information using a generative model; a means for marking parts containing exaggeration or bias; a means for evaluating differences from the original information; a means for generating a transparency report for the user; a means for providing advice for verifying the truth of the information; and a means for acquiring and analyzing advertising content, identifying exaggeration or bias, and providing a transparency report and specific advice. This allows the reliability of advertisements and news articles to be automatically evaluated, enabling users to make honest and unbiased decisions.

[0557] A "user" is someone who uses the system to input or receive information.

[0558] "Information" refers to news articles, advertising content, and other text data and content in general that are subject to analysis.

[0559] A "server" is a computer system that receives, analyzes, and evaluates information sent by users and returns the results.

[0560] A "generative model" refers to an algorithm or data model that uses artificial intelligence to analyze data.

[0561] "Exaggeration" refers to parts that are exaggerated in a way that is greater than the actual content.

[0562] "Bias" refers to a part that contains a particular perspective or preconception.

[0563] "Marking" refers to the act of identifying exaggerations or biases and explicitly labeling them.

[0564] "Differential evaluation" is the process of evaluating the differences between the original information and the analyzed information.

[0565] A "Transparency Report" is a report that identifies areas that contain exaggeration or bias and presents them in a user-friendly format.

[0566] "Advice" means specific instructions or suggestions provided to a user to help them make a good faith decision.

[0567] "Advertising Content" refers to promotional text and multimedia content relating to products and services.

[0568] "Analysis" is the process of evaluating and examining user-provided information for a specific purpose.

[0569] The system embodying this invention analyzes input information from users, identifies exaggeration and bias, generates transparency reports, and provides specific advice. Specific steps for implementing this invention are described below.

[0570] Configuration and operation explanation

[0571] Hardware configuration:

[0572] User devices: Mobile devices such as smartphones and tablets

[0573] Server: A high-performance computer or cloud server

[0574] Network: Internet

[0575] Software configuration:

[0576] Generative AI models: Transformer-based AI model libraries (e.g., Hugging Face Transformers)

[0577] Analysis program: Python script

[0578] News app: an interface where users enter information

[0579] A user accesses a news app or ad-checking app and enters the information about the ad or news article they want to analyze. The entered information is sent from the user's device to a server. The server then supplies the received information to a generative AI model, which analyzes it for exaggeration and bias.

[0580] The server uses a generative AI model to analyze each piece of an article or ad, taking into account the context and assessing whether a particular phrase or expression is exaggerated or biased.

[0581] If exaggeration or bias is identified, the server will flag these and evaluate the deviation from the original information. For example, if a news article mentions "the best product on the market," it will be deemed exaggerated.

[0582] Additionally, the server generates a transparency report that explicitly shows users where there is exaggeration or bias, detailing the alleged exaggeration or bias and explaining why it is problematic.

[0583] Finally, the server provides specific advice to the user, such as recommending verification from a trusted source.

[0584] Specific examples

[0585] Example: A user enters the following ad text in a news app: "The miracle diet! Lose 10kg in just 10 days."

[0586] The server analyzes this text using a generative AI model and identifies phrases like "miracle" and "lose 10kg in just 10 days" as exaggerations. It then generates a transparency report explaining to the user how these phrases are exaggerated. It also provides the user with specific advice: "This information is exaggerated. Please check other reliable sources."

[0587] Example prompt sentence:

[0588] "Below is the content of the advertisement. Please analyze this content for exaggeration or bias and provide a transparent report.

[0589] Advertisement: Miracle diet method! Lose 10kg in just 10 days.

[0590] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0591] Program processing flow and detailed explanation of each step

[0592] Step 1:

[0593] The user enters the information.

[0594] The user opens a news app or ad check app and enters the text of the ad or news article to be analyzed, or specifies a URL. This is the input information. The input information is sent from the user's device to the server.

[0595] Step 2:

[0596] The server receives the information.

[0597] The server receives the information sent by the user. The received information is then directly supplied to the generative AI model for analysis. Through this receiving operation, the server obtains input data (news articles and advertising text).

[0598] Step 3:

[0599] The server analyzes the information using a generative AI model.

[0600] The server inputs the received information into a generative AI model (e.g., Hugging Face Transformers) for analysis. Specifically, the generative AI model evaluates each piece of text and determines whether it contains exaggeration or bias. The input data is processed and an analysis result is generated.

[0601] Step 4:

[0602] The server marks the parts that contain exaggeration or bias.

[0603] The server marks parts of the original text that are identified as exaggerated or biased based on the analysis results obtained from the generative AI model. The marking uses tags (e.g., <exaggeration>) to clearly indicate which parts of the original text are exaggerated. The marked text is then output.

[0604] Step 5:

[0605] The server evaluates the difference from the original information.

[0606] The server evaluates the differences between the original information and the marked information, and sorts out which parts have been exaggerated and how. At this stage, the difference evaluation results are generated.

[0607] Step 6:

[0608] The server generates a transparency report.

[0609] The server generates a transparency report based on the difference evaluation results, detailing any exaggerations or biases and explaining why. The report is output as feedback to the user.

[0610] Step 7:

[0611] The server provides advice to verify the truth of the information.

[0612] Based on the transparency report, the server generates specific advice for users to verify accurate information, including recommendations for reliable sources, and provides this advice to users along with the report.

[0613] For example, if a user enters the ad copy "Miracle Diet! Lose 10kg in just 10 days," the server receives it and analyzes it using a generative AI model. It identifies and marks the exaggerated parts "miracle" and "Lose 10kg in just 10 days." It then evaluates the difference between the original text and the marked text, and generates a transparency report with specific advice, such as "This text contains exaggerated parts. Please check other reliable sources," and provides it to the user.

[0614] 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.

[0615] The present invention is a system that identifies exaggeration and bias in the information entered by a user and provides a transparency report and advice that takes into account the user's feelings. The program processing of the system will be described in detail below.

[0616] Get news articles

[0617] When a user opens a news app and enters the URL or text of a news article, or selects an article within the news app, the device sends this to the server.

[0618] News article analysis

[0619] The server receives news articles and passes them to an internal generative model, which uses artificial intelligence algorithms to analyze the content of the article, evaluate the context of the sentences, and identify any parts that may contain exaggeration or bias.

[0620] Examples:

[0621] News article text:

[0622] "The company's stock price has plummeted. Experts say this is the worst event in the last 20 years."

[0623] The AI ​​model marks the "worst case scenario" as having a high exaggeration score.

[0624] Extracting exaggerated and biased sentences

[0625] The server extracts the exaggerated or biased parts identified by the generative model and marks them, explicitly indicating them to the user with specific tags (<exaggerated>, <biased>).

[0626] Emotion Analysis

[0627] The server uses an emotion engine to analyze emotions based on the text entered by the user and the selections made by the user, and this emotion analysis identifies the emotional state (e.g., anger, anxiety, joy, etc.) that the user is in when reading a news article.

[0628] Examples:

[0629] When a user reads a news article, the emotion engine detects "anxiety" from the expression in the text input.

[0630] Evaluating differences and providing transparency

[0631] The server evaluates the differences between the original news article and the marked text and generates a transparency report that clearly shows which parts contain exaggerations or bias, in a user-friendly format.

[0632] Examples:

[0633] Original news article: The company's stock price has plummeted, in what experts say is the worst it's seen in 20 years.

[0634] Transparency Report: "The worst thing that could happen" is an exaggeration. Please check with another source.

[0635] Providing advice

[0636] The server generates specific advice for the user to confirm the truth of the information. This advice reflects the analysis results of the emotion engine and provides appropriate instructions according to the user's emotional state.

[0637] Examples:

[0638] If users are feeling "uneasy," they should be warned: "This news contains exaggerated statements. Please check the following reliable sources and try not to worry too much:

[0639] 1. Official statistical data

[0640] 2. Other trusted news sites

[0641] 3. In-depth analysis by experts

[0642] Integration into news apps

[0643] Sarva will integrate the "Insight Detector" function into the news app interface. This will allow users to click a specific button when viewing news to receive analysis results and advice in real time. By providing highly transparent information that takes user emotions into consideration, Sarva will support honest and unbiased decision-making.

[0644] Thus, the present invention is a system that takes into account the user's emotional state while providing highly transparent information without exaggeration or bias.

[0645] The processing flow will be explained below.

[0646] Step 1:

[0647] A user opens a news app and enters the URL or text of a news article, or selects an article within the news app.

[0648] Step 2:

[0649] The device sends the URL or text entered by the user to the server.

[0650] Step 3:

[0651] The server passes the received news articles to an internal generative model, which uses artificial intelligence algorithms to analyze the article's content.

[0652] Step 4:

[0653] A generative model in the server evaluates each sentence in the article based on its context, identifying any parts that may contain exaggeration or bias.

[0654] Step 5:

[0655] The server extracts the exaggerated or biased parts identified by the generative model and marks them, for example, with specific tags (<exaggerated>, <biased>).

[0656] Step 6:

[0657] The server activates an emotion engine to recognize the user's emotions while the user is reading the article.

[0658] Step 7:

[0659] The emotion engine in the server analyzes the user's current emotional state (anger, anxiety, joy, etc.) based on the user's text input and selection operations.

[0660] Examples:

[0661] When a user reads a news article, the emotion engine recognizes that they are feeling "anxiety."

[0662] Step 8:

[0663] The server evaluates the differences between the original news article and the marked text and generates a transparency report that clearly shows which parts contain exaggerations or bias.

[0664] Examples:

[0665] Original news article: The company's stock price has plummeted, in what experts say is the worst it's seen in 20 years.

[0666] Transparency Report: "The worst thing that could happen" is an exaggeration. Please check with another source.

[0667] Step 9:

[0668] The server generates appropriate advice for the user based on the results of the analysis by the emotion engine. For example, if the user is feeling anxious, the server will provide advice to calm the user.

[0669] Examples:

[0670] "This news contains exaggerated statements. Please check the following reliable sources and don't worry too much:

[0671] 1. Official statistical data

[0672] 2. Other trusted news sites

[0673] 3. Detailed analysis by experts

[0674] Step 10:

[0675] The server integrates the "Insight Detector" function into the news app interface, allowing users to view analysis results and advice in real time as they browse the news.

[0676] In this way, the server takes into account the user's emotional state and provides transparent information without exaggeration or bias, thereby supporting honest and unbiased decision-making.

[0677] Example 2

[0678] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0679] In recent years, the amount of information on the Internet has increased dramatically, and the news and articles users encounter often contain exaggerations and bias. As a result, users are at greater risk of receiving incorrect information or holding biased viewpoints. Furthermore, when reading news and articles, users' emotions can significantly influence how they perceive the information, making it difficult for users to objectively evaluate it. In such a situation, there is a need for accurate and transparent information provision.

[0680] 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.

[0681] In this invention, the server includes a means for a user to input information, a means for a terminal to send information to the server, a means for the server to receive information, a means for the server to analyze the information using a generative model, a means for the server to mark parts containing exaggeration or bias, a means for the server to analyze the user's emotions, a means for the server to evaluate differences from the original information, a means for the server to generate a transparency report for the user, and a means for the server to provide advice for verifying the truth of information. This allows users to recognize exaggeration and bias when reading news or articles, understand their own emotional state, and obtain more accurate and transparent information.

[0682] "User" refers to a person who uses this system to input information and view news and articles.

[0683] "Terminal" refers to a device used by a user that provides a means for transmitting information to a server.

[0684] "Server" refers to the main computing system that receives and analyzes information sent from users and terminals.

[0685] A "generative model" refers to an algorithm or system that uses artificial intelligence to analyze and evaluate documents.

[0686] "Exaggeration or bias" refers to exaggerated or biased viewpoints in texts or news articles.

[0687] "Marking" refers to the act of explicitly indicating exaggerated or biased parts using specific tags or highlighting.

[0688] "Emotion analysis" refers to the process of identifying a user's emotional state (anger, anxiety, joy, etc.) based on the user's input text and actions.

[0689] "Differential assessment" refers to assessing the difference between the original information and the information that has been marked with exaggeration or bias.

[0690] A "transparency report" refers to a report that clearly shows users which parts contain exaggeration or bias.

