System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting social media to identify false information or Propaganda; means for cleansing the collected social media; means for analyzing the cleansed social media to detect false information or Propaganda; and means for displaying an alert if false information or Propaganda is detected.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] In recent years, the spread of disinformation and propaganda on social media platforms has become a societal problem. Such disinformation significantly hinders the formation of fair public opinion, especially during election periods and important social movements, posing a risk to the very foundations of democracy. Currently, monitoring and countering disinformation is extremely costly, and manual monitoring and response is inefficient, necessitating rapid countermeasures. Furthermore, automated systems for detecting disinformation are still limited, and their accuracy and effectiveness remain issues. [Means for solving the problem]

[0005] To address the above-mentioned challenges, the present invention provides a system that includes a means for collecting social media data to identify false information and propaganda, a means for cleansing the collected data, a means for analyzing the cleansed data to detect false information and propaganda, and a means for displaying a warning when false information and propaganda are detected. The system further includes a means for performing sentiment analysis on the cleansed data and a means for recording the results of the detection of false information and propaganda and generating statistical information, thereby enabling more accurate and effective detection and response to false information. This system can automatically and in real time suppress the spread of false information and support social decision-making based on accurate information.

[0006] "Disinformation" is false information created with the purpose of deliberately distorting the truth and causing misunderstanding.

[0007] "Propaganda" is advertising or information used to manipulate people's perceptions and behavior with a specific purpose.

[0008] "Social media" is an online platform that allows individuals to disseminate, share, and communicate information via the Internet.

[0009] "Data collection methods" are methods or systems for automatically obtaining data such as social media posts and comments.

[0010] A "cleansing means" is a method or system for removing unnecessary information from collected data to make it ready for analysis.

[0011] "Means for analysis" refers to a method or system for using cleansed data to perform evaluation or verification according to a specific purpose.

[0012] A "means of displaying a warning" is a method or system for notifying users and administrators when false information or propaganda is detected.

[0013] "Sentiment analysis" is an analytical method for assessing emotional states such as positive, negative, or neutral from text data.

[0014] A "means for generating statistical information" is a method or system that numerically demonstrates certain patterns or trends based on collected and analyzed data. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0023] [First embodiment]

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

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

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

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

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

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

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

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

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

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

[0034] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

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

[0036] The present invention relates to a social media monitoring system for identifying disinformation and propaganda and taking countermeasures. Specifically, the system provides a server that collects social media data, cleanses the collected data, analyzes the cleansed data, detects disinformation and propaganda, and displays warnings or notifications based on the detection results.

[0037] System Overview

[0038] 1. Data collection methods:

[0039] The server collects post data from the social media API, including metadata such as the post content (text), poster information, and post date and time.

[0040] 2. Data cleansing methods:

[0041] The server cleanses the collected data by removing unnecessary elements such as links, user tags, and hashtags, resulting in text data suitable for analysis.

[0042] 3. Data analysis methods:

[0043] The server uses the cleansed data to perform sentiment analysis and content validation, using libraries such as TextBlob to assess the text's emotional state as positive, negative, or neutral.

[0044] 4. Detecting Disinformation:

[0045] Based on the analysis results, the server sets thresholds for positive and negative scores to identify posts that are likely to contain false information or propaganda, and displays a warning if the negative score exceeds a certain level.

[0046] 5. Warning display means:

[0047] If the server detects false information or propaganda, it will display or notify the user of the post's ID and reason as a warning.

[0048] Explaining program processing in natural language

[0049] 1. Data Collection:

[0050] The server collects the latest post data from the social media API, sends an HTTP request to the API endpoint, and parses the JSON data received as a response.

[0051] 2. Data cleansing:

[0052] The server uses regular expressions to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data. By cleansing, only the text is extracted and formatted for analysis.

[0053] 3. Data Analysis:

[0054] The server performs sentiment analysis on the cleansed text data, using libraries such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0055] 4. Detecting Disinformation:

[0056] Based on the calculated sentiment score, if the negative score is above a certain level, the server determines that the post is likely to be false information or propaganda.

[0057] 5. Warning display:

[0058] The server will issue a warning for posts that contain false information or propaganda, and will notify the user with the post ID and the reason for the detection, for example by displaying a warning message on the console or sending a notification email to an administrator.

[0059] Specific examples

[0060] For example, suppose your server collects the following post data from an API:

[0061] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[0062] The server cleanses this submission and formats it as follows:

[0063] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[0064] The server then performs sentiment analysis on this cleansed text data to detect high negative scores. If the negative score exceeds a certain threshold, the server determines that the post is misinformation or propaganda and displays a warning to the user:

[0065] "Post ID: 12345 - Alert: Detected potential misinformation or propaganda"

[0066] In this way, the present invention is a system that suppresses the spread of false information and propaganda and supports decision-making based on accurate information.

[0067] The processing flow will be explained below.

[0068] Step 1:

[0069] The server sends an HTTP GET request to the specified social media API URL, collects post data from the API, and receives JSON formatted data as a response.

[0070] Step 2:

[0071] The server parses the JSON data obtained from the response and extracts metadata such as the content (text) of each post, poster information, and posting date and time.

[0072] Step 3:

[0073] The server cleanses the text data of each extracted post by removing links (URLs) using regular expressions, removing user tags (@username) and hashtags using regular expressions, and finally, converting redundant whitespace into a single space.

[0074] Step 4:

[0075] The server performs sentiment analysis on the cleansed text data, using libraries such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0076] Step 5:

[0077] The server evaluates the resulting sentiment score and determines that a post is likely to be misinformation or propaganda, especially if the negative score exceeds a predefined threshold, specifically if the negative score is less than -0.5.

[0078] Step 6:

[0079] If misinformation or propaganda is detected, the server generates a warning message containing the post's ID and the reason for the detection (a high negative score).

[0080] Step 7:

[0081] The server notifies the user of the generated warning message by displaying the warning message on the console or by sending a notification email to the administrator.

[0082] For example, if the server collects a post that reads "The government is lying. Don't trust anyone!", the cleansed text will be "The government is lying. Don't trust anyone!". Sentiment analysis will determine that this text has a high negative score and is misinformation or propaganda. The server will generate a warning message that reads "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and notify the user.

[0083] Example 1

[0084] 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."

[0085] In modern society, social media has become a major means of communication, but at the same time, the spread of false information and propaganda has become a social problem. This false information and propaganda originates from unreliable sources and spreads rapidly, damaging trust in society. There is a need for a system that can solve this problem and provide accurate and reliable information.

[0086] 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.

[0087] In this invention, the server includes means for collecting social media data, means for cleansing the collected data, means for analyzing the cleansed data to detect false information and propaganda, means for performing sentiment analysis, means for displaying a warning when false information and propaganda are detected, and means for sending a notification. This enables early detection of false information and propaganda and notification thereof, thereby enabling the provision of highly reliable information.

[0088] "Social media" refers to online services that allow people to share, exchange, and communicate with each other via the Internet.

[0089] "Data collection means" refers to the functions and devices used to obtain information from social media.

[0090] "Data cleansing means" refers to a function or device for organizing collected information and removing unnecessary elements.

[0091] "Data analysis means" refers to a function or device for analyzing cleansed information and extracting specific patterns or features.

[0092] "Disinformation" refers to information that is intentionally or falsely stated and that is not factually correct.

[0093] "Propaganda" refers to information that is spread with a specific intention or purpose, and that is not necessarily factual.

[0094] "Sentiment analysis means" refers to a function or device for assessing the emotional state (positive, negative, neutral) of text data.

[0095] "Warning display means" means a feature or device that provides a visual or other warning when disinformation or propaganda is detected.

[0096] "Notification sending means" refers to a function or device for sending a notification to the outside when a specific condition is met.

[0097] The present invention relates to a social media monitoring system for identifying and combating false information and propaganda. Specifically, the system provides a server that collects social media data, cleanses the collected data, analyzes the cleansed data, detects false information and propaganda, and displays warnings or notifications based on the detection results.

[0098] First, the server collects the latest post data through the social media API. For example, when using the Twitter API, the server sends an HTTP request to the appropriate API endpoint and analyzes the JSON data received as a response. This data includes metadata such as the post content (text), poster information, and the post date and time.

[0099] Next, the server uses regular expressions to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data. This process extracts only the text and formats it into a form suitable for analysis. A specific example of cleansing is a method of removing links and tags using Python's regular expression library.

[0100] The server then performs sentiment analysis on the cleansed text data. For sentiment analysis, it uses a natural language processing library such as TextBlob. Using TextBlob, it is possible to calculate a positive, negative, or neutral sentiment score for the text. For example, it calculates that the negative score for the text "Breaking news! Government is lying! Do not trust anyone! FakeNews" is -0.8.

[0101] Next, the server makes a judgment to detect misinformation or propaganda based on the sentiment score. If the negative score exceeds a certain threshold, the post is judged to be misinformation or propaganda. Based on this judgment result, the server issues a warning for posts in which misinformation or propaganda has been detected. The warning is notified to the user with the post ID and the reason for the detection. For example, a warning message such as "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" can be generated and a notification email can be sent to the administrator.

[0102] As a concrete example, a server collects the post "Breaking news! Government is lying! Do not trust anyone! FakeNews" from the Twitter API, cleanses it, and formats it as "Breaking news! Government is lying! Do not trust anyone! FakeNews." Next, it performs sentiment analysis using TextBlob and detects a high negative score, determining that the post is misinformation or propaganda. As a result, the server generates a warning "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and notifies the user.

[0103] In this way, the present invention realizes a system that suppresses the spread of false information and propaganda and provides accurate information.

[0104] Example prompts for generative AI models

[0105] "Use Twitter's API to remove links and hashtags from posts collected, and use TextBlob to perform sentiment analysis. Generate warning messages for posts with high negative scores."

[0106] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0107] Step 1:

[0108] The server collects post data from social media APIs. The server sends an HTTP request to the API endpoint and receives JSON format data as a response. The input is a request that includes an API key and query parameters, and the output is an array of the received post data. As a specific example, the Twitter API endpoint "statuses / user_timeline" is used to obtain the latest posts from a specific user.

[0109] Step 2:

[0110] The server cleanses the received post data. It uses regular expressions to remove links (URLs), user tags (@username), and hashtags. This cleansing process passes the received JSON data post content as input, and outputs text data with unnecessary elements removed. As a specific example of how it works, the Python regular expression library is used to format "Breaking news! Government is lying! Do not trust anyone! FakeNews" to "Breaking news! Government is lying! Do not trust anyone! FakeNews".

[0111] Step 3:

[0112] The server performs sentiment analysis on the cleansed text data. It uses a natural language processing library such as TextBlob to calculate a sentiment score. It receives the cleansed text as input and outputs a positive, negative, or neutral sentiment score. As a specific example, TextBlob is used to calculate a negative score of -0.8 for "Breaking news! Government is lying! Do not trust anyone! FakeNews."

[0113] Step 4:

[0114] The server detects false information and propaganda based on the sentiment score. The server compares the sentiment score with a set negative score threshold to identify posts that are likely to be false information or propaganda. The server receives the sentiment score as input and obtains posts determined to be false information as output. As a specific example of operation, if the negative score is -0.5 or less, the server determines that the post is false information.

[0115] Step 5:

[0116] The server displays a warning if misinformation or propaganda is detected. The server generates a warning message including the detected post ID, username, and reason for detection, and notifies the user. The input is the detected post information, and the output is the warning message. As a specific example of operation, the server generates a warning message saying "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and sends a notification email to the user.

[0117] (Application example 1)

[0118] 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."

[0119] This invention aims to solve the technical challenges of detecting and quickly addressing disinformation and propaganda spreading on social media. In particular, there is a need to collect and analyze information in real time, visualize it, and provide prompt warnings. However, current systems have difficulty integrating multiple data collection methods and complex processes such as speech recognition, text analysis, and sentiment analysis.

[0120] 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.

[0121] In this invention, the server includes means for collecting social media information to identify false information and propaganda, means for cleansing the collected information, means for analyzing the cleansed information to detect false information and propaganda, means for displaying a warning when false information and propaganda are detected, means for collecting the latest media information through speech recognition and text analysis, means for performing sentiment analysis based on the speech recognition and text analysis, and means for displaying a warning on the display of the smart device. This makes it possible to quickly mitigate the impact of false information and propaganda by evaluating the authenticity of information in real time and displaying a warning.

[0122] "Disinformation" is information that is not based on facts and is spread intentionally or unintentionally with the intent to mislead or confuse.

[0123] "Propaganda" is a method of selectively providing or manipulating information with a specific political, social, or religious intent, and is used to influence people's opinions and actions.

[0124] "Social media" refers to a platform that allows users to share, exchange, and disseminate information via the Internet, and a typical example is SNS (social networking service).

[0125] "Information collection means" refers to the means for collecting post data and metadata sent on social media, and is realized using technologies such as APIs.

[0126] "Cleansing methods" are methods used to remove unnecessary data (e.g., links, user tags, hashtags) from collected information and prepare it in a format suitable for analysis.

[0127] "Analysis methods" are means for detecting false information and propaganda by conducting sentiment analysis and content verification on cleansed information.

[0128] "Warning display means" refers to means for displaying or notifying users of false information or propaganda as a warning when such information is detected.

[0129] "Speech recognition and text analysis means" refers to means for converting speech data into text data and analyzing the text data, and extracting information using natural language processing techniques.

[0130] "Sentiment analysis means" is a means for evaluating the emotional state (positive, negative, neutral) of text data, and analyzes the credibility of information by assigning an emotional score.