[0691] "Advice" refers to the act of providing specific instructions or recommendations for users to verify truthful information.

[0692] The present invention provides a system that identifies exaggerations and biases in information entered by a user and provides transparency reports and advice that take the user's emotions into account. This system is configured by combining a user, a terminal, a server, and a generative model.

[0693] Get news articles

[0694] The user opens the news app and enters the URL or text of a news article, or can select an article within the news app, and the device then sends the information to the server.

[0695] News article analysis

[0696] The server passes the received news article to an internal generative model (e.g., GPT-4), which uses natural language processing techniques to analyze the article's content, specifically evaluating the context of each sentence and identifying parts that may contain exaggeration or bias.

[0697] Examples:

[0698] News article text:

[0699] "The company's stock price has plummeted. Experts say this is the worst event in the last 20 years."

[0700] The AI ​​model marks the "worst case scenario" as having a high exaggeration score.

[0701] Extracting exaggerated and biased sentences

[0702] The server extracts the exaggerated or biased parts identified by the generative model and marks them explicitly, using tags such as "<exaggerated>" and "<biased>".

[0703] Emotion Analysis

[0704] The server uses an emotion engine to analyze emotions based on the text entered by the user and the selections made, thereby identifying the emotional state (e.g., anger, anxiety, joy) the user is in when reading a news article.

[0705] Examples:

[0706] When a user reads a news article, the emotion engine detects "anxiety" from the expression in the text input.

[0707] Evaluating differences and providing transparency

[0708] The server evaluates the differences between the original news article and the marked text and generates a transparency report that clearly shows users which parts contain exaggerations or bias, in an easy-to-understand format.

[0709] Examples:

[0710] Original news article: "The company's stock price has plummeted, with experts saying it's the worst in 20 years."

[0711] Transparency Report: "The worst thing that could happen" is an exaggeration. Please check with another source.

[0712] Providing advice

[0713] The server generates specific advice for the user to confirm the truth of the information. This advice reflects the analysis results of the emotion engine and provides appropriate instructions according to the user's emotional state.

[0714] Examples:

[0715] If users are feeling "uneasy," they should be advised that "This news contains exaggerated statements. Please check the following reliable sources and do not worry too much:

[0716] 1. Official statistical data

[0717] 2. Other trusted news sites

[0718] 3. In-depth analysis by experts

[0719] Integration into news apps

[0720] Sarva will integrate the "Insight Detector" feature into the news app interface. This will allow users to click a specific button while browsing the news to receive analysis results and advice in real time. This feature will enable users to make honest and unbiased decisions while receiving transparent information that takes emotions into account.

[0721] This system takes into account the user's emotions and provides highly transparent information without exaggeration or bias.

[0722] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0723] Step 1:

[0724] The user launches a news app and enters the URL or text of a news article, or selects an article within the news app. The device then sends this to the server.

[0725] Input: News article URL or text entered by the user.

[0726] Output: The URL or text of the news article entered is sent to the server.

[0727] Specific behavior:

[0728] When a user pastes a URL into the search bar of a news app or taps on a specific article from the "Recent News" section, the device instantly sends it to the server.

[0729] Step 2:

[0730] The server passes the received news article to a generative model, which uses natural language processing techniques to analyze the article's content.

[0731] Input: News article text data.

[0732] Output: Contextual data analyzed using natural language processing techniques.

[0733] Specific behavior:

[0734] The server feeds the text of a news article into a generative model, which evaluates the context of each sentence to identify potential exaggerations or biases.

[0735] Step 3:

[0736] The server extracts the exaggerated or biased parts identified by the generative model and marks them to indicate this explicitly.

[0737] Input: Analyzed contextual data, identifying exaggerations and biases.

[0738] Output: Marked text data.

[0739] Specific behavior:

[0740] The server tags the identified phrases with tags such as "<exaggeration>" or "<bias>." For example, the phrase "the worst thing that could happen" in a sentence might be tagged with "<exaggeration>."

[0741] Step 4:

[0742] The server uses an emotion engine to analyze emotions based on the user's text input and selections.

[0743] Input: User input text and operation data.

[0744] Output: Sentiment analysis results (e.g., anxiety, anger, joy, etc.).

[0745] Specific behavior:

[0746] The emotion engine analyzes comments and reactions when users read news articles and identifies emotions such as anxiety and vigilance.

[0747] Step 5:

[0748] The server evaluates the differences between the original news article and the marked text and generates the results as a transparency report.

[0749] Input: Original news article, marked text data.

[0750] Output: Transparency report.

[0751] Specific behavior:

[0752] The server compares the original sentence with the marked sentence and compiles a report that the expression "this is the worst thing that could happen" is an exaggeration.

[0753] Step 6:

[0754] The server generates advice to help users confirm the truth of information, reflecting the analysis results of the emotion engine and providing appropriate instructions according to the user's emotional state.

[0755] Input: Sentiment analysis results.

[0756] Output: Text data of advice.

[0757] Specific behavior:

[0758] If the user feels "anxious," the server generates advice such as "This news contains exaggerated statements. Please check the reliable sources below and try not to worry too much," and provides links to reliable sources.

[0759] Step 7:

[0760] The server integrates the "Insight Detector" function into the news app interface, allowing users to receive analysis results and advice in real time by clicking a specific button.

[0761] Input: User action (clicking a button).

[0762] Output: Display data of analysis results and advice.

[0763] Specific behavior:

[0764] When a user clicks the "Insight Detector" button in a news app, the analysis results and advice received from the server are immediately displayed. This feature allows users to obtain information while being aware of exaggeration and bias when reading the news.

[0765] (Application example 2)

[0766] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0767] Current information provision systems have difficulty detecting exaggerations and biases in the information users are exposed to and providing it in a transparent manner. As a result, users often make decisions based on incorrect information and believe unreliable information. Furthermore, because systems do not take the user's emotional state into account, the advice provided may not be appropriate for the user's psychological state. A comprehensive system is needed to solve these problems.

[0768] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for the user to input information, a means for the server to receive information, a means for the server to analyze the information using a generative model, a means for the server to mark parts containing exaggeration or bias, a means for the server to evaluate differences from the original information, a means for the server to generate a transparency report for the user, a means for the server to analyze the emotional state, and a means for the server to provide advice based on the emotional state. This makes it possible to detect exaggeration or bias in the information the user comes into contact with in real time and provide it in a highly transparent manner, and also to provide appropriate advice according to the user's emotional state.

[0769] A "user" is an entity that inputs information and receives feedback from the system.

[0770] "Information" is text data such as news and advertisements entered by the user.

[0771] A "server" is a computer system that analyzes information received from a user and performs the necessary processing.

[0772] A "generative model" is software that uses artificial intelligence algorithms to analyze information.

[0773] "Analysis" is the process of using generative models to evaluate information and identify its properties.

[0774] "Exaggeration" is when a piece of information is exaggerated in a way that is greater than the actual content.

[0775] "Bias" is a state in which some information is influenced by a particular perspective or prejudice.

[0776] "Marking" is tagging to explicitly indicate exaggerated or biased parts.

[0777] "Difference" refers to the difference between the original information and the analyzed information.

[0778] A "Transparency Report" is a report that clearly shows any exaggeration or bias in the information for users.

[0779] An "emotional state" is the psychological state a user feels when viewing information.

[0780] "Advice" is a set of instructions or suggestions that help users verify information without exaggeration or bias.

[0781] The present invention is a system that detects exaggeration and bias in the information a user comes into contact with in real time, provides the information in a highly transparent manner, and provides advice that takes into account the user's emotional state. Specific embodiments for realizing this system are described below.

[0782] Program processing overview

[0783] The system is implemented as an "Ad-Transparency Detector" application installed on smartphones or smart glasses. When a user views an ad, it analyzes the ad text and marks any exaggerated or biased parts, displaying them in real time. It also analyzes the user's emotions based on their browsing behavior and comments, and provides advice based on those emotions.

[0784] Hardware and Software

[0785] Hardware:

[0786] Smartphone or smart glasses: The device through which the user views the advertisement.

[0787] Server: A central management system that analyzes information and runs generative models.

[0788] software:

[0789] Generative models: Artificial intelligence algorithms that analyze information and identify exaggeration and bias.

[0790] Natural Language Processing Library (nltk): A library for analyzing the emotional state of the user.

[0791] Data Acquisition Library (requests): A library for acquiring text data for advertisements.

[0792] Specific examples

[0793] A user launches the "Ad-Transparency Detector" application and enters the URL of an ad. The ad includes a claim that "This product has achieved record sales!" The system's server receives the ad text and analyzes it using a generative model. The generative model determines that the "record sales" part is exaggerated and marks it.

[0794] The server then analyzes the user's input comments and determines that the user is feeling "anxious." Based on the user's emotional state, the server provides advice such as, "This ad contains exaggerated statements. Please also check the following reliable sources: official website, third-party review sites, etc. Don't worry too much."

[0795] Example of a prompt to be entered

[0796] There is an advertisement that claims, "This product has achieved record sales!" Please evaluate whether this is an exaggeration and explain why. Furthermore, if a user viewing this advertisement feels "anxious," what advice would you give them?

[0797] In this way, users can easily assess the trustworthiness of the advertisements they come into contact with and make appropriate decisions.

[0798] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0799] Step 1:

[0800] The user enters the URL of an ad. The user launches the "Ad-Transparency Detector" application and enters or pastes the URL of the ad. At this time, the URL entered by the user is sent from the device to the server. The input data is the URL of the ad, and the output data is the URL sent to the server.

[0801] Step 2:

[0802] The server retrieves the ad text. Using the received URL, the server uses a data retrieval library (requests) to retrieve the HTML content of the corresponding ad page. The input data is the ad URL, and the output data is the retrieved ad text. Specifically, an HTTP request is sent and the server downloads the HTML source of the ad page.

[0803] Step 3:

[0804] The server analyzes the ad text. The acquired ad text is passed to a generative model, which analyzes the information. The generative model uses an artificial intelligence algorithm to evaluate the ad text and identify parts that contain exaggeration or bias. The input data is the ad text, and the output data is the analysis results. Specifically, natural language processing technology is used to evaluate the context.

[0805] Step 4:

[0806] The server marks exaggerated or biased parts. Parts with high levels of exaggeration or bias identified by the generative model are tagged with specific tags (e.g., <exaggerated>, <biased>) to make them easy for users to understand. The input data is the analysis results, and the output data is the marked advertising text. Specifically, tags are inserted into the relevant parts of the text.

[0807] Step 5:

[0808] The server analyzes the user's emotional state. It uses a natural language processing library (nltk) to analyze the emotional state based on the comments and feedback entered by the user when viewing the ad text. The input data is the user's comments, and the output data is the emotional state (e.g., anxiety, joy). Specifically, it applies a sentiment analysis algorithm to calculate an emotional score.

[0809] Step 6:

[0810] The server evaluates the differences from the original information. It compares the original ad text with the marked ad text and generates a transparency report to indicate to the user any exaggerations or biases. The input data is the original ad text and the marked ad text, and the output data is a transparency report. Specifically, it uses a text comparison algorithm to extract the differences.

[0811] Step 7:

[0812] The server provides advice based on the emotional state. Advice reflecting the emotional state is generated and presented to the user. The input data is the emotional state and a transparency report, and the output data is advice. Specifically, an appropriate message is generated according to the emotional state and presented to the user as feedback.

[0813] Step 8:

[0814] The server sends the transparency report and advice to the terminal. The generated transparency report and advice are sent to the device used by the user (smartphone or smart glasses) and displayed in real time. The input data are the transparency report and advice, and the output data is the information displayed on the user's terminal. Specifically, the data is sent via network communication.

[0815] 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.

[0816] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0817] 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.

[0818] [Third embodiment]

[0819] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0820] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0821] 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).

[0822] 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.

[0823] 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.

[0824] 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).

[0825] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0826] 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.

[0827] 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.

[0828] 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.

[0829] 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.

[0830] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0831] The present invention is a system in which users input information, a server analyzes the information, identifies exaggerations and biases, and provides users with transparent reports and advice. The following describes the program processing of this system in detail.