[0131] "Smart device display means" means a means for visually displaying important information or warnings to a user on smart glasses or other wearable devices.

[0132] This invention relates to a system for identifying and countering false information and propaganda, and specifically includes the following means:

[0133] The server first uses social media APIs to collect post data and related metadata, then uses a cleansing method to remove unnecessary information such as links, user tags, and hashtags from the collected data, converting it into text suitable for analysis.

[0134] Next, speech recognition and text analysis means are used to acquire audio and video data from a smart device (e.g., smart glasses) and convert them into text data. Here, OpenAI®'s speech recognition API is used.

[0135] The cleansed text data is then evaluated using sentiment analysis tools. Natural language processing libraries such as the TextBlob library and Hugging Face's Transformers are used to calculate a sentiment score for the text. If the sentiment score exceeds a certain threshold, the text is marked as misinformation or propaganda.

[0136] When disinformation or propaganda is detected, a warning will be displayed on the smart device's display, including the ID of the detected information and the reason for the detection, and the user will be visually notified. For example, a warning message will be displayed in real time on the display of smart glasses.

[0137] Consider the following scenario: Smart glasses recognize the speech phrase "Breaking news! The government is lying!" and convert it into text. Sentiment analysis of the text detects a high negative score, identifying it as disinformation or propaganda, and a warning is displayed on the glasses' display.

[0138] In addition to cleansing and sentiment analysis of the generated data, the following example prompts can be used to feed the generative AI model:

[0139] Example prompt sentence:

[0140] Transcribe the speech "Breaking news! The government is lying! Trust no one!" and perform sentiment analysis on this text. If the negative score is high, show a warning to the user.

[0141] This allows users to make informed decisions in real time, quickly reducing the impact of disinformation and propaganda.

[0142] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0143] Step 1:

[0144] The server uses the social media API to collect the latest post data and related metadata. Specifically, it sends an HTTP request to the API endpoint and parses the JSON data received as a response. The input is data from the API, and the output is the parsed post data and metadata.

[0145] Step 2:

[0146] The server cleanses the collected post data by using regular expressions to remove links (URLs), user tags (@username), hashtags, etc. The input is the analyzed post data, and the output is cleansed text data with unnecessary information removed.

[0147] Step 3:

[0148] The server evaluates the cleansed text data using speech recognition and text analysis. First, the voice data is converted into text data using OpenAI's speech recognition API, and then the converted text data is analyzed. The input is the voice data and the cleansed text data, and the output is the analyzed text data.

[0149] Step 4:

[0150] The server performs sentiment analysis on the parsed text data, using the TextBlob library and Hugging Face Transformers to calculate a positive, negative, or neutral sentiment score for the text. The input is the parsed text data, and the output is the sentiment score.

[0151] Step 5:

[0152] The server determines the likelihood of misinformation or propaganda based on the sentiment score. If the sentiment score exceeds a certain threshold, the post is identified as misinformation or propaganda. The input is the sentiment score, and the output is the results of identifying the post as misinformation.

[0153] Step 6:

[0154] The server displays a warning when false information or propaganda is identified. Specifically, it sends and displays a warning message in real time on the display of a smart device (such as smart glasses). The input is the result of identifying false information, and the output is the display of a warning message.

[0155] Step 7:

[0156] Users can check the warnings displayed on their smart devices and take measures to prevent the effects of disinformation and propaganda. Specific actions include receiving the alert and checking reliable sources of information. The input is the warning message, and the output is the user's response.

[0157] 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.

[0158] The present invention relates to a social media monitoring system for identifying false information and propaganda and implementing countermeasures. Furthermore, by combining an emotion engine that recognizes user emotions, the system is equipped with the function of assessing the impact of false information and propaganda and adjusting the warning content. The system operates mainly on a server and includes the following main means:

[0159] System Overview

[0160] 1. Data collection methods:

[0161] The server collects post data from the social media API, including metadata such as the post content (text), poster information, and post date and time.

[0162] 2. Data cleansing methods:

[0163] The server cleanses the collected data by removing unnecessary elements such as links, user tags, and hashtags, resulting in text data suitable for analysis.

[0164] 3. Data analysis methods:

[0165] The server uses the cleansed data to perform sentiment analysis and content validation, using libraries such as TextBlob to assess the text's emotional state as positive, negative, or neutral.

[0166] 4. Detecting Disinformation:

[0167] Based on the analysis results, the server sets thresholds for positive and negative scores to identify posts that are likely to contain false information or propaganda, and displays a warning if the negative score exceeds a certain level.

[0168] 5. Warning display means:

[0169] If the server detects false information or propaganda, it will display or notify the user of the post's ID and reason as a warning.

[0170] 6. Emotion Engine:

[0171] The server uses an emotion engine to recognize users' emotions and assess the impact of misinformation and propaganda, analyzing text inputs and behavioral data when users interact with the system.

[0172] 7. Adjustment of warning content:

[0173] The server adjusts the warning content based on the results of the emotion engine, for example, strengthening the warning if the user is already emotionally affected, and appropriately adjusting the warning content to avoid excessively negative impact.

[0174] 8. Recording and statistical information generation means:

[0175] The server will record the results of detecting false information and propaganda and generate statistical information based on this information, which will be useful for analyzing trends in the spread of false information and developing effective countermeasures.

[0176] Explaining program processing in natural language

[0177] 1. Data Collection:

[0178] The server collects the latest post data from the social media API, sends an HTTP request to the API endpoint, and parses the JSON data received as a response.

[0179] 2. Data cleansing:

[0180] The server uses regular expressions to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data. By cleansing, only the text is extracted and formatted for analysis.

[0181] 3. Data Analysis:

[0182] The server performs sentiment analysis on the cleansed text data, using libraries such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0183] 4. Detecting Disinformation:

[0184] Based on the sentiment score, the server determines that there is a high possibility of misinformation or propaganda if the negative score exceeds a predetermined threshold.

[0185] 5. User emotion recognition using emotion engine:

[0186] The server analyzes the user's input and behavioral data and uses an emotion engine to recognize the user's emotional state, which can include joy, sadness, anger, surprise, etc.

[0187] 6. Adjustment of warning content:

[0188] The server adjusts the warning content appropriately based on the analysis results of the emotion engine. For example, if the user is already in a negative emotional state, the server displays a warning message that is careful not to further exacerbate the negative emotional state.

[0189] 7. Warning display:

[0190] The server issues a warning to posts that contain false information or propaganda, including the post's ID and the reason for the detection, and reflects the evaluation results of the emotion engine.

[0191] 8. Recording and Statistics Generation:

[0192] The server records the results of detecting false information and propaganda and generates statistical information that includes the results of the emotion engine, which allows for analysis of the spread of false information and countermeasures based on the emotional state of users.

[0193] Specific examples

[0194] For example, suppose your server collects the following post data from an API:

[0195] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[0196] The server cleanses this submission and formats it as follows:

[0197] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[0198] The server then performs sentiment analysis on this cleansed text data to detect high negative scores, and if the negative score exceeds a certain threshold, the server identifies the post as misinformation or propaganda.

[0199] The server uses an emotion engine to recognize the user's emotional state based on the user's behavioral data. For example, if the user has seen many negative posts in the past 24 hours, the emotion engine will recognize that the user is in a negative emotional state.

[0200] Based on the results, the server generates a warning message "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and notifies the user after adjusting the warning content.

[0201] In this way, the present invention is a system that curbs the spread of disinformation and propaganda, taking into account the emotional state of the user and taking appropriate measures.

[0202] The processing flow will be explained below.

[0203] Step 1:

[0204] The server sends an HTTP GET request to the specified social media API URL, collects post data from the API, and receives JSON formatted data as a response.

[0205] Step 2:

[0206] The server parses the JSON data obtained from the response and extracts metadata such as the content (text) of each post, poster information, and posting date and time.

[0207] Step 3:

[0208] The server cleanses the text data of each extracted post. Specifically, it removes links (URLs) using regular expressions, and also removes user tags (@username) and hashtags using regular expressions. By cleansing, the text data becomes suitable for analysis.

[0209] Step 4:

[0210] The server performs sentiment analysis on the cleansed text data, using libraries such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0211] Step 5:

[0212] The server evaluates the resulting sentiment score and determines that a post is likely to be misinformation or propaganda, especially if the negative score exceeds a predefined threshold, specifically if the negative score is less than -0.5.

[0213] Step 6:

[0214] The server analyzes user input and behavioral data and uses an emotion engine to recognize the user's emotional state. For example, if a user has viewed many negative posts in the past 24 hours, the emotion engine will determine that the user is in a negative emotional state.

[0215] Step 7:

[0216] The server adjusts the warning content based on the emotion recognition results. If the user is already in a negative emotional state, the warning will be strengthened. The server also adjusts the warning content appropriately to avoid excessively negative impact.

[0217] Step 8:

[0218] If misinformation or propaganda is detected, the server generates a warning message containing the post's ID and the reason for the detection (that it was determined to have a high negative score).

[0219] Step 9:

[0220] The server notifies the user of the generated warning message by displaying the warning message on the console or by sending a notification email to the administrator.

[0221] Step 10:

[0222] The server records the results of detecting false information and propaganda and generates statistical information that includes the results of the emotion engine. This statistical information is useful for taking measures based on the spread of false information and the emotional state of users.

[0223] As a concrete example, suppose a server collects a post with the title "Breaking news! Government is lying! Do not trust anyone! FakeNews." After cleansing this post, the resulting text is formatted as "Breaking news! Government is lying! Do not trust anyone! FakeNews." Sentiment analysis determines that this text has a high negative score and is misinformation or propaganda. If the emotion engine subsequently determines that the user is in a negative emotional state, the server generates a warning message such as "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" to notify the user.

[0224] Example 2

[0225] 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."

[0226] In today's social media, false information and propaganda often spread quickly and have a significant impact on society. However, current systems are insufficient in detecting and dealing with false information, and do not take appropriate measures that take into account the emotional state of users. Therefore, it is necessary to solve the problem of quickly detecting false information and propaganda and minimizing their emotional impact on users.

[0227] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting social media data, a means for cleansing the collected data, a means for analyzing the cleansed data and detecting false information or propaganda, a means for displaying a warning when false information or propaganda is detected, a means for analyzing user behavior data and recognizing the user's emotional state using an emotion engine, and a means for adjusting the content of the warning based on the results of the emotion engine. This enables the rapid detection of false information and propaganda and the display of an appropriate warning that takes the user's emotional state into consideration.

[0228] "Data collection means" refers to a device or program that uses a social media API to obtain data such as the content of posts, poster information, and posting date and time.

[0229] A "data cleansing tool" is a device or program that removes unnecessary elements such as links, user tags, and hashtags from collected data and formats it into a form suitable for analysis.

[0230] The "data analysis means" is a device or program that performs sentiment analysis and content verification on the cleansed data and calculates a positive, negative, or neutral sentiment score.

[0231] A "fake news detection tool" is a device or program that, based on the results of data analysis, identifies posts that may be fake or propaganda based on a set threshold.

[0232] A "warning display means" is a device or program that generates and displays a message or alert to notify the user when false information or propaganda is detected.

[0233] An "emotion engine" is an algorithm or device that analyzes a user's text input and behavioral data to recognize the user's emotional state.

[0234] The "alert content adjustment means" is a device or program that optimizes the alert content based on the results of the emotion engine and taking into account the emotional state of the user.

[0235] The present invention relates to a social media monitoring system for identifying disinformation and propaganda and taking countermeasures. The system is server-based and includes the following main means:

[0236] The server first collects post data from the social media API. Specifically, it sends an HTTP request to the API endpoint and analyzes the JSON data received as a response. The collected data includes the post content (text), poster information, and posting date and time. This allows the server to store the most recent post data in a local database.

[0237] Next, the server uses a data cleansing method to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data using regular expressions, extracting only the text and preparing it in a form suitable for analysis.

[0238] The server then performs sentiment analysis on the cleansed data using a data analysis tool. For sentiment analysis, it uses a library such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text, which is then used to evaluate and record the emotional state of the post.

[0239] The server then uses a false information detection method based on the sentiment score to determine that a message is likely to be false information or propaganda if the negative score exceeds a predetermined threshold. If false information or propaganda is detected, the result is recorded.

[0240] The system also analyzes user behavior data and uses an emotion engine to recognize the user's emotional state. Based on the user's input and past behavioral data, the system identifies the user's emotional state. For example, if a user has seen many negative posts, the system will recognize the user's emotional state as negative.

[0241] Based on the analysis results of the emotion engine, the server uses the warning content adjustment means to appropriately adjust the warning content. For example, if the user is already in a negative emotional state, the server may take measures such as changing the warning text to a softer one.

[0242] Finally, for posts that contain false information or propaganda, the server issues a warning using a warning display method. The warning includes the post's ID and the reason for the detection, and reflects the evaluation results of the emotion engine. This allows the server to notify users of an appropriate warning.

[0243] To illustrate this system, here is an example prompt:

[0244] Please explain how your system would perform sentiment analysis on the following post data, and if the negative score exceeds a certain threshold, display a warning about misinformation or propaganda and adjust the warning appropriately. Post data: 'Breaking news! Government is lying! Do not trust anyone! FakeNews'

[0245] In this way, the present invention can effectively curb the spread of false information and propaganda, taking into account the emotional state of the user and taking appropriate measures.

[0246] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0247] Step 1: Data collection

[0248] The server collects post data from the social media API. Specifically, it sends an HTTP GET request to the API endpoint and analyzes the JSON data received as a response. This process obtains metadata such as the post content (text), poster information, and posting date and time as input, and stores them in a local database.