[0832] Get news articles

[0833] The user opens the news app, enters the URL or text of the news article they want to read, and when the user selects an article within the news app, the device sends it to the server.

[0834] News article analysis

[0835] The server passes the received news article to an internal generative model, which uses artificial intelligence to analyze the article's content by evaluating each sentence in context and identifying any parts that may contain exaggeration or bias.

[0836] Examples:

[0837] News article text:

[0838] "The company's stock price has plummeted. Experts say this is the worst event in the last 20 years."

[0839] The AI ​​model marks the "worst case scenario" as having a high exaggeration score.

[0840] Extracting exaggerated and biased sentences

[0841] The server extracts exaggerated or highly biased parts identified by the generative model and marks them, explicitly indicating them to the user with a specific tag (e.g., <exaggerated>).

[0842] Evaluating differences and providing transparency

[0843] The server evaluates the differences between the original news article and the marked text and generates a transparency report that clearly shows which parts contain exaggerations or bias, in a format that is easy for users to understand.

[0844] Examples:

[0845] Original news article: The company's stock price has plummeted, in what experts say is the worst it's seen in 20 years.

[0846] Transparency Report: "The worst thing that could happen" is an exaggeration. Please check with another source.

[0847] Providing advice

[0848] The server generates specific advice to help users verify the truth of the information, for example, recommending reference to other reliable sources or encouraging users to check official statistics.

[0849] Examples:

[0850] This news story contains exaggerated statements. Please verify with the following independent sources:

[0851] 1. Official statistical data

[0852] 2. Other trusted news sites

[0853] 3. In-depth analysis by experts

[0854] Integration into news apps

[0855] The server will integrate the "Insight Detector" function into the news app interface, allowing users to click a specific button while browsing the news to receive analysis results and advice in real time.

[0856] In this way, the present invention provides users with highly transparent information and supports honest and unbiased decision making.

[0857] The processing flow will be explained below.

[0858] Step 1:

[0859] A user opens a news app and enters the URL or text of a news article, or selects an article within the news app.

[0860] Step 2:

[0861] The device sends the URL or text entered by the user to the server.

[0862] Step 3:

[0863] The server passes the received news articles to an internal generative model, which uses artificial intelligence algorithms to analyze the article's content.

[0864] Step 4:

[0865] A generative model in the server evaluates each sentence in the article based on its context, identifying any parts that may contain exaggeration or bias.

[0866] Step 5:

[0867] The server extracts the exaggerated or biased parts identified by the generative model and marks them, for example, with specific tags (<exaggerated>, <biased>).

[0868] Step 6:

[0869] The server evaluates the differences between the original news article and the marked text and generates the results as a transparency report.

[0870] Step 7:

[0871] The server generates a transparency report in text format and displays it to the user.

[0872] Step 8:

[0873] The server generates and provides textual advice to the user on how to verify the truth of the information, such as by recommending a list of other reliable sources or official statistical data.

[0874] Step 9:

[0875] The server integrates the "Insight Detector" function into the news app interface, allowing users to view analysis results and advice in real time as they browse the news.

[0876] Example 1

[0877] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0878] In today's information society, amidst the vast amount of news articles and information being distributed, much of the information contains exaggeration and bias. This makes it difficult for users to obtain accurate and objective information, which can lead to incorrect decision-making. There is a need to solve this problem and provide users with highly transparent and reliable information.

[0879] 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.

[0880] In this invention, the server includes: a means for a user to input information; a means for a terminal to send the input information to the server; a means for the server to receive information; a means for the server to analyze the information using a generative AI model; a means for the server to evaluate the content of an article and identify exaggerations or biases; a means for the server to mark parts containing exaggerations or biases; a means for the server to evaluate differences from the original information and generate a transparency report; a means for the server to provide advice for verifying the truth of the information; and a means for the server to display the analysis results and advice on a user interface. This makes it easier for users to obtain reliable information and enables them to make accurate and unbiased decisions.

[0881] "User" means a human or end-user who uses the system to input information and receive analytical results and transparency reports.

[0882] A "terminal" is an electronic device that a user uses to input information and that has the function of transmitting information to a server.

[0883] A "server" is a computer system that processes information received from a terminal and analyzes it using a generative AI model, and is a device that generates and provides various reports.

[0884] A "generative AI model" is a model that uses artificial intelligence technology to analyze information such as news articles, and is an algorithm used in particular to identify exaggeration or bias in information.

[0885] "Information" is a general term for news articles, text data, etc. entered by users, and is the data to be analyzed.

[0886] "Exaggeration" refers to information that overly exaggerates reality and lacks accuracy.

[0887] "Bias" refers to information that is biased toward a particular position or perspective, and is an unfair representation.

[0888] A "Transparency Report" is a document generated by the server that clearly shows the difference between the original information and the analysis results, and aims to improve the reliability of the information for users.

[0889] "Advice" is a recommendation or instruction provided by the server to help users verify truthful information, pointing to other reliable sources or verification methods.

[0890] "User interface" refers to the screen or operating means that allows the user to visually check the analysis results and advice from the server.

[0891] The present invention is a system in which a user inputs news information, a server analyzes the information to identify exaggerations and biases, and provides a highly transparent report and specific advice to the user. Hereinafter, specific embodiments of this system will be described.

[0892] This system begins when a user inputs information through a news app and the information is sent to the server. When the user inputs the URL or text of a news article, the terminal sends this information to the server. The server passes the received news article to an internal generative AI model, which analyzes the content of the article. As the generative AI model, for example, a model using natural language processing (NLP) is used.

[0893] In the analysis of news articles, the server evaluates each sentence of the article based on the context and identifies parts containing exaggerations or biases. For example, assume there is a news article as follows:

[0894] "The company's stock price has plummeted. Experts say this is the worst event in the past 20 years."

[0895] The generative AI model marks the expression "the worst event" with a high exaggeration score. After parts with exaggerations or biases are identified, the server marks these parts using specific tags (e.g., <exaggeration>). At this time, as a specific example, it is as follows:

[0896] "The company's stock price has plummeted. Experts say this is <exaggeration>the worst event< / exaggeration> in the past 20 years."

[0897] Subsequently, the server evaluates the difference between the original news article and the marked article and generates the result as a transparency report. The transparency report clearly shows which parts contain exaggerations or biases and is a document for improving the reliability of information for the user. As a specific example, it is as follows:

[0898] Original news article: "The company's stock price has plummeted, with experts calling it the worst in 20 years."

[0899] Transparency Report: "'The worst thing that could have happened' is an exaggeration. Please verify with another source."

[0900] Additionally, the server generates specific advice to help users verify the truth of the information, such as recommending other reliable sources or checking official statistics. Examples of advice provided to users include:

[0901] "This news story contains exaggerated statements. Please check with the following independent sources:

[0902] 1. Official statistical data

[0903] 2. Other trusted news sites

[0904] 3. Detailed analysis by experts

[0905] Finally, the server will integrate the "Insight Detector" feature into the news app interface, allowing users to click a specific button while browsing the news to receive real-time analysis and advice.

[0906] In this way, the present invention provides users with highly transparent information and supports honest and unbiased decision making.

[0907] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0908] Step 1:

[0909] The user opens a news app, which is ready to input information. The user enters the URL or text of the news article they want to read. The news app passes the information to the device.

[0910] Input: News article URL or text

[0911] Output: News article information sent to device

[0912] Step 2:

[0913] The terminal sends the entered news article information to the server. Specifically, the terminal sends the information to the server via an HTTP request.

[0914] Input: News article URL or text (on device)

[0915] Output: News article information sent to the server

[0916] Step 3:

[0917] A server receives news article information, which it prepares for input into an analytical model.

[0918] Input: News article information sent from the device

[0919] Output: News article information formatted for analysis

[0920] Step 4:

[0921] The server inputs the news article into the generative AI model, which then makes an API request to the model to begin analysis.

[0922] Input: News article information formatted for analysis

[0923] Output: News article information fed into the generative AI model

[0924] Step 5:

[0925] The generative AI model evaluates each sentence in a news article based on its context and identifies parts that may contain exaggeration or bias. Specifically, the model analyzes the text using natural language processing techniques.

[0926] Input: News article information

[0927] Output: Identification of exaggeration or bias

[0928] Step 6:

[0929] The server extracts exaggerated or highly biased parts from the output of the generative AI model, and marks the identified parts with a corresponding tag (e.g., <exaggerated>).

[0930] Input: Results identifying exaggeration or bias

[0931] Output: Tagged news article information

[0932] Step 7:

[0933] The server evaluates the differences between the original news article and the marked sentences, making it clear which parts contain exaggeration or bias.

[0934] Input: tagged news article information

[0935] Output: The result of the difference evaluation.

[0936] Step 8:

[0937] The server generates a transparency report based on the results of the difference evaluation, which is output in a format that is easy for users to understand.

[0938] Input: Result of the difference evaluation

[0939] Output: Transparency Report

[0940] Step 9:

[0941] The server generates specific advice to provide to the user, including a recommendation to refer to a trusted source.

[0942] Input: Difference Assessment Results and Transparency Report

[0943] Output: Specific advice

[0944] Step 10:

[0945] The server displays the analysis results and advice on the news app interface. Specifically, when the user clicks a specific button, the analysis results and advice are displayed in real time.

[0946] Input: Transparency reports and specific advice

[0947] Output: Analysis results and advice displayed in the user interface

[0948] Through this series of processing steps, users can obtain highly transparent information, which can support honest and unbiased decision-making.

[0949] (Application example 1)

[0950] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0951] In today's world, there is an increasing risk that users will be misled by exaggerations and bias in advertisements and news articles. As a result, users may make decisions based on inaccurate information. In particular, in an information-overloaded environment, unreliable information is easily mixed in, making it difficult for users to discern accurate information. The objective of this invention is to automatically identify exaggerations and biases in advertisement content and news articles, and provide users with transparent reports and specific advice to support honest and unbiased decision-making.

[0952] 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.

[0953] In this invention, the server includes: a means for a user to input information; a means for receiving information; a means for analyzing the information using a generative model; a means for marking parts containing exaggeration or bias; a means for evaluating differences from the original information; a means for generating a transparency report for the user; a means for providing advice for verifying the truth of the information; and a means for acquiring and analyzing advertising content, identifying exaggeration or bias, and providing a transparency report and specific advice. This allows the reliability of advertisements and news articles to be automatically evaluated, enabling users to make honest and unbiased decisions.

[0954] A "user" is someone who uses the system to input or receive information.

[0955] "Information" refers to news articles, advertising content, and other text data and content in general that are subject to analysis.

[0956] A "server" is a computer system that receives, analyzes, and evaluates information sent by users and returns the results.

[0957] A "generative model" refers to an algorithm or data model that uses artificial intelligence to analyze data.

[0958] "Exaggeration" refers to parts that are exaggerated in a way that is greater than the actual content.

[0959] "Bias" refers to a part that contains a particular perspective or preconception.

[0960] "Marking" refers to the act of identifying exaggerations or biases and explicitly labeling them.

[0961] "Differential evaluation" is the process of evaluating the differences between the original information and the analyzed information.

[0962] A "Transparency Report" is a report that identifies areas that contain exaggeration or bias and presents them in a user-friendly format.

[0963] "Advice" means specific instructions or suggestions provided to a user to help them make a good faith decision.

[0964] "Advertising Content" refers to promotional text and multimedia content relating to products and services.

[0965] "Analysis" is the process of evaluating and examining user-provided information for a specific purpose.

[0966] The system embodying this invention analyzes input information from users, identifies exaggeration and bias, generates transparency reports, and provides specific advice. Specific steps for implementing this invention are described below.

[0967] Configuration and operation explanation

[0968] Hardware configuration:

[0969] User devices: Mobile devices such as smartphones and tablets

[0970] Server: A high-performance computer or cloud server

[0971] Network: Internet

[0972] Software configuration:

[0973] Generative AI models: Transformer-based AI model libraries (e.g., Hugging Face Transformers)

[0974] Analysis program: Python script

[0975] News app: an interface where users enter information

[0976] A user accesses a news app or ad-checking app and enters the information about the ad or news article they want to analyze. The entered information is sent from the user's device to a server. The server then supplies the received information to a generative AI model, which analyzes it for exaggeration and bias.