[0249] Step 2: Data cleansing

[0250] The server uses regular expressions to remove unnecessary elements such as links (URLs), user tags (@username), and hashtags from the collected JSON data. The input to this process is the collected post data, and the output is cleansed text data. Specifically, it applies regular expression patterns to remove unnecessary strings.

[0251] Step 3: Data analysis

[0252] The server performs sentiment analysis on the cleansed text data as input. It uses a library such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text. The output of this process is a sentiment score for each post. Specifically, it calls a library to analyze the text and calculate the sentiment score.

[0253] Step 4: Detecting misinformation

[0254] The server takes the sentiment score as input and checks whether the negative score exceeds a predetermined threshold. If the negative score exceeds the threshold, the post is deemed to be false information or propaganda. The output of this process is a list of posts that have been deemed to be false information. Specifically, the sentiment score is compared with the threshold, and posts that meet the criteria are added to the list.

[0255] Step 5: User Emotion Recognition

[0256] The server collects user behavior data and uses an emotion engine to recognize the user's emotional state. The input is the user's text input and behavior log, and the output is the user's emotional state (joy, sadness, anger, etc.). Specifically, the server analyzes the user's behavior data and identifies the user's emotional state through the emotion engine.

[0257] Step 6: Adjust the warning content

[0258] The server uses the results of the emotion engine as input and adjusts the warning content appropriately. For example, if the user is already in a negative emotional state, the warning text is changed to a softer expression. The output of this process is an adjusted warning message. Specifically, the content of the warning text is generated and changed based on the user's emotional state.

[0259] Step 7: Warning Display

[0260] The server issues a warning for posts that are determined to be false. The input is the post that needs a warning and the adjusted warning message, and the output is a warning that is displayed to the user. Specifically, the generated warning message is displayed through an interface to notify the user.

[0261] Step 8: Recording and generating statistics

[0262] The server records the results of the detection of false information and propaganda and the results of the emotion engine, and generates statistical information. The detection results and the emotion engine results are used as input, and the output is statistical information. Specifically, statistical analysis is performed based on this data, and a report is created to analyze the spread of false information and effective countermeasures.

[0263] (Application example 2)

[0264] 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."

[0265] In modern society, the spread of false information and propaganda through social media has become a major problem. This can lead to users being influenced by false information, causing anxiety and confusion. Furthermore, existing systems only detect false information and do not display warnings that take into account the user's emotional state, which can cause excessive stress. To solve this situation, a system that not only detects false information but also takes into account the user's emotional state and displays appropriate warnings is needed.

[0266] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0267] In this invention, the server includes a data collection means for identifying false information and propaganda, a means for cleansing the collected data, a means for analyzing the cleansed data and detecting false information and propaganda, a means for displaying a warning when false information and propaganda are detected, a means for processing the collected data in real time, a means for analyzing the emotional state of the user, and a means for adjusting the content of the warning based on the emotional state of the user, thereby enabling appropriate detection of false information and propaganda and display of a warning that takes into account the emotional state of the user.

[0268] "Data collection means" refers to the means used to collect posting data and metadata from social media.

[0269] A "cleansing method" is a method for removing unnecessary elements from collected data and preparing text data in a form suitable for analysis.

[0270] The "data analysis means" is a means for analyzing the cleansed text data and determining sentiment scores and false information.

[0271] "Disinformation and propaganda detection measures" are measures that identify posts that are likely to be disinformation or propaganda based on the results of data analysis.

[0272] "Warning display measures" are measures that display a warning on a post when false information or propaganda is detected.

[0273] "Real-time processing means" refers to means for updating and analyzing collected data in real time and responding immediately.

[0274] The "emotional state analysis means" is a means for analyzing the user's emotional state based on the user's past behavioral data and text input.

[0275] The "warning content adjustment means" is a means for adjusting the warning content according to the emotional state of the user based on the result of the emotional state analysis means.

[0276] This paper describes an embodiment of the present invention, which describes a system for identifying false information and propaganda and displaying warnings based on a user's emotional state. The system includes means for data collection, data cleansing, data analysis, false information detection, emotional state analysis, warning content adjustment, and warning display.

[0277] Data collection methods

[0278] The server collects post data from the social media API. Specifically, it sends an HTTP request to the API endpoint and analyzes the JSON data received as a response. This process uses a communication library such as the requests library.

[0279] Data Cleansing Methods

[0280] The server uses regular expressions to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected data, thereby obtaining text data suitable for analysis.

[0281] Data Analysis Methods

[0282] The server performs sentiment analysis on the cleansed text data, using a natural language processing library such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0283] Measures to detect disinformation and propaganda

[0284] The server determines the likelihood of misinformation or propaganda if the negative score exceeds a predetermined threshold based on the sentiment score. This determination is made using a threshold to determine whether a warning should be sent to the user.

[0285] Emotional state analysis means

[0286] The server analyzes user input and behavioral data and uses an emotion engine to recognize the user's emotional state. Based on the user's behavioral data within the past 24 hours, it counts the number of negative posts and evaluates the user's current emotional state.

[0287] Warning content adjustment means

[0288] The server adjusts the warning content based on the results of the emotional state analysis means. For example, if the user is already in a negative emotional state, the server displays a warning message that takes care not to further exacerbate the user's negative emotional state.

[0289] Warning display means

[0290] The server issues a warning to posts that contain false information or propaganda, including the post's ID and the reason for the detection, and reflects the evaluation results of the emotion engine.

[0291] Specific examples

[0292] For example, if your server collects the following post data from an API:

[0293] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[0294] The server cleanses this submission and formats it as follows:

[0295] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[0296] The server then performs sentiment analysis on this cleansed text data to detect high negative scores, and if the negative score exceeds a certain threshold, the server identifies the post as misinformation or propaganda.

[0297] The server uses an emotion engine to recognize the user's emotional state based on the user's behavioral data. For example, if the user has seen many negative posts in the past 24 hours, the emotion engine will recognize that the user is in a negative emotional state.

[0298] Based on the results, the server generates a warning message "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and notifies the user after adjusting the warning content.

[0299] Prompt Sentence Examples

[0300] Example prompts to apply to your API endpoints:

[0301] "Please provide the latest posts data from the social media API endpoint: 'https: / / api.socialmedia.com / latest_posts' additionally, gather the user's past 24-hour activity data for sentiment analysis: [ {"text": "I'm so sad about the news.", "timestamp": "2023-10-20T14:00:00Z"}, {"text": "Angry about the government's actions.", "timestamp": "2023-10-20T16:00:00Z"}]

[0302] In this way, the present invention makes it possible to curb the spread of false information and propaganda and to take appropriate measures taking into account the emotional state of the user.

[0303] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0304] Processing flow

[0305] Step 1: Data collection

[0306] The server sends an HTTP request to the social media API to collect the latest post data, and receives a JSON response from the API endpoint.

[0307] Input: API endpoint

[0308] Output: Post data (JSON format)

[0309] Step 2: Data cleansing

[0310] The server removes unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data using regular expressions.

[0311] Input: Post data (JSON format)

[0312] Output: Cleansed text data

[0313] Step 3: Data analysis

[0314] The server performs sentiment analysis on the cleansed text data, using a natural language processing library such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0315] Input: Cleansed text data

[0316] Output: Sentiment score (positive, negative, neutral)

[0317] Step 4: Detecting disinformation and propaganda

[0318] The server determines that the likelihood of the information being false information or propaganda is high if the negative score exceeds a predetermined threshold based on the emotion score.

[0319] Input: Sentiment score

[0320] Output: Disinformation and propaganda judgement result (True / False)

[0321] Step 5: Analyze emotional state

[0322] The server analyzes the user's behavioral data and uses an emotion engine to recognize the user's emotional state, for example, by counting the number of negative posts the user has made in the past 24 hours.

[0323] Input: User behavior data (post data from the past 24 hours)

[0324] Output: User's emotional state (number of negative comments)

[0325] Step 6: Adjust the warning content

[0326] The server adjusts the warning content based on the results of the emotional state analysis. If the user is already in a negative emotional state, the server generates a warning message that is careful not to further upset the user.

[0327] Input: User's emotional state, misinformation and propaganda judgment results

[0328] Output: Adjusted warning content

[0329] Step 7: View warnings

[0330] The server issues a warning to users when a post contains false information or propaganda, and notifies them. The warning includes the post's ID, the reason for the detection, and the evaluation result by the emotion engine.

[0331] Input: Adjusted warning content, post ID

[0332] Output: Display of warning message

[0333] This process allows the server to effectively detect false information and propaganda and display warnings in real time that take into account the user's emotional state.

[0334] 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.

[0335] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (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.

[0336] 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.

[0337] [Second embodiment]

[0338] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0339] 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.

[0340] 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).

[0341] 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.

[0342] 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.

[0343] 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).

[0344] 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.

[0345] 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.

[0346] 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.

[0347] 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.

[0348] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0349] 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."

[0350] The present invention relates to a social media monitoring system for identifying disinformation and propaganda and taking countermeasures. Specifically, the system provides a server that collects social media data, cleanses the collected data, analyzes the cleansed data, detects disinformation and propaganda, and displays warnings or notifications based on the detection results.

[0351] System Overview

[0352] 1. Data collection methods:

[0353] The server collects post data from the social media API, including metadata such as the post content (text), poster information, and post date and time.

[0354] 2. Data cleansing methods:

[0355] The server cleanses the collected data by removing unnecessary elements such as links, user tags, and hashtags, resulting in text data suitable for analysis.

[0356] 3. Data analysis methods:

[0357] The server uses the cleansed data to perform sentiment analysis and content validation, using libraries such as TextBlob to assess the text's emotional state as positive, negative, or neutral.

[0358] 4. Detecting Disinformation:

[0359] Based on the analysis results, the server sets thresholds for positive and negative scores to identify posts that are likely to contain false information or propaganda, and displays a warning if the negative score exceeds a certain level.

[0360] 5. Warning display means:

[0361] If the server detects false information or propaganda, it will display or notify the user of the post's ID and reason as a warning.

[0362] Explaining program processing in natural language

[0363] 1. Data Collection:

[0364] The server collects the latest post data from the social media API, sends an HTTP request to the API endpoint, and parses the JSON data received as a response.

[0365] 2. Data cleansing:

[0366] The server uses regular expressions to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data. By cleansing, only the text is extracted and formatted for analysis.

[0367] 3. Data Analysis:

[0368] The server performs sentiment analysis on the cleansed text data, using libraries such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0369] 4. Detecting Disinformation:

[0370] Based on the calculated sentiment score, if the negative score is above a certain level, the server determines that the post is likely to be false information or propaganda.

[0371] 5. Warning display:

[0372] The server will issue a warning for posts that contain false information or propaganda, and will notify the user with the post ID and the reason for the detection, for example by displaying a warning message on the console or sending a notification email to an administrator.

[0373] Specific examples

[0374] For example, suppose your server collects the following post data from an API:

[0375] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[0376] The server cleanses this submission and formats it as follows:

[0377] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[0378] The server then performs sentiment analysis on this cleansed text data to detect high negative scores. If the negative score exceeds a certain threshold, the server determines that the post is misinformation or propaganda and displays a warning to the user:

[0379] "Post ID: 12345 - Alert: Detected potential misinformation or propaganda"

[0380] In this way, the present invention is a system that suppresses the spread of false information and propaganda and supports decision-making based on accurate information.

[0381] The processing flow will be explained below.

[0382] Step 1:

[0383] The server sends an HTTP GET request to the specified social media API URL, collects post data from the API, and receives JSON formatted data as a response.

[0384] Step 2:

[0385] The server parses the JSON data obtained from the response and extracts metadata such as the content (text) of each post, poster information, and posting date and time.

[0386] Step 3:

[0387] The server cleanses the text data of each extracted post by removing links (URLs) using regular expressions, removing user tags (@username) and hashtags using regular expressions, and finally, converting redundant whitespace into a single space.

[0388] Step 4:

[0389] The server performs sentiment analysis on the cleansed text data, using libraries such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0390] Step 5:

[0391] The server evaluates the resulting sentiment score and determines that a post is likely to be misinformation or propaganda, especially if the negative score exceeds a predefined threshold, specifically if the negative score is less than -0.5.

[0392] Step 6:

[0393] If misinformation or propaganda is detected, the server generates a warning message containing the post's ID and the reason for the detection (a high negative score).

[0394] Step 7:

[0395] The server notifies the user of the generated warning message by displaying the warning message on the console or by sending a notification email to the administrator.

[0396] For example, if the server collects a post that reads "The government is lying. Don't trust anyone!", the cleansed text will be "The government is lying. Don't trust anyone!". Sentiment analysis will determine that this text has a high negative score and is misinformation or propaganda. The server will generate a warning message that reads "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and notify the user.

[0397] Example 1

[0398] 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."

[0399] In modern society, social media has become a major means of communication, but at the same time, the spread of false information and propaganda has become a social problem. This false information and propaganda originates from unreliable sources and spreads rapidly, damaging trust in society. There is a need for a system that can solve this problem and provide accurate and reliable information.

[0400] 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.

[0401] In this invention, the server includes means for collecting social media data, means for cleansing the collected data, means for analyzing the cleansed data to detect false information and propaganda, means for performing sentiment analysis, means for displaying a warning when false information and propaganda are detected, and means for sending a notification. This enables early detection of false information and propaganda and notification thereof, thereby enabling the provision of highly reliable information.