[0977] The server uses a generative AI model to analyze each piece of an article or ad, taking into account the context and assessing whether a particular phrase or expression is exaggerated or biased.

[0978] If exaggeration or bias is identified, the server will flag these and evaluate the deviation from the original information. For example, if a news article mentions "the best product on the market," it will be deemed exaggerated.

[0979] Additionally, the server generates a transparency report that explicitly shows users where there is exaggeration or bias, detailing the alleged exaggeration or bias and explaining why it is problematic.

[0980] Finally, the server provides specific advice to the user, such as recommending verification from a trusted source.

[0981] Specific examples

[0982] Example: A user enters the following ad text in a news app: "The miracle diet! Lose 10kg in just 10 days."

[0983] The server analyzes this text using a generative AI model and identifies phrases like "miracle" and "lose 10kg in just 10 days" as exaggerations. It then generates a transparency report explaining to the user how these phrases are exaggerated. It also provides the user with specific advice: "This information is exaggerated. Please check other reliable sources."

[0984] Example prompt sentence:

[0985] "Below is the content of the advertisement. Please analyze this content for exaggeration or bias and provide a transparent report.

[0986] Advertisement: Miracle diet method! Lose 10kg in just 10 days.

[0987] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0988] Program processing flow and detailed explanation of each step

[0989] Step 1:

[0990] The user enters the information.

[0991] The user opens a news app or ad check app and enters the text of the ad or news article to be analyzed, or specifies a URL. This is the input information. The input information is sent from the user's device to the server.

[0992] Step 2:

[0993] The server receives the information.

[0994] The server receives the information sent by the user. The received information is then directly supplied to the generative AI model for analysis. Through this receiving operation, the server obtains input data (news articles and advertising text).

[0995] Step 3:

[0996] The server analyzes the information using a generative AI model.

[0997] The server inputs the received information into a generative AI model (e.g., Hugging Face Transformers) for analysis. Specifically, the generative AI model evaluates each piece of text and determines whether it contains exaggeration or bias. The input data is processed and an analysis result is generated.

[0998] Step 4:

[0999] The server marks the parts that contain exaggeration or bias.

[1000] The server marks parts of the original text that are identified as exaggerated or biased based on the analysis results obtained from the generative AI model. The marking uses tags (e.g., <exaggeration>) to clearly indicate which parts of the original text are exaggerated. The marked text is then output.

[1001] Step 5:

[1002] The server evaluates the difference from the original information.

[1003] The server evaluates the differences between the original information and the marked information, and sorts out which parts have been exaggerated and how. At this stage, the difference evaluation results are generated.

[1004] Step 6:

[1005] The server generates a transparency report.

[1006] The server generates a transparency report based on the difference evaluation results, detailing any exaggerations or biases and explaining why. The report is then output as feedback to the user.

[1007] Step 7:

[1008] The server provides advice to verify the truth of the information.

[1009] Based on the transparency report, the server generates specific advice for users to verify accurate information, including recommendations for reliable sources, and provides this advice to users along with the report.

[1010] For example, if a user enters the ad copy "Miracle Diet! Lose 10kg in just 10 days," the server receives it and analyzes it using a generative AI model. It identifies and marks the exaggerated parts "miracle" and "Lose 10kg in just 10 days." It then evaluates the difference between the original text and the marked text, and generates a transparency report with specific advice, such as "This text contains exaggerated parts. Please check other reliable sources," and provides it to the user.

[1011] 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.

[1012] The present invention is a system that identifies exaggeration and bias in the information entered by a user and provides a transparency report and advice that takes into account the user's feelings. The program processing of the system will be described in detail below.

[1013] Get news articles

[1014] When a user opens a news app and enters the URL or text of a news article, or selects an article within the news app, the device sends this to the server.

[1015] News article analysis

[1016] The server receives news articles and passes them to an internal generative model, which uses artificial intelligence algorithms to analyze the content of the article, evaluate the context of the sentences, and identify any parts that may contain exaggeration or bias.

[1017] Examples:

[1018] News article text:

[1019] "The company's stock price has plummeted. Experts say this is the worst event in the last 20 years."

[1020] The AI ​​model marks the "worst case scenario" as having a high exaggeration score.

[1021] Extracting exaggerated and biased sentences

[1022] The server extracts the exaggerated or biased parts identified by the generative model and marks them, explicitly indicating them to the user with specific tags (<exaggerated>, <biased>).

[1023] Emotion Analysis

[1024] The server uses an emotion engine to analyze emotions based on the text entered by the user and the selections made by the user, and this emotion analysis identifies the emotional state (e.g., anger, anxiety, joy, etc.) that the user is in when reading a news article.

[1025] Examples:

[1026] When a user reads a news article, the emotion engine detects "anxiety" from the expression in the text input.

[1027] Evaluating differences and providing transparency

[1028] The server evaluates the differences between the original news article and the marked text and generates a transparency report that clearly shows which parts contain exaggerations or bias, in a user-friendly format.

[1029] Examples:

[1030] Original news article: The company's stock price has plummeted, in what experts say is the worst it's seen in 20 years.

[1031] Transparency Report: "The worst thing that could happen" is an exaggeration. Please check with another source.

[1032] Providing advice

[1033] The server generates specific advice for the user to confirm the truth of the information. This advice reflects the analysis results of the emotion engine and provides appropriate instructions according to the user's emotional state.

[1034] Examples:

[1035] If users are feeling "uneasy," they should be warned: "This news contains exaggerated statements. Please check the following reliable sources and try not to worry too much:

[1036] 1. Official statistical data

[1037] 2. Other trusted news sites

[1038] 3. In-depth analysis by experts

[1039] Integration into news apps

[1040] Sarva will integrate the "Insight Detector" function into the news app interface. This will allow users to click a specific button when viewing news to receive analysis results and advice in real time. By providing highly transparent information that takes user emotions into consideration, Sarva will support honest and unbiased decision-making.

[1041] Thus, the present invention is a system that takes into account the user's emotional state while providing highly transparent information without exaggeration or bias.

[1042] The processing flow will be explained below.

[1043] Step 1:

[1044] A user opens a news app and enters the URL or text of a news article, or selects an article within the news app.

[1045] Step 2:

[1046] The device sends the URL or text entered by the user to the server.

[1047] Step 3:

[1048] The server passes the received news articles to an internal generative model, which uses artificial intelligence algorithms to analyze the article's content.

[1049] Step 4:

[1050] A generative model in the server evaluates each sentence in the article based on its context, identifying any parts that may contain exaggeration or bias.

[1051] Step 5:

[1052] The server extracts the exaggerated or biased parts identified by the generative model and marks them, for example, with specific tags (<exaggerated>, <biased>).

[1053] Step 6:

[1054] The server activates an emotion engine to recognize the user's emotions while the user is reading the article.

[1055] Step 7:

[1056] The emotion engine in the server analyzes the user's current emotional state (anger, anxiety, joy, etc.) based on the user's text input and selection operations.

[1057] Examples:

[1058] When a user reads a news article, the emotion engine recognizes that they are feeling "anxiety."

[1059] Step 8:

[1060] The server evaluates the differences between the original news article and the marked text and generates a transparency report that clearly shows which parts contain exaggerations or bias.

[1061] Examples:

[1062] Original news article: The company's stock price has plummeted, in what experts say is the worst it's seen in 20 years.

[1063] Transparency Report: "The worst thing that could happen" is an exaggeration. Please check with another source.

[1064] Step 9:

[1065] The server generates appropriate advice for the user based on the results of the analysis by the emotion engine. For example, if the user is feeling anxious, the server will provide advice to calm the user.

[1066] Examples:

[1067] "This news contains exaggerated statements. Please check the following reliable sources and don't worry too much:

[1068] 1. Official statistical data

[1069] 2. Other trusted news sites

[1070] 3. Detailed analysis by experts

[1071] Step 10:

[1072] The server integrates the "Insight Detector" function into the news app interface, allowing users to view analysis results and advice in real time as they browse the news.

[1073] In this way, the server takes into account the user's emotional state and provides transparent information without exaggeration or bias, thereby supporting honest and unbiased decision-making.

[1074] Example 2

[1075] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1076] In recent years, the amount of information on the Internet has increased dramatically, and the news and articles users encounter often contain exaggerations and bias. As a result, users are at greater risk of receiving incorrect information or holding biased viewpoints. Furthermore, when reading news and articles, users' emotions can significantly influence how they perceive the information, making it difficult for users to objectively evaluate it. In such a situation, there is a need for accurate and transparent information provision.

[1077] 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.

[1078] In this invention, the server includes a means for a user to input information, a means for a terminal to send information to the server, a means for the server to receive information, a means for the server to analyze the information using a generative model, a means for the server to mark parts containing exaggeration or bias, a means for the server to analyze the user's emotions, a means for the server to evaluate differences from the original information, a means for the server to generate a transparency report for the user, and a means for the server to provide advice for verifying the truth of information. This allows users to recognize exaggeration and bias when reading news or articles, understand their own emotional state, and obtain more accurate and transparent information.

[1079] "User" refers to a person who uses this system to input information and view news and articles.

[1080] "Terminal" refers to a device used by a user that provides a means for transmitting information to a server.

[1081] "Server" refers to the main computing system that receives and analyzes information sent from users and terminals.

[1082] A "generative model" refers to an algorithm or system that uses artificial intelligence to analyze and evaluate documents.

[1083] "Exaggeration or bias" refers to exaggerated or biased viewpoints in texts or news articles.

[1084] "Marking" refers to the act of explicitly indicating exaggerated or biased parts using specific tags or highlighting.

[1085] "Emotion analysis" refers to the process of identifying a user's emotional state (anger, anxiety, joy, etc.) based on the user's input text and actions.

[1086] "Differential assessment" refers to assessing the difference between the original information and the information that has been marked with exaggeration or bias.

[1087] A "transparency report" refers to a report that clearly shows users which parts contain exaggeration or bias.

[1088] "Advice" refers to the act of providing specific instructions or recommendations for users to verify truthful information.

[1089] The present invention provides a system that identifies exaggerations and biases in information entered by a user and provides transparency reports and advice that take the user's emotions into account. This system is configured by combining a user, a terminal, a server, and a generative model.

[1090] Get news articles

[1091] The user opens the news app and enters the URL or text of a news article, or can select an article within the news app, and the device then sends the information to the server.

[1092] News article analysis

[1093] The server passes the received news article to an internal generative model (e.g., GPT-4), which uses natural language processing techniques to analyze the article's content, specifically evaluating the context of each sentence and identifying parts that may contain exaggeration or bias.

[1094] Examples:

[1095] News article text:

[1096] "The company's stock price has plummeted. Experts say this is the worst event in the last 20 years."

[1097] The AI ​​model marks the "worst case scenario" as having a high exaggeration score.

[1098] Extracting exaggerated and biased sentences

[1099] The server extracts the exaggerated or biased parts identified by the generative model and marks them explicitly, using tags such as "<exaggerated>" and "<biased>".

[1100] Emotion Analysis

[1101] The server uses an emotion engine to analyze emotions based on the text entered by the user and the selections made, thereby identifying the emotional state (e.g., anger, anxiety, joy) the user is in when reading a news article.

[1102] Examples:

[1103] When a user reads a news article, the emotion engine detects "anxiety" from the expression in the text input.

[1104] Evaluating differences and providing transparency

[1105] The server evaluates the differences between the original news article and the marked text and generates a transparency report that clearly shows users which parts contain exaggerations or bias, in an easy-to-understand format.

[1106] Examples:

[1107] Original news article: "The company's stock price has plummeted, with experts saying it's the worst in 20 years."

[1108] Transparency Report: "The worst thing that could happen" is an exaggeration. Please check with another source.

[1109] Providing advice

[1110] The server generates specific advice for the user to confirm the truth of the information. This advice reflects the analysis results of the emotion engine and provides appropriate instructions according to the user's emotional state.