[0402] "Social media" refers to online services that allow people to share, exchange, and communicate with each other via the Internet.

[0403] "Data collection means" refers to the functions and devices used to obtain information from social media.

[0404] "Data cleansing means" refers to a function or device for organizing collected information and removing unnecessary elements.

[0405] "Data analysis means" refers to a function or device for analyzing cleansed information and extracting specific patterns or features.

[0406] "Disinformation" refers to information that is intentionally or falsely stated and that is not factually correct.

[0407] "Propaganda" refers to information that is spread with a specific intention or purpose, and that is not necessarily factual.

[0408] "Sentiment analysis means" refers to a function or device for assessing the emotional state (positive, negative, neutral) of text data.

[0409] "Warning display means" means a feature or device that provides a visual or other warning when disinformation or propaganda is detected.

[0410] "Notification sending means" refers to a function or device for sending a notification to the outside when a specific condition is met.

[0411] The present invention relates to a social media monitoring system for identifying and combating false information and propaganda. Specifically, the system provides a server that collects social media data, cleanses the collected data, analyzes the cleansed data, detects false information and propaganda, and displays warnings or notifications based on the detection results.

[0412] First, the server collects the latest post data through the social media API. For example, when using the Twitter API, the server sends an HTTP request to the appropriate API endpoint and analyzes the JSON data received as a response. This data includes metadata such as the post content (text), poster information, and the post date and time.

[0413] Next, the server uses regular expressions to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data. This process extracts only the text and formats it into a form suitable for analysis. A specific example of cleansing is a method of removing links and tags using Python's regular expression library.

[0414] The server then performs sentiment analysis on the cleansed text data. For sentiment analysis, it uses a natural language processing library such as TextBlob. Using TextBlob, it is possible to calculate a positive, negative, or neutral sentiment score for the text. For example, it calculates that the negative score for the text "Breaking news! Government is lying! Do not trust anyone! FakeNews" is -0.8.

[0415] Next, the server makes a judgment to detect misinformation or propaganda based on the sentiment score. If the negative score exceeds a certain threshold, the post is judged to be misinformation or propaganda. Based on this judgment result, the server issues a warning for posts in which misinformation or propaganda has been detected. The warning is notified to the user with the post ID and the reason for the detection. For example, a warning message such as "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" can be generated and a notification email can be sent to the administrator.

[0416] As a concrete example, a server collects the post "Breaking news! Government is lying! Do not trust anyone! FakeNews" from the Twitter API, cleanses it, and formats it as "Breaking news! Government is lying! Do not trust anyone! FakeNews." Next, it performs sentiment analysis using TextBlob and detects a high negative score, determining that the post is misinformation or propaganda. As a result, the server generates a warning "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and notifies the user.

[0417] In this way, the present invention realizes a system that suppresses the spread of false information and propaganda and provides accurate information.

[0418] Example prompts for generative AI models

[0419] "Use Twitter's API to remove links and hashtags from posts collected, and use TextBlob to perform sentiment analysis. Generate warning messages for posts with high negative scores."

[0420] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0421] Step 1:

[0422] The server collects post data from social media APIs. The server sends an HTTP request to the API endpoint and receives JSON format data as a response. The input is a request that includes an API key and query parameters, and the output is an array of the received post data. As a specific example, the Twitter API endpoint "statuses / user_timeline" is used to obtain the latest posts from a specific user.

[0423] Step 2:

[0424] The server cleanses the received post data. It uses regular expressions to remove links (URLs), user tags (@username), and hashtags. This cleansing process passes the received JSON data post content as input, and outputs text data with unnecessary elements removed. As a specific example of how it works, the Python regular expression library is used to format "Breaking news! Government is lying! Do not trust anyone! FakeNews" to "Breaking news! Government is lying! Do not trust anyone! FakeNews".

[0425] Step 3:

[0426] The server performs sentiment analysis on the cleansed text data. It uses a natural language processing library such as TextBlob to calculate a sentiment score. It receives the cleansed text as input and outputs a positive, negative, or neutral sentiment score. As a specific example, TextBlob is used to calculate a negative score of -0.8 for "Breaking news! Government is lying! Do not trust anyone! FakeNews."

[0427] Step 4:

[0428] The server detects false information and propaganda based on the sentiment score. The server compares the sentiment score with a set negative score threshold to identify posts that are likely to be false information or propaganda. The server receives the sentiment score as input and obtains posts determined to be false information as output. As a specific example of operation, if the negative score is -0.5 or less, the server determines that the post is false information.

[0429] Step 5:

[0430] The server displays a warning if misinformation or propaganda is detected. The server generates a warning message including the detected post ID, username, and reason for detection, and notifies the user. The input is the detected post information, and the output is the warning message. As a specific example of operation, the server generates a warning message saying "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and sends a notification email to the user.

[0431] (Application example 1)

[0432] 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."

[0433] This invention aims to solve the technical challenges of detecting and quickly addressing disinformation and propaganda spreading on social media. In particular, there is a need to collect and analyze information in real time, visualize it, and provide prompt warnings. However, current systems have difficulty integrating multiple data collection methods and complex processes such as speech recognition, text analysis, and sentiment analysis.

[0434] 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.

[0435] In this invention, the server includes means for collecting social media information to identify false information and propaganda, means for cleansing the collected information, means for analyzing the cleansed information to detect false information and propaganda, means for displaying a warning when false information and propaganda are detected, means for collecting the latest media information through speech recognition and text analysis, means for performing sentiment analysis based on the speech recognition and text analysis, and means for displaying a warning on the display of the smart device. This makes it possible to quickly mitigate the impact of false information and propaganda by evaluating the authenticity of information in real time and displaying a warning.

[0436] "Disinformation" is information that is not based on facts and is spread intentionally or unintentionally with the intent to mislead or confuse.

[0437] "Propaganda" is a method of selectively providing or manipulating information with a specific political, social, or religious intent, and is used to influence people's opinions and actions.

[0438] "Social media" refers to a platform that allows users to share, exchange, and disseminate information via the Internet, and a typical example is SNS (social networking service).

[0439] "Information collection means" refers to the means for collecting post data and metadata sent on social media, and is realized using technologies such as APIs.

[0440] "Cleansing methods" are methods used to remove unnecessary data (e.g., links, user tags, hashtags) from collected information and prepare it in a format suitable for analysis.

[0441] "Analysis methods" are means for detecting false information and propaganda by conducting sentiment analysis and content verification on cleansed information.

[0442] "Warning display means" refers to means for displaying or notifying users of false information or propaganda as a warning when such information is detected.

[0443] "Speech recognition and text analysis means" refers to means for converting speech data into text data and analyzing the text data, and extracting information using natural language processing techniques.

[0444] "Sentiment analysis means" is a means for evaluating the emotional state (positive, negative, neutral) of text data, and analyzes the credibility of information by assigning an emotional score.

[0445] "Smart device display means" means a means for visually displaying important information or warnings to a user on smart glasses or other wearable devices.

[0446] This invention relates to a system for identifying and countering false information and propaganda, and specifically includes the following means:

[0447] The server first uses social media APIs to collect post data and related metadata, then uses a cleansing method to remove unnecessary information such as links, user tags, and hashtags from the collected data, converting it into text suitable for analysis.

[0448] Next, speech recognition and text analysis tools are used to capture audio and video data from smart devices (e.g., smart glasses) and convert them into text data. OpenAI's speech recognition API is used here.

[0449] The cleansed text data is then evaluated using sentiment analysis tools. Natural language processing libraries such as the TextBlob library and Hugging Face's Transformers are used to calculate a sentiment score for the text. If the sentiment score exceeds a certain threshold, the text is marked as misinformation or propaganda.

[0450] When disinformation or propaganda is detected, a warning will be displayed on the smart device's display, including the ID of the detected information and the reason for the detection, and the user will be visually notified. For example, a warning message will be displayed in real time on the display of smart glasses.

[0451] Consider the following scenario: Smart glasses recognize the speech phrase "Breaking news! The government is lying!" and convert it into text. Sentiment analysis of the text detects a high negative score, identifying it as disinformation or propaganda, and a warning is displayed on the glasses' display.

[0452] In addition to cleansing and sentiment analysis of the generated data, the following example prompts can be used to feed the generative AI model:

[0453] Example prompt sentence:

[0454] Transcribe the speech "Breaking news! The government is lying! Trust no one!" and perform sentiment analysis on this text. If the negative score is high, show a warning to the user.

[0455] This allows users to make informed decisions in real time, quickly reducing the impact of disinformation and propaganda.

[0456] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0457] Step 1:

[0458] The server uses the social media API to collect the latest post data and related metadata. Specifically, it sends an HTTP request to the API endpoint and parses the JSON data received as a response. The input is data from the API, and the output is the parsed post data and metadata.

[0459] Step 2:

[0460] The server cleanses the collected post data by using regular expressions to remove links (URLs), user tags (@username), hashtags, etc. The input is the analyzed post data, and the output is cleansed text data with unnecessary information removed.

[0461] Step 3:

[0462] The server evaluates the cleansed text data using speech recognition and text analysis. First, the voice data is converted into text data using OpenAI's speech recognition API, and then the converted text data is analyzed. The input is the voice data and the cleansed text data, and the output is the analyzed text data.

[0463] Step 4:

[0464] The server performs sentiment analysis on the parsed text data, using the TextBlob library and Hugging Face Transformers to calculate a positive, negative, or neutral sentiment score for the text. The input is the parsed text data, and the output is the sentiment score.

[0465] Step 5:

[0466] The server determines the likelihood of misinformation or propaganda based on the sentiment score. If the sentiment score exceeds a certain threshold, the post is identified as misinformation or propaganda. The input is the sentiment score, and the output is the results of identifying the post as misinformation.

[0467] Step 6:

[0468] The server displays a warning when false information or propaganda is identified. Specifically, it sends and displays a warning message in real time on the display of a smart device (such as smart glasses). The input is the result of identifying false information, and the output is the display of a warning message.

[0469] Step 7:

[0470] Users can check the warnings displayed on their smart devices and take measures to prevent the effects of disinformation and propaganda. Specific actions include receiving the alert and checking reliable sources of information. The input is the warning message, and the output is the user's response.

[0471] 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.

[0472] The present invention relates to a social media monitoring system for identifying false information and propaganda and implementing countermeasures. Furthermore, by combining an emotion engine that recognizes user emotions, the system is equipped with the function of assessing the impact of false information and propaganda and adjusting the warning content. The system operates mainly on a server and includes the following main means:

[0473] System Overview

[0474] 1. Data collection methods:

[0475] The server collects post data from the social media API, including metadata such as the post content (text), poster information, and post date and time.

[0476] 2. Data cleansing methods:

[0477] The server cleanses the collected data by removing unnecessary elements such as links, user tags, and hashtags, resulting in text data suitable for analysis.

[0478] 3. Data analysis methods:

[0479] The server uses the cleansed data to perform sentiment analysis and content validation, using libraries such as TextBlob to assess the text's emotional state as positive, negative, or neutral.

[0480] 4. Detecting Disinformation:

[0481] Based on the analysis results, the server sets thresholds for positive and negative scores to identify posts that are likely to contain false information or propaganda, and displays a warning if the negative score exceeds a certain level.

[0482] 5. Warning display means:

[0483] If the server detects false information or propaganda, it will display or notify the user of the post's ID and reason as a warning.

[0484] 6. Emotion Engine:

[0485] The server uses an emotion engine to recognize users' emotions and assess the impact of misinformation and propaganda, analyzing text inputs and behavioral data when users interact with the system.

[0486] 7. Adjustment of warning content:

[0487] The server adjusts the warning content based on the results of the emotion engine, for example, strengthening the warning if the user is already emotionally affected, and appropriately adjusting the warning content to avoid excessively negative impact.

[0488] 8. Recording and statistical information generation means:

[0489] The server will record the results of detecting false information and propaganda and generate statistical information based on this information, which will be useful for analyzing trends in the spread of false information and developing effective countermeasures.

[0490] Explaining program processing in natural language

[0491] 1. Data Collection:

[0492] The server collects the latest post data from the social media API, sends an HTTP request to the API endpoint, and parses the JSON data received as a response.

[0493] 2. Data cleansing:

[0494] The server uses regular expressions to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data. By cleansing, only the text is extracted and formatted for analysis.

[0495] 3. Data Analysis:

[0496] The server performs sentiment analysis on the cleansed text data, using libraries such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0497] 4. Detecting Disinformation:

[0498] Based on the sentiment score, the server determines that there is a high possibility of misinformation or propaganda if the negative score exceeds a predetermined threshold.

[0499] 5. User emotion recognition using emotion engine:

[0500] The server analyzes the user's input and behavioral data and uses an emotion engine to recognize the user's emotional state, which can include joy, sadness, anger, surprise, etc.

[0501] 6. Adjustment of warning content:

[0502] The server adjusts the warning content appropriately based on the analysis results of the emotion engine. For example, if the user is already in a negative emotional state, the server displays a warning message that is careful not to further exacerbate the negative emotional state.

[0503] 7. Warning display:

[0504] The server issues a warning to posts that contain false information or propaganda, including the post's ID and the reason for the detection, and reflects the evaluation results of the emotion engine.

[0505] 8. Recording and Statistics Generation:

[0506] The server records the results of detecting false information and propaganda and generates statistical information that includes the results of the emotion engine, which allows for analysis of the spread of false information and countermeasures based on the emotional state of users.

[0507] Specific examples

[0508] For example, suppose your server collects the following post data from an API:

[0509] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[0510] The server cleanses this submission and formats it as follows:

[0511] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[0512] The server then performs sentiment analysis on this cleansed text data to detect high negative scores, and if the negative score exceeds a certain threshold, the server identifies the post as misinformation or propaganda.

[0513] The server uses an emotion engine to recognize the user's emotional state based on the user's behavioral data. For example, if the user has seen many negative posts in the past 24 hours, the emotion engine will recognize that the user is in a negative emotional state.

[0514] Based on the results, the server generates a warning message "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and notifies the user after adjusting the warning content.

[0515] In this way, the present invention is a system that curbs the spread of disinformation and propaganda, taking into account the emotional state of the user and taking appropriate measures.

[0516] The processing flow will be explained below.

[0517] Step 1:

[0518] The server sends an HTTP GET request to the specified social media API URL, collects post data from the API, and receives JSON formatted data as a response.

[0519] Step 2:

[0520] The server parses the JSON data obtained from the response and extracts metadata such as the content (text) of each post, poster information, and posting date and time.

[0521] Step 3:

[0522] The server cleanses the text data of each extracted post. Specifically, it removes links (URLs) using regular expressions, and also removes user tags (@username) and hashtags using regular expressions. By cleansing, the text data becomes suitable for analysis.

[0523] Step 4:

[0524] The server performs sentiment analysis on the cleansed text data, using libraries such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0525] Step 5:

[0526] The server evaluates the resulting sentiment score and determines that a post is likely to be misinformation or propaganda, especially if the negative score exceeds a predefined threshold, specifically if the negative score is less than -0.5.

[0527] Step 6:

[0528] The server analyzes user input and behavioral data and uses an emotion engine to recognize the user's emotional state. For example, if a user has viewed many negative posts in the past 24 hours, the emotion engine will determine that the user is in a negative emotional state.

[0529] Step 7:

[0530] The server adjusts the warning content based on the emotion recognition results. If the user is already in a negative emotional state, the warning will be strengthened. The server also adjusts the warning content appropriately to avoid excessively negative impact.

[0531] Step 8:

[0532] If misinformation or propaganda is detected, the server generates a warning message containing the post's ID and the reason for the detection (that it was determined to have a high negative score).

[0533] Step 9:

[0534] The server notifies the user of the generated warning message by displaying the warning message on the console or by sending a notification email to the administrator.

[0535] Step 10:

[0536] The server records the results of detecting false information and propaganda and generates statistical information that includes the results of the emotion engine. This statistical information is useful for taking measures based on the spread of false information and the emotional state of users.

[0537] As a concrete example, suppose a server collects a post with the title "Breaking news! Government is lying! Do not trust anyone! FakeNews." After cleansing this post, the resulting text is formatted as "Breaking news! Government is lying! Do not trust anyone! FakeNews." Sentiment analysis determines that this text has a high negative score and is misinformation or propaganda. If the emotion engine subsequently determines that the user is in a negative emotional state, the server generates a warning message such as "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" to notify the user.

[0538] Example 2

[0539] 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."

[0540] In today's social media, false information and propaganda often spread quickly and have a significant impact on society. However, current systems are insufficient in detecting and dealing with false information, and do not take appropriate measures that take into account the emotional state of users. Therefore, it is necessary to solve the problem of quickly detecting false information and propaganda and minimizing their emotional impact on users.

[0541] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting social media data, a means for cleansing the collected data, a means for analyzing the cleansed data and detecting false information or propaganda, a means for displaying a warning when false information or propaganda is detected, a means for analyzing user behavior data and recognizing the user's emotional state using an emotion engine, and a means for adjusting the content of the warning based on the results of the emotion engine. This enables the rapid detection of false information and propaganda and the display of an appropriate warning that takes the user's emotional state into consideration.

[0542] "Data collection means" refers to a device or program that uses a social media API to obtain data such as the content of posts, poster information, and posting date and time.

[0543] A "data cleansing tool" is a device or program that removes unnecessary elements such as links, user tags, and hashtags from collected data and formats it into a form suitable for analysis.

[0544] The "data analysis means" is a device or program that performs sentiment analysis and content verification on the cleansed data and calculates a positive, negative, or neutral sentiment score.

[0545] A "fake news detection tool" is a device or program that, based on the results of data analysis, identifies posts that may be fake or propaganda based on a set threshold.

[0546] A "warning display means" is a device or program that generates and displays a message or alert to notify the user when false information or propaganda is detected.

[0547] An "emotion engine" is an algorithm or device that analyzes a user's text input and behavioral data to recognize the user's emotional state.

[0548] The "alert content adjustment means" is a device or program that optimizes the alert content based on the results of the emotion engine and taking into account the emotional state of the user.

[0549] The present invention relates to a social media monitoring system for identifying disinformation and propaganda and taking countermeasures. The system is server-based and includes the following main means:

[0550] The server first collects post data from the social media API. Specifically, it sends an HTTP request to the API endpoint and analyzes the JSON data received as a response. The collected data includes the post content (text), poster information, and posting date and time. This allows the server to store the most recent post data in a local database.

[0551] Next, the server uses a data cleansing method to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data using regular expressions, extracting only the text and preparing it in a form suitable for analysis.

[0552] The server then performs sentiment analysis on the cleansed data using a data analysis tool. For sentiment analysis, it uses a library such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text, which is then used to evaluate and record the emotional state of the post.

[0553] The server then uses a false information detection method based on the sentiment score to determine that a message is likely to be false information or propaganda if the negative score exceeds a predetermined threshold. If false information or propaganda is detected, the result is recorded.

[0554] The system also analyzes user behavior data and uses an emotion engine to recognize the user's emotional state. Based on the user's input and past behavioral data, the system identifies the user's emotional state. For example, if a user has seen many negative posts, the system will recognize the user's emotional state as negative.

[0555] Based on the analysis results of the emotion engine, the server uses the warning content adjustment means to appropriately adjust the warning content. For example, if the user is already in a negative emotional state, the server may take measures such as changing the warning text to a softer one.

[0556] Finally, for posts that contain false information or propaganda, the server issues a warning using a warning display method. The warning includes the post's ID and the reason for the detection, and reflects the evaluation results of the emotion engine. This allows the server to notify users of an appropriate warning.

[0557] To illustrate this system, here is an example prompt:

[0558] Please explain how your system would perform sentiment analysis on the following post data, and if the negative score exceeds a certain threshold, display a warning about misinformation or propaganda and adjust the warning appropriately. Post data: 'Breaking news! Government is lying! Do not trust anyone! FakeNews'

[0559] In this way, the present invention can effectively curb the spread of false information and propaganda, taking into account the emotional state of the user and taking appropriate measures.

[0560] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0561] Step 1: Data collection

[0562] The server collects post data from the social media API. Specifically, it sends an HTTP GET request to the API endpoint and analyzes the JSON data received as a response. This process obtains metadata such as the post content (text), poster information, and posting date and time as input, and stores them in a local database.

[0563] Step 2: Data cleansing

[0564] The server uses regular expressions to remove unnecessary elements such as links (URLs), user tags (@username), and hashtags from the collected JSON data. The input to this process is the collected post data, and the output is cleansed text data. Specifically, it applies regular expression patterns to remove unnecessary strings.

[0565] Step 3: Data analysis

[0566] The server performs sentiment analysis on the cleansed text data as input. It uses a library such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text. The output of this process is a sentiment score for each post. Specifically, it calls a library to analyze the text and calculate the sentiment score.

[0567] Step 4: Detecting misinformation

[0568] The server takes the sentiment score as input and checks whether the negative score exceeds a predetermined threshold. If the negative score exceeds the threshold, the post is deemed to be false information or propaganda. The output of this process is a list of posts that have been deemed to be false information. Specifically, the sentiment score is compared with the threshold, and posts that meet the criteria are added to the list.

[0569] Step 5: User Emotion Recognition

[0570] The server collects user behavior data and uses an emotion engine to recognize the user's emotional state. The input is the user's text input and behavior log, and the output is the user's emotional state (joy, sadness, anger, etc.). Specifically, the server analyzes the user's behavior data and identifies the user's emotional state through the emotion engine.

[0571] Step 6: Adjust the warning content

[0572] The server uses the results of the emotion engine as input and adjusts the warning content appropriately. For example, if the user is already in a negative emotional state, the warning text is changed to a softer expression. The output of this process is an adjusted warning message. Specifically, the content of the warning text is generated and changed based on the user's emotional state.

[0573] Step 7: Warning Display

[0574] The server issues a warning for posts that are determined to be false. The input is the post that needs a warning and the adjusted warning message, and the output is a warning that is displayed to the user. Specifically, the generated warning message is displayed through an interface to notify the user.

[0575] Step 8: Recording and generating statistics

[0576] The server records the results of the detection of false information and propaganda and the results of the emotion engine, and generates statistical information. The detection results and the emotion engine results are used as input, and the output is statistical information. Specifically, statistical analysis is performed based on this data, and a report is created to analyze the spread of false information and effective countermeasures.

[0577] (Application example 2)

[0578] 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."

[0579] In modern society, the spread of false information and propaganda through social media has become a major problem. This can lead to users being influenced by false information, causing anxiety and confusion. Furthermore, existing systems only detect false information and do not display warnings that take into account the user's emotional state, which can cause excessive stress. To solve this situation, a system that not only detects false information but also takes into account the user's emotional state and displays appropriate warnings is needed.

[0580] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0581] In this invention, the server includes a data collection means for identifying false information and propaganda, a means for cleansing the collected data, a means for analyzing the cleansed data and detecting false information and propaganda, a means for displaying a warning when false information and propaganda are detected, a means for processing the collected data in real time, a means for analyzing the emotional state of the user, and a means for adjusting the content of the warning based on the emotional state of the user, thereby enabling appropriate detection of false information and propaganda and display of a warning that takes into account the emotional state of the user.

[0582] "Data collection means" refers to the means used to collect posting data and metadata from social media.

[0583] A "cleansing method" is a method for removing unnecessary elements from collected data and preparing text data in a form suitable for analysis.

[0584] The "data analysis means" is a means for analyzing the cleansed text data and determining sentiment scores and false information.

[0585] "Disinformation and propaganda detection measures" are measures that identify posts that are likely to be disinformation or propaganda based on the results of data analysis.

[0586] "Warning display measures" are measures that display a warning on a post when false information or propaganda is detected.

[0587] "Real-time processing means" refers to means for updating and analyzing collected data in real time and responding immediately.

[0588] The "emotional state analysis means" is a means for analyzing the user's emotional state based on the user's past behavioral data and text input.

[0589] The "warning content adjustment means" is a means for adjusting the warning content according to the emotional state of the user based on the result of the emotional state analysis means.

[0590] This paper describes an embodiment of the present invention, which describes a system for identifying false information and propaganda and displaying warnings based on a user's emotional state. The system includes means for data collection, data cleansing, data analysis, false information detection, emotional state analysis, warning content adjustment, and warning display.

[0591] Data collection methods

[0592] The server collects post data from the social media API. Specifically, it sends an HTTP request to the API endpoint and analyzes the JSON data received as a response. This process uses a communication library such as the requests library.

[0593] Data Cleansing Methods

[0594] The server uses regular expressions to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected data, thereby obtaining text data suitable for analysis.

[0595] Data Analysis Methods

[0596] The server performs sentiment analysis on the cleansed text data, using a natural language processing library such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0597] Measures to detect disinformation and propaganda

[0598] The server determines the likelihood of misinformation or propaganda if the negative score exceeds a predetermined threshold based on the sentiment score. This determination is made using a threshold to determine whether a warning should be sent to the user.

[0599] Emotional state analysis means

[0600] The server analyzes user input and behavioral data and uses an emotion engine to recognize the user's emotional state. Based on the user's behavioral data within the past 24 hours, it counts the number of negative posts and evaluates the user's current emotional state.

[0601] Warning content adjustment means

[0602] The server adjusts the warning content based on the results of the emotional state analysis means. For example, if the user is already in a negative emotional state, the server displays a warning message that takes care not to further exacerbate the user's negative emotional state.

[0603] Warning display means

[0604] The server issues a warning to posts that contain false information or propaganda, including the post's ID and the reason for the detection, and reflects the evaluation results of the emotion engine.

[0605] Specific examples

[0606] For example, if your server collects the following post data from an API:

[0607] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[0608] The server cleanses this submission and formats it as follows:

[0609] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[0610] The server then performs sentiment analysis on this cleansed text data to detect high negative scores, and if the negative score exceeds a certain threshold, the server identifies the post as misinformation or propaganda.

[0611] The server uses an emotion engine to recognize the user's emotional state based on the user's behavioral data. For example, if the user has seen many negative posts in the past 24 hours, the emotion engine will recognize that the user is in a negative emotional state.

[0612] Based on the results, the server generates a warning message "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and notifies the user after adjusting the warning content.

[0613] Prompt Sentence Examples

[0614] Example prompts to apply to your API endpoints:

[0615] "Please provide the latest posts data from the social media API endpoint: 'https: / / api.socialmedia.com / latest_posts' additionally, gather the user's past 24-hour activity data for sentiment analysis: [ {"text": "I'm so sad about the news.", "timestamp": "2023-10-20T14:00:00Z"}, {"text": "Angry about the government's actions.", "timestamp": "2023-10-20T16:00:00Z"}]

[0616] In this way, the present invention makes it possible to curb the spread of false information and propaganda and to take appropriate measures taking into account the emotional state of the user.

[0617] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0618] Processing flow

[0619] Step 1: Data collection

[0620] The server sends an HTTP request to the social media API to collect the latest post data, and receives a JSON response from the API endpoint.