[1111] Examples:

[1112] If users are feeling "uneasy," they should be advised that "This news contains exaggerated statements. Please check the following reliable sources and do not worry too much:

[1113] 1. Official statistical data

[1114] 2. Other trusted news sites

[1115] 3. In-depth analysis by experts

[1116] Integration into news apps

[1117] Sarva will integrate the "Insight Detector" feature into the news app interface. This will allow users to click a specific button while browsing the news to receive analysis results and advice in real time. This feature will enable users to make honest and unbiased decisions while receiving transparent information that takes emotions into account.

[1118] This system takes into account the user's emotions and provides highly transparent information without exaggeration or bias.

[1119] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1120] Step 1:

[1121] The user launches a news app and enters the URL or text of a news article, or selects an article within the news app. The device then sends this to the server.

[1122] Input: News article URL or text entered by the user.

[1123] Output: The URL or text of the news article entered is sent to the server.

[1124] Specific behavior:

[1125] When a user pastes a URL into the search bar of a news app or taps on a specific article from the "Recent News" section, the device instantly sends it to the server.

[1126] Step 2:

[1127] The server passes the received news article to a generative model, which uses natural language processing techniques to analyze the article's content.

[1128] Input: News article text data.

[1129] Output: Contextual data analyzed using natural language processing techniques.

[1130] Specific behavior:

[1131] The server feeds the text of a news article into a generative model, which evaluates the context of each sentence to identify potential exaggerations or biases.

[1132] Step 3:

[1133] The server extracts the exaggerated or biased parts identified by the generative model and marks them to indicate this explicitly.

[1134] Input: Analyzed contextual data, identifying exaggerations and biases.

[1135] Output: Marked text data.

[1136] Specific behavior:

[1137] The server tags the identified phrases with tags such as "<exaggeration>" or "<bias>." For example, the phrase "the worst thing that could happen" in a sentence might be tagged with "<exaggeration>."

[1138] Step 4:

[1139] The server uses an emotion engine to analyze emotions based on the user's text input and selections.

[1140] Input: User input text and operation data.

[1141] Output: Sentiment analysis results (e.g., anxiety, anger, joy, etc.).

[1142] Specific behavior:

[1143] The emotion engine analyzes comments and reactions when users read news articles and identifies emotions such as anxiety and vigilance.

[1144] Step 5:

[1145] The server evaluates the differences between the original news article and the marked text and generates the results as a transparency report.

[1146] Input: Original news article, marked text data.

[1147] Output: Transparency report.

[1148] Specific behavior:

[1149] The server compares the original sentence with the marked sentence and compiles a report that the expression "this is the worst thing that could happen" is an exaggeration.

[1150] Step 6:

[1151] The server generates advice to help users confirm the truth of information, reflecting the analysis results of the emotion engine and providing appropriate instructions according to the user's emotional state.

[1152] Input: Sentiment analysis results.

[1153] Output: Text data of advice.

[1154] Specific behavior:

[1155] If the user feels "anxious," the server generates advice such as, "This news contains exaggerated statements. Please check the reliable sources below and try not to worry too much," and provides links to reliable sources.

[1156] Step 7:

[1157] The server integrates the "Insight Detector" function into the news app interface, allowing users to receive analysis results and advice in real time by clicking a specific button.

[1158] Input: User action (clicking a button).

[1159] Output: Display data of analysis results and advice.

[1160] Specific behavior:

[1161] When a user clicks the "Insight Detector" button in a news app, the analysis results and advice received from the server are immediately displayed. This feature allows users to obtain information while being aware of exaggeration and bias when reading the news.

[1162] (Application example 2)

[1163] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1164] Current information provision systems have difficulty detecting exaggerations and biases in the information users are exposed to and providing it in a transparent manner. As a result, users often make decisions based on incorrect information and believe unreliable information. Furthermore, because systems do not take the user's emotional state into account, the advice provided may not be appropriate for the user's psychological state. A comprehensive system is needed to solve these problems.

[1165] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for the user to input information, a means for the server to receive information, a means for the server to analyze the information using a generative model, a means for the server to mark parts containing exaggeration or bias, a means for the server to evaluate differences from the original information, a means for the server to generate a transparency report for the user, a means for the server to analyze the emotional state, and a means for the server to provide advice based on the emotional state. This makes it possible to detect exaggeration or bias in the information the user comes into contact with in real time and provide it in a highly transparent manner, and also to provide appropriate advice according to the user's emotional state.

[1166] A "user" is an entity that inputs information and receives feedback from the system.

[1167] "Information" is text data such as news and advertisements entered by the user.

[1168] A "server" is a computer system that analyzes information received from a user and performs the necessary processing.

[1169] A "generative model" is software that uses artificial intelligence algorithms to analyze information.

[1170] "Analysis" is the process of using generative models to evaluate information and identify its properties.

[1171] "Exaggeration" is when a piece of information is exaggerated in a way that is greater than the actual content.

[1172] "Bias" is a state in which some information is influenced by a particular perspective or prejudice.

[1173] "Marking" is tagging to explicitly indicate exaggerated or biased parts.

[1174] "Difference" refers to the difference between the original information and the analyzed information.

[1175] A "Transparency Report" is a report that clearly shows any exaggeration or bias in the information for users.

[1176] An "emotional state" is the psychological state a user feels when viewing information.

[1177] "Advice" is a set of instructions or suggestions that help users verify information without exaggeration or bias.

[1178] The present invention is a system that detects exaggeration and bias in the information a user comes into contact with in real time, provides the information in a highly transparent manner, and provides advice that takes into account the user's emotional state. Specific embodiments for realizing this system are described below.

[1179] Program processing overview

[1180] The system is implemented as an "Ad-Transparency Detector" application installed on smartphones or smart glasses. When a user views an ad, it analyzes the ad text and marks any exaggerated or biased parts, displaying them in real time. It also analyzes the user's emotions based on their browsing behavior and comments, and provides advice based on those emotions.

[1181] Hardware and Software

[1182] Hardware:

[1183] Smartphone or smart glasses: The device through which the user views the advertisement.

[1184] Server: A central management system that analyzes information and runs generative models.

[1185] software:

[1186] Generative models: Artificial intelligence algorithms that analyze information and identify exaggeration and bias.

[1187] Natural Language Processing Library (nltk): A library for analyzing the emotional state of the user.

[1188] Data Acquisition Library (requests): A library for acquiring text data for advertisements.

[1189] Specific examples

[1190] A user launches the "Ad-Transparency Detector" application and enters the URL of an ad. The ad includes a claim that "This product has achieved record sales!" The system's server receives the ad text and analyzes it using a generative model. The generative model determines that the "record sales" part is exaggerated and marks it.

[1191] The server then analyzes the user's input comments and determines that the user is feeling "anxious." Based on the user's emotional state, the server provides advice such as, "This ad contains exaggerated statements. Please also check the following reliable sources: official website, third-party review sites, etc. Don't worry too much."

[1192] Example of a prompt to be entered

[1193] There is an advertisement that claims, "This product has achieved record sales!" Please evaluate whether this is an exaggeration and explain why. Furthermore, if a user viewing this advertisement feels "anxious," what advice would you give them?

[1194] In this way, users can easily assess the trustworthiness of the advertisements they come into contact with and make appropriate decisions.

[1195] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1196] Step 1:

[1197] The user enters the URL of an ad. The user launches the "Ad-Transparency Detector" application and enters or pastes the URL of the ad. At this time, the URL entered by the user is sent from the device to the server. The input data is the URL of the ad, and the output data is the URL sent to the server.

[1198] Step 2:

[1199] The server retrieves the ad text. Using the received URL, the server uses a data retrieval library (requests) to retrieve the HTML content of the corresponding ad page. The input data is the ad URL, and the output data is the retrieved ad text. Specifically, an HTTP request is sent and the server downloads the HTML source of the ad page.

[1200] Step 3:

[1201] The server analyzes the ad text. The acquired ad text is passed to a generative model, which analyzes the information. The generative model uses an artificial intelligence algorithm to evaluate the ad text and identify parts that contain exaggeration or bias. The input data is the ad text, and the output data is the analysis results. Specifically, natural language processing technology is used to evaluate the context.

[1202] Step 4:

[1203] The server marks exaggerated or biased parts. Parts with high levels of exaggeration or bias identified by the generative model are tagged with specific tags (e.g., <exaggerated>, <biased>) to make them easy for users to understand. The input data is the analysis results, and the output data is the marked advertising text. Specifically, tags are inserted into the relevant parts of the text.

[1204] Step 5:

[1205] The server analyzes the user's emotional state. It uses a natural language processing library (nltk) to analyze the emotional state based on the comments and feedback entered by the user when viewing the ad text. The input data is the user's comments, and the output data is the emotional state (e.g., anxiety, joy). Specifically, it applies a sentiment analysis algorithm to calculate an emotional score.

[1206] Step 6:

[1207] The server evaluates the differences from the original information. It compares the original ad text with the marked ad text and generates a transparency report to indicate to the user any exaggerations or biases. The input data is the original ad text and the marked ad text, and the output data is a transparency report. Specifically, it uses a text comparison algorithm to extract the differences.

[1208] Step 7:

[1209] The server provides advice based on the emotional state. Advice reflecting the emotional state is generated and presented to the user. The input data is the emotional state and a transparency report, and the output data is advice. Specifically, an appropriate message is generated according to the emotional state and presented to the user as feedback.

[1210] Step 8:

[1211] The server sends the transparency report and advice to the terminal. The generated transparency report and advice are sent to the device used by the user (smartphone or smart glasses) and displayed in real time. The input data are the transparency report and advice, and the output data is the information displayed on the user's terminal. Specifically, the data is sent via network communication.

[1212] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1213] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1214] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1215] [Fourth embodiment]

[1216] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1217] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1218] 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).

[1219] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1220] 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.

[1221] 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).

[1222] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1223] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1224] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1225] 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.

[1226] 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.

[1227] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1228] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1229] The present invention is a system in which users input information, a server analyzes the information, identifies exaggerations and biases, and provides users with transparent reports and advice. The following describes the program processing of this system in detail.

[1230] Get news articles

[1231] The user opens the news app, enters the URL or text of the news article they want to read, and when the user selects an article within the news app, the device sends it to the server.

[1232] News article analysis

[1233] The server passes the received news article to an internal generative model, which uses artificial intelligence to analyze the article's content by evaluating each sentence in context and identifying any parts that may contain exaggeration or bias.

[1234] Examples:

[1235] News article text:

[1236] "The company's stock price has plummeted. Experts say this is the worst event in the last 20 years."

[1237] The AI ​​model marks the "worst case scenario" as having a high exaggeration score.

[1238] Extracting exaggerated and biased sentences

[1239] The server extracts exaggerated or highly biased parts identified by the generative model and marks them, explicitly indicating them to the user with a specific tag (e.g., <exaggerated>).

[1240] Evaluating differences and providing transparency

[1241] The server evaluates the differences between the original news article and the marked text and generates a transparency report that clearly shows which parts contain exaggerations or bias, in a format that is easy for users to understand.

[1242] Examples:

[1243] Original news article: The company's stock price has plummeted, in what experts say is the worst it's seen in 20 years.

[1244] Transparency Report: "The worst thing that could happen" is an exaggeration. Please check with another source.

[1245] Providing advice

[1246] The server generates specific advice to help users verify the truth of the information, for example, recommending reference to other reliable sources or encouraging users to check official statistics.

[1247] Examples:

[1248] This news story contains exaggerated statements. Please verify with the following independent sources:

[1249] 1. Official statistical data

[1250] 2. Other trusted news sites

[1251] 3. In-depth analysis by experts

[1252] Integration into news apps

[1253] The server will integrate the "Insight Detector" function into the news app interface, allowing users to click a specific button while browsing the news to receive analysis results and advice in real time.

[1254] In this way, the present invention provides users with highly transparent information and supports honest and unbiased decision making.

[1255] The processing flow will be explained below.

[1256] Step 1:

[1257] A user opens a news app and enters the URL or text of a news article, or selects an article within the news app.

[1258] Step 2:

[1259] The device sends the URL or text entered by the user to the server.