[0621] Input: API endpoint

[0622] Output: Post data (JSON format)

[0623] Step 2: Data cleansing

[0624] The server removes unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data using regular expressions.

[0625] Input: Post data (JSON format)

[0626] Output: Cleansed text data

[0627] Step 3: Data analysis

[0628] The server performs sentiment analysis on the cleansed text data, using a natural language processing library such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0629] Input: Cleansed text data

[0630] Output: Sentiment score (positive, negative, neutral)

[0631] Step 4: Detecting disinformation and propaganda

[0632] The server determines that the likelihood of the information being false information or propaganda is high if the negative score exceeds a predetermined threshold based on the emotion score.

[0633] Input: Sentiment score

[0634] Output: Disinformation and propaganda judgement result (True / False)

[0635] Step 5: Analyze emotional state

[0636] The server analyzes the user's behavioral data and uses an emotion engine to recognize the user's emotional state, for example, by counting the number of negative posts the user has made in the past 24 hours.

[0637] Input: User behavior data (post data from the past 24 hours)

[0638] Output: User's emotional state (number of negative comments)

[0639] Step 6: Adjust the warning content

[0640] The server adjusts the warning content based on the results of the emotional state analysis. If the user is already in a negative emotional state, the server generates a warning message that is careful not to further upset the user.

[0641] Input: User's emotional state, misinformation and propaganda judgment results

[0642] Output: Adjusted warning content

[0643] Step 7: View warnings

[0644] The server issues a warning to users when a post contains false information or propaganda, and notifies them. The warning includes the post's ID, the reason for the detection, and the evaluation result by the emotion engine.

[0645] Input: Adjusted warning content, post ID

[0646] Output: Display of warning message

[0647] This process allows the server to effectively detect false information and propaganda and display warnings in real time that take into account the user's emotional state.

[0648] 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.

[0649] 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.

[0650] 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.

[0651] [Third embodiment]

[0652] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0653] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0654] 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).

[0655] 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.

[0656] 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.

[0657] 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).

[0658] 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.

[0659] 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.

[0660] 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.

[0661] 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.

[0662] 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.

[0663] 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."

[0664] The present invention relates to a social media monitoring system for identifying disinformation and propaganda and taking countermeasures. Specifically, the system provides a server that collects social media data, cleanses the collected data, analyzes the cleansed data, detects disinformation and propaganda, and displays warnings or notifications based on the detection results.

[0665] System Overview

[0666] 1. Data collection methods:

[0667] The server collects post data from the social media API, including metadata such as the post content (text), poster information, and post date and time.

[0668] 2. Data cleansing methods:

[0669] The server cleanses the collected data by removing unnecessary elements such as links, user tags, and hashtags, resulting in text data suitable for analysis.

[0670] 3. Data analysis methods:

[0671] The server uses the cleansed data to perform sentiment analysis and content validation, using libraries such as TextBlob to assess the text's emotional state as positive, negative, or neutral.

[0672] 4. Detecting Disinformation:

[0673] Based on the analysis results, the server sets thresholds for positive and negative scores to identify posts that are likely to contain false information or propaganda, and displays a warning if the negative score exceeds a certain level.

[0674] 5. Warning display means:

[0675] If the server detects false information or propaganda, it will display or notify the user of the post's ID and reason as a warning.

[0676] Explaining program processing in natural language

[0677] 1. Data Collection:

[0678] The server collects the latest post data from the social media API, sends an HTTP request to the API endpoint, and parses the JSON data received as a response.

[0679] 2. Data cleansing:

[0680] The server uses regular expressions to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data. By cleansing, only the text is extracted and formatted for analysis.

[0681] 3. Data Analysis:

[0682] The server performs sentiment analysis on the cleansed text data, using libraries such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0683] 4. Detecting Disinformation:

[0684] Based on the calculated sentiment score, if the negative score is above a certain level, the server determines that the post is likely to be false information or propaganda.

[0685] 5. Warning display:

[0686] The server will issue a warning for posts that contain false information or propaganda, and will notify the user with the post ID and the reason for the detection, for example by displaying a warning message on the console or sending a notification email to an administrator.

[0687] Specific examples

[0688] For example, suppose your server collects the following post data from an API:

[0689] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[0690] The server cleanses this submission and formats it as follows:

[0691] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[0692] The server then performs sentiment analysis on this cleansed text data to detect high negative scores. If the negative score exceeds a certain threshold, the server determines that the post is misinformation or propaganda and displays a warning to the user:

[0693] "Post ID: 12345 - Alert: Detected potential misinformation or propaganda"

[0694] In this way, the present invention is a system that suppresses the spread of false information and propaganda and supports decision-making based on accurate information.

[0695] The processing flow will be explained below.

[0696] Step 1:

[0697] The server sends an HTTP GET request to the specified social media API URL, collects post data from the API, and receives JSON formatted data as a response.

[0698] Step 2:

[0699] The server parses the JSON data obtained from the response and extracts metadata such as the content (text) of each post, poster information, and posting date and time.

[0700] Step 3:

[0701] The server cleanses the text data of each extracted post by removing links (URLs) using regular expressions, removing user tags (@username) and hashtags using regular expressions, and finally, converting redundant whitespace into a single space.

[0702] Step 4:

[0703] The server performs sentiment analysis on the cleansed text data, using libraries such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0704] Step 5:

[0705] The server evaluates the resulting sentiment score and determines that a post is likely to be misinformation or propaganda, especially if the negative score exceeds a predefined threshold, specifically if the negative score is less than -0.5.

[0706] Step 6:

[0707] If misinformation or propaganda is detected, the server generates a warning message containing the post's ID and the reason for the detection (a high negative score).

[0708] Step 7:

[0709] The server notifies the user of the generated warning message by displaying the warning message on the console or by sending a notification email to the administrator.

[0710] For example, if the server collects a post that reads "The government is lying. Don't trust anyone!", the cleansed text will be "The government is lying. Don't trust anyone!". Sentiment analysis will determine that this text has a high negative score and is misinformation or propaganda. The server will generate a warning message that reads "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and notify the user.

[0711] Example 1

[0712] 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."

[0713] In modern society, social media has become a major means of communication, but at the same time, the spread of false information and propaganda has become a social problem. This false information and propaganda originates from unreliable sources and spreads rapidly, damaging trust in society. There is a need for a system that can solve this problem and provide accurate and reliable information.

[0714] 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.

[0715] In this invention, the server includes means for collecting social media data, means for cleansing the collected data, means for analyzing the cleansed data to detect false information and propaganda, means for performing sentiment analysis, means for displaying a warning when false information and propaganda are detected, and means for sending a notification. This enables early detection of false information and propaganda and notification thereof, thereby enabling the provision of highly reliable information.

[0716] "Social media" refers to online services that allow people to share, exchange, and communicate with each other via the Internet.

[0717] "Data collection means" refers to the functions and devices used to obtain information from social media.

[0718] "Data cleansing means" refers to a function or device for organizing collected information and removing unnecessary elements.

[0719] "Data analysis means" refers to a function or device for analyzing cleansed information and extracting specific patterns or features.

[0720] "Disinformation" refers to information that is intentionally or falsely stated and that is not factually correct.

[0721] "Propaganda" refers to information that is spread with a specific intention or purpose, and that is not necessarily factual.

[0722] "Sentiment analysis means" refers to a function or device for assessing the emotional state (positive, negative, neutral) of text data.

[0723] "Warning display means" means a feature or device that provides a visual or other warning when disinformation or propaganda is detected.

[0724] "Notification sending means" refers to a function or device for sending a notification to the outside when a specific condition is met.

[0725] The present invention relates to a social media monitoring system for identifying and combating false information and propaganda. Specifically, the system provides a server that collects social media data, cleanses the collected data, analyzes the cleansed data, detects false information and propaganda, and displays warnings or notifications based on the detection results.

[0726] First, the server collects the latest post data through the social media API. For example, when using the Twitter API, the server sends an HTTP request to the appropriate API endpoint and analyzes the JSON data received as a response. This data includes metadata such as the post content (text), poster information, and the post date and time.

[0727] Next, the server uses regular expressions to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data. This process extracts only the text and formats it into a form suitable for analysis. A specific example of cleansing is a method of removing links and tags using Python's regular expression library.

[0728] The server then performs sentiment analysis on the cleansed text data. For sentiment analysis, it uses a natural language processing library such as TextBlob. Using TextBlob, it is possible to calculate a positive, negative, or neutral sentiment score for the text. For example, it calculates that the negative score for the text "Breaking news! Government is lying! Do not trust anyone! FakeNews" is -0.8.

[0729] Next, the server makes a judgment to detect misinformation or propaganda based on the sentiment score. If the negative score exceeds a certain threshold, the post is judged to be misinformation or propaganda. Based on this judgment result, the server issues a warning for posts in which misinformation or propaganda has been detected. The warning is notified to the user with the post ID and the reason for the detection. For example, a warning message such as "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" can be generated and a notification email can be sent to the administrator.

[0730] As a concrete example, a server collects the post "Breaking news! Government is lying! Do not trust anyone! FakeNews" from the Twitter API, cleanses it, and formats it as "Breaking news! Government is lying! Do not trust anyone! FakeNews." Next, it performs sentiment analysis using TextBlob and detects a high negative score, determining that the post is misinformation or propaganda. As a result, the server generates a warning "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and notifies the user.

[0731] In this way, the present invention realizes a system that suppresses the spread of false information and propaganda and provides accurate information.

[0732] Example prompts for generative AI models

[0733] "Use Twitter's API to remove links and hashtags from posts collected, and use TextBlob to perform sentiment analysis. Generate warning messages for posts with high negative scores."

[0734] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0735] Step 1:

[0736] The server collects post data from social media APIs. The server sends an HTTP request to the API endpoint and receives JSON format data as a response. The input is a request that includes an API key and query parameters, and the output is an array of the received post data. As a specific example, the Twitter API endpoint "statuses / user_timeline" is used to obtain the latest posts from a specific user.

[0737] Step 2:

[0738] The server cleanses the received post data. It uses regular expressions to remove links (URLs), user tags (@username), and hashtags. This cleansing process passes the received JSON data post content as input, and outputs text data with unnecessary elements removed. As a specific example of how it works, the Python regular expression library is used to format "Breaking news! Government is lying! Do not trust anyone! FakeNews" to "Breaking news! Government is lying! Do not trust anyone! FakeNews".

[0739] Step 3:

[0740] The server performs sentiment analysis on the cleansed text data. It uses a natural language processing library such as TextBlob to calculate a sentiment score. It receives the cleansed text as input and outputs a positive, negative, or neutral sentiment score. As a specific example, TextBlob is used to calculate a negative score of -0.8 for "Breaking news! Government is lying! Do not trust anyone! FakeNews."

[0741] Step 4:

[0742] The server detects false information and propaganda based on the sentiment score. The server compares the sentiment score with a set negative score threshold to identify posts that are likely to be false information or propaganda. The server receives the sentiment score as input and obtains posts determined to be false information as output. As a specific example of operation, if the negative score is -0.5 or less, the server determines that the post is false information.

[0743] Step 5:

[0744] The server displays a warning if misinformation or propaganda is detected. The server generates a warning message including the detected post ID, username, and reason for detection, and notifies the user. The input is the detected post information, and the output is the warning message. As a specific example of operation, the server generates a warning message saying "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and sends a notification email to the user.

[0745] (Application example 1)

[0746] 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."

[0747] This invention aims to solve the technical challenges of detecting and quickly addressing disinformation and propaganda spreading on social media. In particular, there is a need to collect and analyze information in real time, visualize it, and provide prompt warnings. However, current systems have difficulty integrating multiple data collection methods and complex processes such as speech recognition, text analysis, and sentiment analysis.

[0748] 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.

[0749] In this invention, the server includes means for collecting social media information to identify false information and propaganda, means for cleansing the collected information, means for analyzing the cleansed information to detect false information and propaganda, means for displaying a warning when false information and propaganda are detected, means for collecting the latest media information through speech recognition and text analysis, means for performing sentiment analysis based on the speech recognition and text analysis, and means for displaying a warning on the display of the smart device. This makes it possible to quickly mitigate the impact of false information and propaganda by evaluating the authenticity of information in real time and displaying a warning.

[0750] "Disinformation" is information that is not based on facts and is spread intentionally or unintentionally with the intent to mislead or confuse.

[0751] "Propaganda" is a method of selectively providing or manipulating information with a specific political, social, or religious intent, and is used to influence people's opinions and actions.

[0752] "Social media" refers to a platform that allows users to share, exchange, and disseminate information via the Internet, and a typical example is SNS (social networking service).

[0753] "Information collection means" refers to the means for collecting post data and metadata sent on social media, and is realized using technologies such as APIs.

[0754] "Cleansing methods" are methods used to remove unnecessary data (e.g., links, user tags, hashtags) from collected information and prepare it in a format suitable for analysis.

[0755] "Analysis methods" are means for detecting false information and propaganda by conducting sentiment analysis and content verification on cleansed information.

[0756] "Warning display means" refers to means for displaying or notifying users of false information or propaganda as a warning when such information is detected.

[0757] "Speech recognition and text analysis means" refers to means for converting speech data into text data and analyzing the text data, and extracting information using natural language processing techniques.

[0758] "Sentiment analysis means" is a means for evaluating the emotional state (positive, negative, neutral) of text data, and analyzes the credibility of information by assigning an emotional score.

[0759] "Smart device display means" means a means for visually displaying important information or warnings to a user on smart glasses or other wearable devices.