[1260] Step 3:

[1261] The server passes the received news articles to an internal generative model, which uses artificial intelligence algorithms to analyze the article's content.

[1262] Step 4:

[1263] A generative model in the server evaluates each sentence in the article based on its context, identifying any parts that may contain exaggeration or bias.

[1264] Step 5:

[1265] The server extracts the exaggerated or biased parts identified by the generative model and marks them, for example, with specific tags (<exaggerated>, <biased>).

[1266] Step 6:

[1267] The server evaluates the differences between the original news article and the marked text and generates the results as a transparency report.

[1268] Step 7:

[1269] The server generates a transparency report in text format and displays it to the user.

[1270] Step 8:

[1271] The server generates and provides textual advice to the user on how to verify the truth of the information, such as by recommending a list of other reliable sources or official statistical data.

[1272] Step 9:

[1273] The server integrates the "Insight Detector" function into the news app interface, allowing users to view analysis results and advice in real time as they browse the news.

[1274] Example 1

[1275] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1276] In today's information society, amidst the vast amount of news articles and information being distributed, much of the information contains exaggeration and bias. This makes it difficult for users to obtain accurate and objective information, which can lead to incorrect decision-making. There is a need to solve this problem and provide users with highly transparent and reliable information.

[1277] 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.

[1278] In this invention, the server includes: a means for a user to input information; a means for a terminal to send the input information to the server; a means for the server to receive information; a means for the server to analyze the information using a generative AI model; a means for the server to evaluate the content of an article and identify exaggerations or biases; a means for the server to mark parts containing exaggerations or biases; a means for the server to evaluate differences from the original information and generate a transparency report; a means for the server to provide advice for verifying the truth of the information; and a means for the server to display the analysis results and advice on a user interface. This makes it easier for users to obtain reliable information and enables them to make accurate and unbiased decisions.

[1279] "User" means a human or end-user who uses the system to input information and receive analytical results and transparency reports.

[1280] A "terminal" is an electronic device that a user uses to input information and that has the function of transmitting information to a server.

[1281] A "server" is a computer system that processes information received from a terminal and analyzes it using a generative AI model, and is a device that generates and provides various reports.

[1282] A "generative AI model" is a model that uses artificial intelligence technology to analyze information such as news articles, and is an algorithm used in particular to identify exaggeration or bias in information.

[1283] "Information" is a general term for news articles, text data, etc. entered by users, and is the data to be analyzed.

[1284] "Exaggeration" refers to information that overly exaggerates reality and lacks accuracy.

[1285] "Bias" refers to information that is biased toward a particular position or perspective, and is an unfair representation.

[1286] A "Transparency Report" is a document generated by the server that clearly shows the difference between the original information and the analysis results, and aims to improve the reliability of the information for users.

[1287] "Advice" is a recommendation or instruction provided by the server to help users verify truthful information, pointing to other reliable sources or verification methods.

[1288] "User interface" refers to the screen or operating means that allows the user to visually check the analysis results and advice from the server.

[1289] The present invention is a system in which a user inputs news information, a server analyzes the information, identifies exaggerations and bias, and provides the user with a highly transparent report and specific advice. Specific embodiments of this system are described below.

[1290] This system begins when a user inputs information through a news app and sends it to a server. When a user inputs the URL or text of a news article, the device sends this information to the server. The server passes the received news article to an internal generative AI model, which then analyzes the content of the article. For example, a model using natural language processing (NLP) is used as the generative AI model.

[1291] When analyzing a news article, the server evaluates each sentence in the article based on its context to identify exaggerations or bias. For example, consider the following news article:

[1292] "The company's stock price has plummeted. Experts say this is the worst thing that has happened in the last 20 years."

[1293] The generated AI model marks the expression "the worst-case scenario" with a high exaggeration score. After identifying the exaggerated and biased parts, the server marks these parts using specific tags (e.g., <exaggeration>). At this time, a specific example is as follows:

[1294] "The company's stock price has plummeted. Experts say this is the <exaggeration>worst-case scenario in the past 20 years< / exaggeration>."

[1295] Subsequently, the server evaluates the difference between the original news article and the marked text and generates the result as a transparency report. The transparency report clearly shows which parts contain exaggeration or bias and is a document to improve the reliability of information for users. A specific example is as follows:

[1296] Original news article: "The company's stock price has plummeted. Experts say this is the worst-case scenario in the past 20 years."

[1297] Transparency report: "The expression 'the worst-case scenario' is exaggerated. Please verify from another information source."

[1298] Furthermore, the server generates specific advice for users to verify the true information. For example, it recommends referring to other reliable information sources or encourages checking public statistical data. Examples of advice provided to users include the following content:

[1299] "This news contains exaggerated expressions. Please verify using the following independent information sources:

[1300] 1. Public statistical data

[1301] 2. Other reliable news sites

[1302] 3. Detailed analysis articles by experts"

[1303] Finally, the server will integrate the "Insight Detector" feature into the news app interface, allowing users to click a specific button while browsing the news to receive real-time analysis and advice.

[1304] In this way, the present invention provides users with highly transparent information and supports honest and unbiased decision making.

[1305] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1306] Step 1:

[1307] The user opens a news app, which is ready to input information. The user enters the URL or text of the news article they want to read. The news app passes the information to the device.

[1308] Input: News article URL or text

[1309] Output: News article information sent to device

[1310] Step 2:

[1311] The terminal sends the entered news article information to the server. Specifically, the terminal sends the information to the server via an HTTP request.

[1312] Input: News article URL or text (on device)

[1313] Output: News article information sent to the server

[1314] Step 3:

[1315] A server receives news article information, which it prepares for input into an analytical model.

[1316] Input: News article information sent from the device

[1317] Output: News article information formatted for analysis

[1318] Step 4:

[1319] The server inputs the news article into the generative AI model, which then makes an API request to the model to begin analysis.

[1320] Input: News article information formatted for analysis

[1321] Output: News article information fed into the generative AI model

[1322] Step 5:

[1323] The generative AI model evaluates each sentence in a news article based on its context and identifies parts that may contain exaggeration or bias. Specifically, the model analyzes the text using natural language processing techniques.

[1324] Input: News article information

[1325] Output: Identification of exaggeration or bias

[1326] Step 6:

[1327] The server extracts exaggerated or highly biased parts from the output of the generative AI model, and marks the identified parts with a corresponding tag (e.g., <exaggerated>).

[1328] Input: Results identifying exaggeration or bias

[1329] Output: Tagged news article information

[1330] Step 7:

[1331] The server evaluates the differences between the original news article and the marked sentences, making it clear which parts contain exaggeration or bias.

[1332] Input: tagged news article information

[1333] Output: The result of the difference evaluation.

[1334] Step 8:

[1335] The server generates a transparency report based on the results of the difference evaluation, which is output in a format that is easy for users to understand.

[1336] Input: Result of the difference evaluation

[1337] Output: Transparency Report

[1338] Step 9:

[1339] The server generates specific advice to provide to the user, including a recommendation to refer to a trusted source.

[1340] Input: Difference Assessment Results and Transparency Report

[1341] Output: Specific advice

[1342] Step 10:

[1343] The server displays the analysis results and advice on the news app interface. Specifically, when the user clicks a specific button, the analysis results and advice are displayed in real time.

[1344] Input: Transparency reports and specific advice

[1345] Output: Analysis results and advice displayed in the user interface

[1346] Through this series of processing steps, users can obtain highly transparent information, which can support honest and unbiased decision-making.

[1347] (Application example 1)

[1348] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1349] In today's world, there is an increasing risk that users will be misled by exaggerations and bias in advertisements and news articles. As a result, users may make decisions based on inaccurate information. In particular, in an information-overloaded environment, unreliable information is easily mixed in, making it difficult for users to discern accurate information. The objective of this invention is to automatically identify exaggerations and biases in advertisement content and news articles, and provide users with transparent reports and specific advice to support honest and unbiased decision-making.

[1350] 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.

[1351] In this invention, the server includes: a means for a user to input information; a means for receiving information; a means for analyzing the information using a generative model; a means for marking parts containing exaggeration or bias; a means for evaluating differences from the original information; a means for generating a transparency report for the user; a means for providing advice for verifying the truth of the information; and a means for acquiring and analyzing advertising content, identifying exaggeration or bias, and providing a transparency report and specific advice. This allows the reliability of advertisements and news articles to be automatically evaluated, enabling users to make honest and unbiased decisions.

[1352] A "user" is someone who uses the system to input or receive information.

[1353] "Information" refers to news articles, advertising content, and other text data and content in general that are subject to analysis.

[1354] A "server" is a computer system that receives, analyzes, and evaluates information sent by users and returns the results.

[1355] A "generative model" refers to an algorithm or data model that uses artificial intelligence to analyze data.

[1356] "Exaggeration" refers to parts that are exaggerated in a way that is greater than the actual content.

[1357] "Bias" refers to a part that contains a particular perspective or preconception.

[1358] "Marking" refers to the act of identifying exaggerations or biases and explicitly labeling them.

[1359] "Differential evaluation" is the process of evaluating the differences between the original information and the analyzed information.

[1360] A "Transparency Report" is a report that identifies areas that contain exaggeration or bias and presents them in a user-friendly format.

[1361] "Advice" means specific instructions or suggestions provided to a user to help them make a good faith decision.

[1362] "Advertising Content" refers to promotional text and multimedia content relating to products and services.

[1363] "Analysis" is the process of evaluating and examining user-provided information for a specific purpose.

[1364] The system embodying this invention analyzes input information from users, identifies exaggeration and bias, generates transparency reports, and provides specific advice. Specific steps for implementing this invention are described below.

[1365] Configuration and operation explanation

[1366] Hardware configuration:

[1367] User devices: Mobile devices such as smartphones and tablets

[1368] Server: A high-performance computer or cloud server

[1369] Network: Internet

[1370] Software configuration:

[1371] Generative AI models: Transformer-based AI model libraries (e.g., Hugging Face Transformers)

[1372] Analysis program: Python script

[1373] News app: an interface where users enter information

[1374] A user accesses a news app or ad-checking app and enters the information about the ad or news article they want to analyze. The entered information is sent from the user's device to a server. The server then supplies the received information to a generative AI model, which analyzes it for exaggeration and bias.

[1375] The server uses a generative AI model to analyze each piece of an article or ad, taking into account the context and assessing whether a particular phrase or expression is exaggerated or biased.

[1376] If exaggeration or bias is identified, the server will flag these and evaluate the deviation from the original information. For example, if a news article mentions "the best product on the market," it will be deemed exaggerated.

[1377] Additionally, the server generates a transparency report that explicitly shows users where there is exaggeration or bias, detailing the alleged exaggeration or bias and explaining why it is problematic.

[1378] Finally, the server provides specific advice to the user, such as recommending verification from a trusted source.

[1379] Specific examples

[1380] Example: A user enters the following ad text in a news app: "The miracle diet! Lose 10kg in just 10 days."

[1381] The server analyzes this text using a generative AI model and identifies phrases like "miracle" and "lose 10kg in just 10 days" as exaggerations. It then generates a transparency report explaining to the user how these phrases are exaggerated. It also provides the user with specific advice: "This information is exaggerated. Please check other reliable sources."

[1382] Example prompt sentence:

[1383] "Below is the content of the advertisement. Please analyze this content for exaggeration or bias and provide a transparent report.

[1384] Advertisement: Miracle diet method! Lose 10kg in just 10 days.

[1385] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1386] Program processing flow and detailed explanation of each step

[1387] Step 1:

[1388] The user enters the information.

[1389] The user opens a news app or ad check app and enters the text of the ad or news article to be analyzed, or specifies a URL. This is the input information. The input information is sent from the user's device to the server.

[1390] Step 2:

[1391] The server receives the information.

[1392] The server receives the information sent by the user. The received information is then directly supplied to the generative AI model for analysis. Through this receiving operation, the server obtains input data (news articles and advertising text).

[1393] Step 3:

[1394] The server analyzes the information using a generative AI model.

[1395] The server inputs the received information into a generative AI model (e.g., Hugging Face Transformers) for analysis. Specifically, the generative AI model evaluates each piece of text and determines whether it contains exaggeration or bias. The input data is processed and an analysis result is generated.

[1396] Step 4:

[1397] The server marks the parts that contain exaggeration or bias.

[1398] The server marks parts of the original text that are identified as exaggerated or biased based on the analysis results obtained from the generative AI model. The marking uses tags (e.g., <exaggeration>) to clearly indicate which parts of the original text are exaggerated. The marked text is then output.

[1399] Step 5:

[1400] The server evaluates the difference from the original information.

[1401] The server evaluates the differences between the original information and the marked information, and sorts out which parts have been exaggerated and how. At this stage, the difference evaluation results are generated.

[1402] Step 6:

[1403] The server generates a transparency report.

[1404] The server generates a transparency report based on the difference evaluation results, detailing any exaggerations or biases and explaining why. The report is then output as feedback to the user.

[1405] Step 7:

[1406] The server provides advice to verify the truth of the information.

[1407] Based on the transparency report, the server generates specific advice for users to verify accurate information, including recommendations for reliable sources, and provides this advice to users along with the report.

[1408] For example, if a user enters the ad copy "Miracle Diet! Lose 10kg in just 10 days," the server receives it and analyzes it using a generative AI model. It identifies and marks the exaggerated parts "miracle" and "Lose 10kg in just 10 days." It then evaluates the difference between the original text and the marked text, and generates a transparency report with specific advice, such as "This text contains exaggerated parts. Please check other reliable sources," and provides it to the user.

[1409] 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.

[1410] The present invention is a system that identifies exaggeration and bias in the information entered by a user and provides a transparency report and advice that takes into account the user's feelings. The program processing of the system will be described in detail below.

[1411] Get news articles

[1412] When a user opens a news app and enters the URL or text of a news article, or selects an article within the news app, the device sends this to the server.

[1413] News article analysis

[1414] The server receives news articles and passes them to an internal generative model, which uses artificial intelligence algorithms to analyze the content of the article, evaluate the context of the sentences, and identify any parts that may contain exaggeration or bias.

[1415] Examples:

[1416] News article text:

[1417] "The company's stock price has plummeted. Experts say this is the worst event in the last 20 years."

[1418] The AI ​​model marks the "worst case scenario" as having a high exaggeration score.

[1419] Extracting exaggerated and biased sentences

[1420] The server extracts the exaggerated or biased parts identified by the generative model and marks them, explicitly indicating them to the user with specific tags (<exaggerated>, <biased>).

[1421] Emotion Analysis

[1422] The server uses an emotion engine to analyze emotions based on the text entered by the user and the selections made by the user, and this emotion analysis identifies the emotional state (e.g., anger, anxiety, joy, etc.) that the user is in when reading a news article.

[1423] Examples:

[1424] When a user reads a news article, the emotion engine detects "anxiety" from the expression in the text input.

[1425] Evaluating differences and providing transparency

[1426] The server evaluates the differences between the original news article and the marked text and generates a transparency report that clearly shows which parts contain exaggeration or bias, in a user-friendly format.

[1427] Examples:

[1428] Original news article: The company's stock price has plummeted, in what experts say is the worst it's seen in 20 years.

[1429] Transparency Report: "The worst thing that could happen" is an exaggeration. Please check with another source.

[1430] Providing advice

[1431] The server generates specific advice for the user to confirm the truth of the information. This advice reflects the analysis results of the emotion engine and provides appropriate instructions according to the user's emotional state.

[1432] Examples:

[1433] If users are feeling "uneasy," they should be advised that "This news contains exaggerated statements. Please check the following reliable sources and do not worry too much:

[1434] 1. Official statistical data

[1435] 2. Other trusted news sites

[1436] 3. In-depth analysis by experts

[1437] Integration into news apps

[1438] Sarva will integrate the "Insight Detector" function into the news app interface. This will allow users to click a specific button when viewing news to receive analysis results and advice in real time. By providing highly transparent information that takes user emotions into consideration, Sarva will support honest and unbiased decision-making.

[1439] Thus, the present invention is a system that takes into account the user's emotional state while providing highly transparent information without exaggeration or bias.

[1440] The processing flow will be explained below.

[1441] Step 1:

[1442] A user opens a news app and enters the URL or text of a news article, or selects an article within the news app.

[1443] Step 2:

[1444] The device sends the URL or text entered by the user to the server.

[1445] Step 3:

[1446] The server passes the received news articles to an internal generative model, which uses artificial intelligence algorithms to analyze the article's content.

[1447] Step 4:

[1448] A generative model in the server evaluates each sentence in the article based on its context, identifying any parts that may contain exaggeration or bias.

[1449] Step 5:

[1450] The server extracts the exaggerated or biased parts identified by the generative model and marks them, for example, with specific tags (<exaggerated>, <biased>).

[1451] Step 6:

[1452] The server activates an emotion engine to recognize the user's emotions while the user is reading the article.

[1453] Step 7:

[1454] The emotion engine in the server analyzes the user's current emotional state (anger, anxiety, joy, etc.) based on the user's text input and selection operations.

[1455] Examples:

[1456] When a user reads a news article, the emotion engine recognizes that they are feeling "anxiety."

[1457] Step 8:

[1458] The server evaluates the differences between the original news article and the marked text and generates a transparency report that clearly shows which parts contain exaggerations or bias.

[1459] Examples:

[1460] Original news article: The company's stock price has plummeted, in what experts say is the worst it's seen in 20 years.

[1461] Transparency Report: "The worst thing that could happen" is an exaggeration. Please check with another source.

[1462] Step 9:

[1463] The server generates appropriate advice for the user based on the results of the analysis by the emotion engine. For example, if the user is feeling anxious, the server will provide advice to calm the user.

[1464] Examples:

[1465] "This news contains exaggerated statements. Please check the following reliable sources and don't worry too much:

[1466] 1. Official statistical data

[1467] 2. Other trusted news sites

[1468] 3. Detailed analysis by experts

[1469] Step 10:

[1470] The server integrates the "Insight Detector" function into the news app interface, allowing users to view analysis results and advice in real time as they browse the news.

[1471] In this way, the server takes into account the user's emotional state and provides transparent information without exaggeration or bias, thereby supporting honest and unbiased decision-making.

[1472] Example 2

[1473] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1474] In recent years, the amount of information on the Internet has increased dramatically, and the news and articles users encounter often contain exaggerations and bias. As a result, users are at greater risk of receiving incorrect information or holding biased viewpoints. Furthermore, when reading news and articles, users' emotions can significantly influence how they perceive the information, making it difficult for users to objectively evaluate it. In such a situation, there is a need for accurate and transparent information provision.

[1475] 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.

[1476] In this invention, the server includes a means for a user to input information, a means for a terminal to send information to the server, a means for the server to receive information, a means for the server to analyze the information using a generative model, a means for the server to mark parts containing exaggeration or bias, a means for the server to analyze the user's emotions, a means for the server to evaluate differences from the original information, a means for the server to generate a transparency report for the user, and a means for the server to provide advice for verifying the truth of information. This allows users to recognize exaggeration and bias when reading news or articles, understand their own emotional state, and obtain more accurate and transparent information.

[1477] "User" refers to a person who uses this system to input information and view news and articles.

[1478] "Terminal" refers to a device used by a user that provides a means for transmitting information to a server.

[1479] "Server" refers to the main computing system that receives and analyzes information sent from users and terminals.

[1480] A "generative model" refers to an algorithm or system that uses artificial intelligence to analyze and evaluate documents.

[1481] "Exaggeration or bias" refers to exaggerated or biased viewpoints in texts or news articles.

[1482] "Marking" refers to the act of explicitly indicating exaggerated or biased parts using specific tags or highlighting.

[1483] "Emotion analysis" refers to the process of identifying a user's emotional state (anger, anxiety, joy, etc.) based on the user's input text and actions.

[1484] "Differential assessment" refers to assessing the difference between the original information and the information that has been marked with exaggeration or bias.

[1485] A "transparency report" refers to a report that clearly shows users which parts contain exaggeration or bias.

[1486] "Advice" refers to the act of providing specific instructions or recommendations for users to verify truthful information.

[1487] The present invention provides a system that identifies exaggerations and biases in information entered by a user and provides transparency reports and advice that take the user's emotions into account. This system is configured by combining a user, a terminal, a server, and a generative model.

[1488] Get news articles

[1489] The user opens the news app and enters the URL or text of a news article, or can select an article within the news app, and the device then sends the information to the server.

[1490] News article analysis

[1491] The server passes the received news article to an internal generative model (e.g., GPT-4), which uses natural language processing techniques to analyze the article's content, specifically evaluating the context of each sentence and identifying parts that may contain exaggeration or bias.

[1492] Examples:

[1493] News article text:

[1494] "The company's stock price has plummeted. Experts say this is the worst event in the last 20 years."

[1495] The AI ​​model marks the "worst case scenario" as having a high exaggeration score.

[1496] Extracting exaggerated and biased sentences

[1497] The server extracts the exaggerated or biased parts identified by the generative model and marks them explicitly, using tags such as "<exaggerated>" and "<biased>".

[1498] Emotion Analysis

[1499] The server uses an emotion engine to analyze emotions based on the text entered by the user and the selections made, thereby identifying the emotional state (e.g., anger, anxiety, joy) the user is in when reading a news article.

[1500] Examples:

[1501] When a user reads a news article, the emotion engine detects "anxiety" from the expression in the text input.

[1502] Evaluating differences and providing transparency

[1503] The server evaluates the differences between the original news article and the marked text and generates a transparency report that clearly shows users which parts contain exaggerations or bias, in an easy-to-understand format.

[1504] Examples:

[1505] Original news article: "The company's stock price has plummeted, with experts saying it's the worst in 20 years."

[1506] Transparency Report: "The worst thing that could happen" is an exaggeration. Please check with another source.

[1507] Providing advice

[1508] The server generates specific advice for the user to confirm the truth of the information. This advice reflects the analysis results of the emotion engine and provides appropriate instructions according to the user's emotional state.

[1509] Examples:

[1510] If users are feeling "uneasy," they should be advised that "This news contains exaggerated statements. Please check the following reliable sources and do not worry too much:

[1511] 1. Official statistical data

[1512] 2. Other trusted news sites

[1513] 3. In-depth analysis by experts

[1514] Integration into news apps

[1515] Sarva will integrate the "Insight Detector" feature into the news app interface. This will allow users to click a specific button while browsing the news to receive analysis results and advice in real time. This feature will enable users to make honest and unbiased decisions while receiving transparent information that takes emotions into account.

[1516] This system takes into account the user's emotions and provides highly transparent information without exaggeration or bias.

[1517] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1518] Step 1:

[1519] The user launches a news app and enters the URL or text of a news article, or selects an article within the news app. The device then sends this to the server.

[1520] Input: News article URL or text entered by the user.

[1521] Output: The URL or text of the news article entered is sent to the server.

[1522] Specific behavior:

[1523] When a user pastes a URL into the search bar of a news app or taps on a specific article from the "Recent News" section, the device instantly sends it to the server.

[1524] Step 2:

[1525] The server passes the received news article to a generative model, which uses natural language processing techniques to analyze the article's content.

[1526] Input: News article text data.

[1527] Output: Contextual data analyzed using natural language processing techniques.

[1528] Specific behavior:

[1529] The server feeds the text of a news article into a generative model, which evaluates the context of each sentence to identify potential exaggerations or biases.

[1530] Step 3:

[1531] The server extracts the exaggerated or biased parts identified by the generative model and marks them to indicate this explicitly.

[1532] Input: Analyzed contextual data, identifying exaggerations and biases.

[1533] Output: Marked text data.

[1534] Specific behavior:

[1535] The server tags the identified phrases with tags such as "<exaggeration>" or "<bias>." For example, the phrase "the worst thing that could happen" in a sentence might be tagged with "<exaggeration>."

[1536] Step 4:

[1537] The server uses an emotion engine to analyze emotions based on the user's text input and selections.

[1538] Input: User input text and operation data.

[1539] Output: Sentiment analysis results (e.g., anxiety, anger, joy, etc.).