[0760] This invention relates to a system for identifying and countering false information and propaganda, and specifically includes the following means:

[0761] The server first uses social media APIs to collect post data and related metadata, then uses a cleansing method to remove unnecessary information such as links, user tags, and hashtags from the collected data, converting it into text suitable for analysis.

[0762] Next, speech recognition and text analysis tools are used to capture audio and video data from smart devices (e.g., smart glasses) and convert them into text data. OpenAI's speech recognition API is used here.

[0763] The cleansed text data is then evaluated using sentiment analysis tools. Natural language processing libraries such as the TextBlob library and Hugging Face's Transformers are used to calculate a sentiment score for the text. If the sentiment score exceeds a certain threshold, the text is marked as misinformation or propaganda.

[0764] When disinformation or propaganda is detected, a warning will be displayed on the smart device's display, including the ID of the detected information and the reason for the detection, and the user will be visually notified. For example, a warning message will be displayed in real time on the display of smart glasses.

[0765] Consider the following scenario: Smart glasses recognize the speech phrase "Breaking news! The government is lying!" and convert it into text. Sentiment analysis of the text detects a high negative score, identifying it as disinformation or propaganda, and a warning is displayed on the glasses' display.

[0766] In addition to cleansing and sentiment analysis of the generated data, the following example prompts can be used to feed the generative AI model:

[0767] Example prompt sentence:

[0768] Transcribe the speech "Breaking news! The government is lying! Trust no one!" and perform sentiment analysis on this text. If the negative score is high, show a warning to the user.

[0769] This allows users to make informed decisions in real time, quickly reducing the impact of disinformation and propaganda.

[0770] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0771] Step 1:

[0772] The server uses the social media API to collect the latest post data and related metadata. Specifically, it sends an HTTP request to the API endpoint and parses the JSON data received as a response. The input is data from the API, and the output is the parsed post data and metadata.

[0773] Step 2:

[0774] The server cleanses the collected post data by using regular expressions to remove links (URLs), user tags (@username), hashtags, etc. The input is the analyzed post data, and the output is cleansed text data with unnecessary information removed.

[0775] Step 3:

[0776] The server evaluates the cleansed text data using speech recognition and text analysis. First, the voice data is converted into text data using OpenAI's speech recognition API, and then the converted text data is analyzed. The input is the voice data and the cleansed text data, and the output is the analyzed text data.

[0777] Step 4:

[0778] The server performs sentiment analysis on the parsed text data, using the TextBlob library and Hugging Face Transformers to calculate a positive, negative, or neutral sentiment score for the text. The input is the parsed text data, and the output is the sentiment score.

[0779] Step 5:

[0780] The server determines the likelihood of misinformation or propaganda based on the sentiment score. If the sentiment score exceeds a certain threshold, the post is identified as misinformation or propaganda. The input is the sentiment score, and the output is the results of identifying the post as misinformation.

[0781] Step 6:

[0782] The server displays a warning when false information or propaganda is identified. Specifically, it sends and displays a warning message in real time on the display of a smart device (such as smart glasses). The input is the result of identifying false information, and the output is the display of a warning message.

[0783] Step 7:

[0784] Users can check the warnings displayed on their smart devices and take measures to prevent the effects of disinformation and propaganda. Specific actions include receiving the alert and checking reliable sources of information. The input is the warning message, and the output is the user's response.

[0785] 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.

[0786] The present invention relates to a social media monitoring system for identifying false information and propaganda and implementing countermeasures. Furthermore, by combining an emotion engine that recognizes user emotions, the system is equipped with the function of assessing the impact of false information and propaganda and adjusting the warning content. The system operates mainly on a server and includes the following main means:

[0787] System Overview

[0788] 1. Data collection methods:

[0789] The server collects post data from the social media API, including metadata such as the post content (text), poster information, and post date and time.

[0790] 2. Data cleansing methods:

[0791] The server cleanses the collected data by removing unnecessary elements such as links, user tags, and hashtags, resulting in text data suitable for analysis.

[0792] 3. Data analysis methods:

[0793] The server uses the cleansed data to perform sentiment analysis and content validation, using libraries such as TextBlob to assess the text's emotional state as positive, negative, or neutral.

[0794] 4. Detecting Disinformation:

[0795] Based on the analysis results, the server sets thresholds for positive and negative scores to identify posts that are likely to contain false information or propaganda, and displays a warning if the negative score exceeds a certain level.

[0796] 5. Warning display means:

[0797] If the server detects false information or propaganda, it will display or notify the user of the post's ID and reason as a warning.

[0798] 6. Emotion Engine:

[0799] The server uses an emotion engine to recognize users' emotions and assess the impact of misinformation and propaganda, analyzing text inputs and behavioral data when users interact with the system.

[0800] 7. Adjustment of warning content:

[0801] The server adjusts the warning content based on the results of the emotion engine, for example, strengthening the warning if the user is already emotionally affected, and appropriately adjusting the warning content to avoid excessively negative impact.

[0802] 8. Recording and statistical information generation means:

[0803] The server will record the results of detecting false information and propaganda and generate statistical information based on this information, which will be useful for analyzing trends in the spread of false information and developing effective countermeasures.

[0804] Explaining program processing in natural language

[0805] 1. Data Collection:

[0806] The server collects the latest post data from the social media API, sends an HTTP request to the API endpoint, and parses the JSON data received as a response.

[0807] 2. Data cleansing:

[0808] The server uses regular expressions to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data. By cleansing, only the text is extracted and formatted for analysis.

[0809] 3. Data Analysis:

[0810] The server performs sentiment analysis on the cleansed text data, using libraries such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0811] 4. Detecting Disinformation:

[0812] Based on the sentiment score, the server determines that there is a high possibility of misinformation or propaganda if the negative score exceeds a predetermined threshold.

[0813] 5. User emotion recognition using emotion engine:

[0814] The server analyzes the user's input and behavioral data and uses an emotion engine to recognize the user's emotional state, which can include joy, sadness, anger, surprise, etc.

[0815] 6. Adjustment of warning content:

[0816] The server adjusts the warning content appropriately based on the analysis results of the emotion engine. For example, if the user is already in a negative emotional state, the server displays a warning message that is careful not to further exacerbate the negative emotional state.

[0817] 7. Warning display:

[0818] The server issues a warning to posts that contain false information or propaganda, including the post's ID and the reason for the detection, and reflects the evaluation results of the emotion engine.

[0819] 8. Recording and Statistics Generation:

[0820] The server records the results of detecting false information and propaganda and generates statistical information that includes the results of the emotion engine, which allows for analysis of the spread of false information and countermeasures based on the emotional state of users.

[0821] Specific examples

[0822] For example, suppose your server collects the following post data from an API:

[0823] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[0824] The server cleanses this submission and formats it as follows:

[0825] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[0826] The server then performs sentiment analysis on this cleansed text data to detect high negative scores, and if the negative score exceeds a certain threshold, the server identifies the post as misinformation or propaganda.

[0827] The server uses an emotion engine to recognize the user's emotional state based on the user's behavioral data. For example, if the user has seen many negative posts in the past 24 hours, the emotion engine will recognize that the user is in a negative emotional state.

[0828] Based on the results, the server generates a warning message "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and notifies the user after adjusting the warning content.

[0829] In this way, the present invention is a system that curbs the spread of disinformation and propaganda, taking into account the emotional state of the user and taking appropriate measures.

[0830] The processing flow will be explained below.

[0831] Step 1:

[0832] The server sends an HTTP GET request to the specified social media API URL, collects post data from the API, and receives JSON formatted data as a response.

[0833] Step 2:

[0834] The server parses the JSON data obtained from the response and extracts metadata such as the content (text) of each post, poster information, and posting date and time.

[0835] Step 3:

[0836] The server cleanses the text data of each extracted post. Specifically, it removes links (URLs) using regular expressions, and also removes user tags (@username) and hashtags using regular expressions. By cleansing, the text data becomes suitable for analysis.

[0837] Step 4:

[0838] The server performs sentiment analysis on the cleansed text data, using libraries such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0839] Step 5:

[0840] The server evaluates the resulting sentiment score and determines that a post is likely to be misinformation or propaganda, especially if the negative score exceeds a predefined threshold, specifically if the negative score is less than -0.5.

[0841] Step 6:

[0842] The server analyzes user input and behavioral data and uses an emotion engine to recognize the user's emotional state. For example, if a user has viewed many negative posts in the past 24 hours, the emotion engine will determine that the user is in a negative emotional state.

[0843] Step 7:

[0844] The server adjusts the warning content based on the emotion recognition results. If the user is already in a negative emotional state, the warning will be strengthened. The server also adjusts the warning content appropriately to avoid excessively negative impact.

[0845] Step 8:

[0846] If misinformation or propaganda is detected, the server generates a warning message containing the post's ID and the reason for the detection (that it was determined to have a high negative score).

[0847] Step 9:

[0848] The server notifies the user of the generated warning message by displaying the warning message on the console or by sending a notification email to the administrator.

[0849] Step 10:

[0850] The server records the results of detecting false information and propaganda and generates statistical information that includes the results of the emotion engine. This statistical information is useful for taking measures based on the spread of false information and the emotional state of users.

[0851] As a concrete example, suppose a server collects a post with the title "Breaking news! Government is lying! Do not trust anyone! FakeNews." After cleansing this post, the resulting text is formatted as "Breaking news! Government is lying! Do not trust anyone! FakeNews." Sentiment analysis determines that this text has a high negative score and is misinformation or propaganda. If the emotion engine subsequently determines that the user is in a negative emotional state, the server generates a warning message such as "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" to notify the user.

[0852] Example 2

[0853] 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."

[0854] In today's social media, false information and propaganda often spread quickly and have a significant impact on society. However, current systems are insufficient in detecting and dealing with false information, and do not take appropriate measures that take into account the emotional state of users. Therefore, it is necessary to solve the problem of quickly detecting false information and propaganda and minimizing their emotional impact on users.

[0855] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting social media data, a means for cleansing the collected data, a means for analyzing the cleansed data and detecting false information or propaganda, a means for displaying a warning when false information or propaganda is detected, a means for analyzing user behavior data and recognizing the user's emotional state using an emotion engine, and a means for adjusting the content of the warning based on the results of the emotion engine. This enables the rapid detection of false information and propaganda and the display of an appropriate warning that takes the user's emotional state into consideration.

[0856] "Data collection means" refers to a device or program that uses a social media API to obtain data such as the content of posts, poster information, and posting date and time.

[0857] A "data cleansing tool" is a device or program that removes unnecessary elements such as links, user tags, and hashtags from collected data and formats it into a form suitable for analysis.

[0858] The "data analysis means" is a device or program that performs sentiment analysis and content verification on the cleansed data and calculates a positive, negative, or neutral sentiment score.

[0859] A "fake news detection tool" is a device or program that, based on the results of data analysis, identifies posts that may be fake or propaganda based on a set threshold.

[0860] A "warning display means" is a device or program that generates and displays a message or alert to notify the user when false information or propaganda is detected.

[0861] An "emotion engine" is an algorithm or device that analyzes a user's text input and behavioral data to recognize the user's emotional state.

[0862] The "alert content adjustment means" is a device or program that optimizes the alert content based on the results of the emotion engine and taking into account the emotional state of the user.

[0863] The present invention relates to a social media monitoring system for identifying disinformation and propaganda and taking countermeasures. The system is server-based and includes the following main means:

[0864] The server first collects post data from the social media API. Specifically, it sends an HTTP request to the API endpoint and analyzes the JSON data received as a response. The collected data includes the post content (text), poster information, and posting date and time. This allows the server to store the most recent post data in a local database.

[0865] Next, the server uses a data cleansing method to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data using regular expressions, extracting only the text and preparing it in a form suitable for analysis.

[0866] The server then performs sentiment analysis on the cleansed data using a data analysis tool. For sentiment analysis, it uses a library such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text, which is then used to evaluate and record the emotional state of the post.

[0867] The server then uses a false information detection method based on the sentiment score to determine that a message is likely to be false information or propaganda if the negative score exceeds a predetermined threshold. If false information or propaganda is detected, the result is recorded.

[0868] The system also analyzes user behavior data and uses an emotion engine to recognize the user's emotional state. Based on the user's input and past behavioral data, the system identifies the user's emotional state. For example, if a user has seen many negative posts, the system will recognize the user's emotional state as negative.

[0869] Based on the analysis results of the emotion engine, the server uses the warning content adjustment means to appropriately adjust the warning content. For example, if the user is already in a negative emotional state, the server may take measures such as changing the warning text to a softer one.

[0870] Finally, for posts that contain false information or propaganda, the server issues a warning using a warning display method. The warning includes the post's ID and the reason for the detection, and reflects the evaluation results of the emotion engine. This allows the server to notify users of an appropriate warning.

[0871] To illustrate this system, here is an example prompt:

[0872] Please explain how your system would perform sentiment analysis on the following post data, and if the negative score exceeds a certain threshold, display a warning about misinformation or propaganda and adjust the warning appropriately. Post data: 'Breaking news! Government is lying! Do not trust anyone! FakeNews'

[0873] In this way, the present invention can effectively curb the spread of false information and propaganda, taking into account the emotional state of the user and taking appropriate measures.

[0874] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0875] Step 1: Data collection

[0876] The server collects post data from the social media API. Specifically, it sends an HTTP GET request to the API endpoint and analyzes the JSON data received as a response. This process obtains metadata such as the post content (text), poster information, and posting date and time as input, and stores them in a local database.