[1540] Specific behavior:

[1541] The emotion engine analyzes comments and reactions when users read news articles and identifies emotions such as anxiety and vigilance.

[1542] Step 5:

[1543] The server evaluates the differences between the original news article and the marked text and generates the results as a transparency report.

[1544] Input: Original news article, marked text data.

[1545] Output: Transparency report.

[1546] Specific behavior:

[1547] The server compares the original sentence with the marked sentence and compiles a report that the expression "this is the worst thing that could happen" is an exaggeration.

[1548] Step 6:

[1549] The server generates advice to help users confirm the truth of information, reflecting the analysis results of the emotion engine and providing appropriate instructions according to the user's emotional state.

[1550] Input: Sentiment analysis results.

[1551] Output: Text data of advice.

[1552] Specific behavior:

[1553] If the user feels "anxious," the server generates advice such as "This news contains exaggerated statements. Please check the reliable sources below and try not to worry too much," and provides links to reliable sources.

[1554] Step 7:

[1555] The server integrates the "Insight Detector" function into the news app interface, allowing users to receive analysis results and advice in real time by clicking a specific button.

[1556] Input: User action (clicking a button).

[1557] Output: Display data of analysis results and advice.

[1558] Specific behavior:

[1559] When a user clicks the "Insight Detector" button in a news app, the analysis results and advice received from the server are immediately displayed. This feature allows users to obtain information while being aware of exaggeration and bias when reading the news.

[1560] (Application example 2)

[1561] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1562] Current information provision systems have difficulty detecting exaggerations and biases in the information users are exposed to and providing it in a transparent manner. As a result, users often make decisions based on incorrect information and believe unreliable information. Furthermore, because systems do not take the user's emotional state into account, the advice provided may not be appropriate for the user's psychological state. A comprehensive system is needed to solve these problems.

[1563] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for the user to input information, a means for the server to receive information, a means for the server to analyze the information using a generative model, a means for the server to mark parts containing exaggeration or bias, a means for the server to evaluate differences from the original information, a means for the server to generate a transparency report for the user, a means for the server to analyze the emotional state, and a means for the server to provide advice based on the emotional state. This makes it possible to detect exaggeration or bias in the information the user comes into contact with in real time and provide it in a highly transparent manner, and also to provide appropriate advice according to the user's emotional state.

[1564] A "user" is an entity that inputs information and receives feedback from the system.

[1565] "Information" is text data such as news and advertisements entered by the user.

[1566] A "server" is a computer system that analyzes information received from a user and performs the necessary processing.

[1567] A "generative model" is software that uses artificial intelligence algorithms to analyze information.

[1568] "Analysis" is the process of using generative models to evaluate information and identify its properties.

[1569] "Exaggeration" is when a piece of information is exaggerated in a way that is greater than the actual content.

[1570] "Bias" is a state in which some information is influenced by a particular perspective or prejudice.

[1571] "Marking" is tagging to explicitly indicate exaggerated or biased parts.

[1572] "Difference" refers to the difference between the original information and the analyzed information.

[1573] A "Transparency Report" is a report that clearly shows any exaggeration or bias in the information for users.

[1574] An "emotional state" is the psychological state a user feels when viewing information.

[1575] "Advice" is a set of instructions or suggestions that help users verify information without exaggeration or bias.

[1576] The present invention is a system that detects exaggeration and bias in the information a user comes into contact with in real time, provides the information in a highly transparent manner, and provides advice that takes into account the user's emotional state. Specific embodiments for realizing this system are described below.

[1577] Program processing overview

[1578] The system is implemented as an "Ad-Transparency Detector" application installed on smartphones or smart glasses. When a user views an ad, it analyzes the ad text and marks any exaggerated or biased parts, displaying them in real time. It also analyzes the user's emotions based on their browsing behavior and comments, and provides advice based on those emotions.

[1579] Hardware and Software

[1580] Hardware:

[1581] Smartphone or smart glasses: The device through which the user views the advertisement.

[1582] Server: A central management system that analyzes information and runs generative models.

[1583] software:

[1584] Generative models: Artificial intelligence algorithms that analyze information and identify exaggeration and bias.

[1585] Natural Language Processing Library (nltk): A library for analyzing the emotional state of the user.

[1586] Data Acquisition Library (requests): A library for acquiring text data for advertisements.

[1587] Specific examples

[1588] A user launches the "Ad-Transparency Detector" application and enters the URL of an ad. The ad includes a claim that "This product has achieved record sales!" The system's server receives the ad text and analyzes it using a generative model. The generative model determines that the "record sales" part is exaggerated and marks it.

[1589] The server then analyzes the user's input comments and determines that the user is feeling "anxious." Based on the user's emotional state, the server provides advice such as, "This ad contains exaggerated statements. Please also check the following reliable sources: official website, third-party review sites, etc. Don't worry too much."

[1590] Example of a prompt to be entered

[1591] There is an advertisement that claims, "This product has achieved record sales!" Please evaluate whether this is an exaggeration and explain why. Furthermore, if a user viewing this advertisement feels "anxious," what advice would you give them?

[1592] In this way, users can easily assess the trustworthiness of the advertisements they come into contact with and make appropriate decisions.

[1593] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1594] Step 1:

[1595] The user enters the URL of an ad. The user launches the "Ad-Transparency Detector" application and enters or pastes the URL of the ad. At this time, the URL entered by the user is sent from the device to the server. The input data is the URL of the ad, and the output data is the URL sent to the server.

[1596] Step 2:

[1597] The server retrieves the ad text. Using the received URL, the server uses a data retrieval library (requests) to retrieve the HTML content of the corresponding ad page. The input data is the ad URL, and the output data is the retrieved ad text. Specifically, an HTTP request is sent and the server downloads the HTML source of the ad page.

[1598] Step 3:

[1599] The server analyzes the ad text. The acquired ad text is passed to a generative model, which analyzes the information. The generative model uses an artificial intelligence algorithm to evaluate the ad text and identify parts that contain exaggeration or bias. The input data is the ad text, and the output data is the analysis results. Specifically, natural language processing technology is used to evaluate the context.

[1600] Step 4:

[1601] The server marks exaggerated or biased parts. Parts with high levels of exaggeration or bias identified by the generative model are tagged with specific tags (e.g., <exaggerated>, <biased>) to make them easy for users to understand. The input data is the analysis results, and the output data is the marked advertising text. Specifically, tags are inserted into the relevant parts of the text.

[1602] Step 5:

[1603] The server analyzes the user's emotional state. It uses a natural language processing library (nltk) to analyze the emotional state based on the comments and feedback entered by the user when viewing the ad text. The input data is the user's comments, and the output data is the emotional state (e.g., anxiety, joy). Specifically, it applies a sentiment analysis algorithm to calculate an emotional score.

[1604] Step 6:

[1605] The server evaluates the differences from the original information. It compares the original ad text with the marked ad text and generates a transparency report to indicate to the user any exaggerations or biases. The input data is the original ad text and the marked ad text, and the output data is a transparency report. Specifically, it uses a text comparison algorithm to extract the differences.

[1606] Step 7:

[1607] The server provides advice based on the emotional state. Advice reflecting the emotional state is generated and presented to the user. The input data is the emotional state and a transparency report, and the output data is advice. Specifically, an appropriate message is generated according to the emotional state and presented to the user as feedback.

[1608] Step 8:

[1609] The server sends the transparency report and advice to the terminal. The generated transparency report and advice are sent to the device used by the user (smartphone or smart glasses) and displayed in real time. The input data are the transparency report and advice, and the output data is the information displayed on the user's terminal. Specifically, the data is sent via network communication.

[1610] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1611] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1612] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1613] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1614] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1615] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1616] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1617] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1618] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1619] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1620] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1621] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1622] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1623] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1624] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1625] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1626] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1627] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1628] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1629] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1630] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1631] The following is further disclosed regarding the above embodiment.

[1632] (Claim 1)

[1633] a means for a user to input information;

[1634] a means for the server to receive the information;

[1635] a means for the server to analyze the information using the generative model;

[1636] a means for the server to mark portions containing exaggeration or bias;

[1637] A means for the server to evaluate the difference from the original information;

[1638] means for the server to generate a transparency report for the user;

[1639] A system including a means for a server to provide advice for verifying the truth of information.

[1640] (Claim 2)

[1641] 10. The system of claim 1, wherein the user uses a news app to enter information.

[1642] (Claim 3)

[1643] 10. The system of claim 1, wherein the generative model utilizes artificial intelligence.

[1644] "Example 1"

[1645] (Claim 1)

[1646] a means for a user to input information;

[1647] A means for transmitting input information from the terminal to a server;

[1648] a means for the server to receive the information;

[1649] a means for the server to analyze the information using the generated AI model;

[1650] A means for the server to evaluate the content of the article and identify exaggeration or bias; and

[1651] a means for the server to mark portions containing exaggeration or bias;

[1652] a means for the server to evaluate the difference from the original information and generate a transparency report;

[1653] a means by which the server provides advice for verifying the truth of the information;

[1654] means for the server to display the analysis results and advice on a user interface;

[1655] A system including:

[1656] (Claim 2)

[1657] 10. The system of claim 1, wherein the URL or text of a news article is input.

[1658] (Claim 3)

[1659] 10. The system of claim 1, wherein the generative AI model utilizes artificial intelligence.

[1660] "Application Example 1"

[1661] (Claim 1)

[1662] a means for a user to input information;

[1663] a means for the server to receive the information;

[1664] a means for the server to analyze the information using the generative model;

[1665] a means for the server to mark portions containing exaggeration or bias;

[1666] A means for the server to evaluate the difference from the original information;

[1667] means for the server to generate a transparency report for the user;

[1668] a means by which the server provides advice for verifying the truth of the information;

[1669] A means to capture and analyze advertising content, identify exaggerations and bias, and provide transparency reports and specific advice;

[1670] A system including:

[1671] (Claim 2)

[1672] 10. The system of claim 1, wherein the user uses a news app to enter information.

[1673] (Claim 3)

[1674] 10. The system of claim 1, wherein the generative model utilizes artificial intelligence.

[1675] (Claim 4)

[1676] 10. The system of claim 1 implemented as an application installed on a smartphone.

[1677] "Example 2: Combining Emotion Engines"

[1678] (Claim 1)

[1679] a means for a user to input information;

[1680] means for the terminal to transmit information to a server;

[1681] a means for the server to receive the information;

[1682] a means for the server to analyze the information using the generative model;

[1683] a means for the server to mark portions containing exaggeration or bias;

[1684] A means for the server to analyze the user's emotions;

[1685] A means for the server to evaluate the difference from the original information;

[1686] means for the server to generate a transparency report for the user;

[1687] A system including a means for a server to provide advice for verifying the truth of information.

[1688] (Claim 2)

[1689] 10. The system of claim 1, wherein the user uses a news app to enter information.

[1690] (Claim 3)

[1691] 10. The system of claim 1, wherein the generative model utilizes artificial intelligence.

[1692] "Application example 2 when combining emotion engines"

[1693] (Claim 1)

[1694] a means for a user to input information;

[1695] a means for the server to receive the information;

[1696] a means for the server to analyze the information using the generative model;

[1697] a means for the server to mark portions containing exaggeration or bias;

[1698] A means for the server to evaluate the difference from the original information;

[1699] means for the server to generate a transparency report for the user;

[1700] a means for the server to analyze the emotional state;

[1701] The system includes a means by which the server provides advice based on emotional state.

[1702] (Claim 2)

[1703] 10. The system of claim 1, wherein the user uses a device application to input information.

[1704] (Claim 3)

[1705] 10. The system of claim 1, wherein the generative model utilizes artificial intelligence. [Explanation of symbols]

[1706] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for a user to input information; a means for the server to receive the information; a means for the server to analyze the information using the generative model; a means for the server to mark portions containing exaggeration or bias; A means for the server to evaluate the difference from the original information; means for the server to generate a transparency report for the user; A system including a means for a server to provide advice for verifying the truth of information.

2. The system of claim 1 , wherein the user uses a news app to input information.

3. The system of claim 1 , wherein the generative model utilizes artificial intelligence.

Images