[0877] Step 2: Data cleansing

[0878] The server uses regular expressions to remove unnecessary elements such as links (URLs), user tags (@username), and hashtags from the collected JSON data. The input to this process is the collected post data, and the output is cleansed text data. Specifically, it applies regular expression patterns to remove unnecessary strings.

[0879] Step 3: Data analysis

[0880] The server performs sentiment analysis on the cleansed text data as input. It uses a library such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text. The output of this process is a sentiment score for each post. Specifically, it calls a library to analyze the text and calculate the sentiment score.

[0881] Step 4: Detecting misinformation

[0882] The server takes the sentiment score as input and checks whether the negative score exceeds a predetermined threshold. If the negative score exceeds the threshold, the post is deemed to be false information or propaganda. The output of this process is a list of posts that have been deemed to be false information. Specifically, the sentiment score is compared with the threshold, and posts that meet the criteria are added to the list.

[0883] Step 5: User Emotion Recognition

[0884] The server collects user behavior data and uses an emotion engine to recognize the user's emotional state. The input is the user's text input and behavior log, and the output is the user's emotional state (joy, sadness, anger, etc.). Specifically, the server analyzes the user's behavior data and identifies the user's emotional state through the emotion engine.

[0885] Step 6: Adjust the warning content

[0886] The server uses the results of the emotion engine as input and adjusts the warning content appropriately. For example, if the user is already in a negative emotional state, the warning text is changed to a softer expression. The output of this process is an adjusted warning message. Specifically, the content of the warning text is generated and changed based on the user's emotional state.

[0887] Step 7: Warning Display

[0888] The server issues a warning for posts that are determined to be false. The input is the post that needs a warning and the adjusted warning message, and the output is a warning that is displayed to the user. Specifically, the generated warning message is displayed through an interface to notify the user.

[0889] Step 8: Recording and generating statistics

[0890] The server records the results of the detection of false information and propaganda and the results of the emotion engine, and generates statistical information. The detection results and the emotion engine results are used as input, and the output is statistical information. Specifically, statistical analysis is performed based on this data, and a report is created to analyze the spread of false information and effective countermeasures.

[0891] (Application example 2)

[0892] 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."

[0893] In modern society, the spread of false information and propaganda through social media has become a major problem. This can lead to users being influenced by false information, causing anxiety and confusion. Furthermore, existing systems only detect false information and do not display warnings that take into account the user's emotional state, which can cause excessive stress. To solve this situation, a system that not only detects false information but also takes into account the user's emotional state and displays appropriate warnings is needed.

[0894] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0895] In this invention, the server includes a data collection means for identifying false information and propaganda, a means for cleansing the collected data, a means for analyzing the cleansed data and detecting false information and propaganda, a means for displaying a warning when false information and propaganda are detected, a means for processing the collected data in real time, a means for analyzing the emotional state of the user, and a means for adjusting the content of the warning based on the emotional state of the user, thereby enabling appropriate detection of false information and propaganda and display of a warning that takes into account the emotional state of the user.

[0896] "Data collection means" refers to the means used to collect posting data and metadata from social media.

[0897] A "cleansing method" is a method for removing unnecessary elements from collected data and preparing text data in a form suitable for analysis.

[0898] The "data analysis means" is a means for analyzing the cleansed text data and determining sentiment scores and false information.

[0899] "Disinformation and propaganda detection measures" are measures that identify posts that are likely to be disinformation or propaganda based on the results of data analysis.

[0900] "Warning display measures" are measures that display a warning on a post when false information or propaganda is detected.

[0901] "Real-time processing means" refers to means for updating and analyzing collected data in real time and responding immediately.

[0902] The "emotional state analysis means" is a means for analyzing the user's emotional state based on the user's past behavioral data and text input.

[0903] The "warning content adjustment means" is a means for adjusting the warning content according to the emotional state of the user based on the result of the emotional state analysis means.

[0904] This paper describes an embodiment of the present invention, which describes a system for identifying false information and propaganda and displaying warnings based on a user's emotional state. The system includes means for data collection, data cleansing, data analysis, false information detection, emotional state analysis, warning content adjustment, and warning display.

[0905] Data collection methods

[0906] The server collects post data from the social media API. Specifically, it sends an HTTP request to the API endpoint and analyzes the JSON data received as a response. This process uses a communication library such as the requests library.

[0907] Data Cleansing Methods

[0908] The server uses regular expressions to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected data, thereby obtaining text data suitable for analysis.

[0909] Data Analysis Methods

[0910] The server performs sentiment analysis on the cleansed text data, using a natural language processing library such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0911] Measures to detect disinformation and propaganda

[0912] The server determines the likelihood of misinformation or propaganda if the negative score exceeds a predetermined threshold based on the sentiment score. This determination is made using a threshold to determine whether a warning should be sent to the user.

[0913] Emotional state analysis means

[0914] The server analyzes user input and behavioral data and uses an emotion engine to recognize the user's emotional state. Based on the user's behavioral data within the past 24 hours, it counts the number of negative posts and evaluates the user's current emotional state.

[0915] Warning content adjustment means

[0916] The server adjusts the warning content based on the results of the emotional state analysis means. For example, if the user is already in a negative emotional state, the server displays a warning message that takes care not to further exacerbate the user's negative emotional state.

[0917] Warning display means

[0918] The server issues a warning to posts that contain false information or propaganda, including the post's ID and the reason for the detection, and reflects the evaluation results of the emotion engine.

[0919] Specific examples

[0920] For example, if your server collects the following post data from an API:

[0921] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[0922] The server cleanses this submission and formats it as follows:

[0923] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[0924] The server then performs sentiment analysis on this cleansed text data to detect high negative scores, and if the negative score exceeds a certain threshold, the server identifies the post as misinformation or propaganda.

[0925] The server uses an emotion engine to recognize the user's emotional state based on the user's behavioral data. For example, if the user has seen many negative posts in the past 24 hours, the emotion engine will recognize that the user is in a negative emotional state.

[0926] Based on the results, the server generates a warning message "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and notifies the user after adjusting the warning content.

[0927] Prompt Sentence Examples

[0928] Example prompts to apply to your API endpoints:

[0929] "Please provide the latest posts data from the social media API endpoint: 'https: / / api.socialmedia.com / latest_posts' additionally, gather the user's past 24-hour activity data for sentiment analysis: [ {"text": "I'm so sad about the news.", "timestamp": "2023-10-20T14:00:00Z"}, {"text": "Angry about the government's actions.", "timestamp": "2023-10-20T16:00:00Z"}]

[0930] In this way, the present invention makes it possible to curb the spread of false information and propaganda and to take appropriate measures taking into account the emotional state of the user.

[0931] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0932] Processing flow

[0933] Step 1: Data collection

[0934] The server sends an HTTP request to the social media API to collect the latest post data, and receives a JSON response from the API endpoint.

[0935] Input: API endpoint

[0936] Output: Post data (JSON format)

[0937] Step 2: Data cleansing

[0938] The server removes unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data using regular expressions.

[0939] Input: Post data (JSON format)

[0940] Output: Cleansed text data

[0941] Step 3: Data analysis

[0942] The server performs sentiment analysis on the cleansed text data, using a natural language processing library such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0943] Input: Cleansed text data

[0944] Output: Sentiment score (positive, negative, neutral)

[0945] Step 4: Detecting disinformation and propaganda

[0946] The server determines that the likelihood of the information being false information or propaganda is high if the negative score exceeds a predetermined threshold based on the emotion score.

[0947] Input: Sentiment score

[0948] Output: Disinformation and propaganda judgement result (True / False)

[0949] Step 5: Analyze emotional state

[0950] The server analyzes the user's behavioral data and uses an emotion engine to recognize the user's emotional state, for example, by counting the number of negative posts the user has made in the past 24 hours.

[0951] Input: User behavior data (post data from the past 24 hours)

[0952] Output: User's emotional state (number of negative comments)

[0953] Step 6: Adjust the warning content

[0954] The server adjusts the warning content based on the results of the emotional state analysis. If the user is already in a negative emotional state, the server generates a warning message that is careful not to further upset the user.

[0955] Input: User's emotional state, misinformation and propaganda judgment results

[0956] Output: Adjusted warning content

[0957] Step 7: View warnings

[0958] The server issues a warning to users when a post contains false information or propaganda, and notifies them. The warning includes the post's ID, the reason for the detection, and the evaluation result by the emotion engine.

[0959] Input: Adjusted warning content, post ID

[0960] Output: Display of warning message

[0961] This process allows the server to effectively detect false information and propaganda and display warnings in real time that take into account the user's emotional state.

[0962] 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.

[0963] 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.

[0964] 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.

[0965] [Fourth embodiment]

[0966] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0967] 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.

[0968] 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).

[0969] 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.

[0970] 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.

[0971] 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).

[0972] 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.

[0973] 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.

[0974] 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.

[0975] 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.

[0976] 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.

[0977] 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.

[0978] 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."

[0979] The present invention relates to a social media monitoring system for identifying disinformation and propaganda and taking countermeasures. Specifically, the system provides a server that collects social media data, cleanses the collected data, analyzes the cleansed data, detects disinformation and propaganda, and displays warnings or notifications based on the detection results.

[0980] System Overview

[0981] 1. Data collection methods:

[0982] The server collects post data from the social media API, including metadata such as the post content (text), poster information, and post date and time.

[0983] 2. Data cleansing methods:

[0984] The server cleanses the collected data by removing unnecessary elements such as links, user tags, and hashtags, resulting in text data suitable for analysis.

[0985] 3. Data analysis methods:

[0986] The server uses the cleansed data to perform sentiment analysis and content validation, using libraries such as TextBlob to assess the text's emotional state as positive, negative, or neutral.

[0987] 4. Detecting Disinformation:

[0988] Based on the analysis results, the server sets thresholds for positive and negative scores to identify posts that are likely to contain false information or propaganda, and displays a warning if the negative score exceeds a certain level.

[0989] 5. Warning display means:

[0990] If the server detects false information or propaganda, it will display or notify the user of the post's ID and reason as a warning.

[0991] Explaining program processing in natural language

[0992] 1. Data Collection:

[0993] The server collects the latest post data from the social media API, sends an HTTP request to the API endpoint, and parses the JSON data received as a response.

[0994] 2. Data cleansing:

[0995] The server uses regular expressions to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data. By cleansing, only the text is extracted and formatted for analysis.

[0996] 3. Data Analysis:

[0997] The server performs sentiment analysis on the cleansed text data, using libraries such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[0998] 4. Detecting Disinformation:

[0999] Based on the calculated sentiment score, if the negative score is above a certain level, the server determines that the post is likely to be false information or propaganda.

[1000] 5. Warning display:

[1001] The server will issue a warning for posts that contain false information or propaganda, and will notify the user with the post ID and the reason for the detection, for example by displaying a warning message on the console or sending a notification email to an administrator.

[1002] Specific examples

[1003] For example, suppose your server collects the following post data from an API:

[1004] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[1005] The server cleanses this submission and formats it as follows:

[1006] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[1007] The server then performs sentiment analysis on this cleansed text data to detect high negative scores. If the negative score exceeds a certain threshold, the server determines that the post is misinformation or propaganda and displays a warning to the user:

[1008] "Post ID: 12345 - Alert: Detected potential misinformation or propaganda"

[1009] In this way, the present invention is a system that suppresses the spread of false information and propaganda and supports decision-making based on accurate information.

[1010] The processing flow will be explained below.

[1011] Step 1:

[1012] The server sends an HTTP GET request to the specified social media API URL, collects post data from the API, and receives JSON formatted data as a response.

[1013] Step 2:

[1014] The server parses the JSON data obtained from the response and extracts metadata such as the content (text) of each post, poster information, and posting date and time.

[1015] Step 3:

[1016] The server cleanses the text data of each extracted post by removing links (URLs) using regular expressions, removing user tags (@username) and hashtags using regular expressions, and finally, converting redundant whitespace into a single space.

[1017] Step 4:

[1018] The server performs sentiment analysis on the cleansed text data, using libraries such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[1019] Step 5:

[1020] The server evaluates the resulting sentiment score and determines that a post is likely to be misinformation or propaganda, especially if the negative score exceeds a predefined threshold, specifically if the negative score is less than -0.5.

[1021] Step 6:

[1022] If misinformation or propaganda is detected, the server generates a warning message containing the post's ID and the reason for the detection (a high negative score).

[1023] Step 7:

[1024] The server notifies the user of the generated warning message by displaying the warning message on the console or by sending a notification email to the administrator.

[1025] For example, if the server collects a post that reads "The government is lying. Don't trust anyone!", the cleansed text will be "The government is lying. Don't trust anyone!". Sentiment analysis will determine that this text has a high negative score and is misinformation or propaganda. The server will generate a warning message that reads "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and notify the user.

[1026] Example 1

[1027] 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."

[1028] In modern society, social media has become a major means of communication, but at the same time, the spread of false information and propaganda has become a social problem. This false information and propaganda originates from unreliable sources and spreads rapidly, damaging trust in society. There is a need for a system that can solve this problem and provide accurate and reliable information.

[1029] 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.

[1030] In this invention, the server includes means for collecting social media data, means for cleansing the collected data, means for analyzing the cleansed data to detect false information and propaganda, means for performing sentiment analysis, means for displaying a warning when false information and propaganda are detected, and means for sending a notification. This enables early detection of false information and propaganda and notification thereof, thereby enabling the provision of highly reliable information.

[1031] "Social media" refers to online services that allow people to share, exchange, and communicate with each other via the Internet.

[1032] "Data collection means" refers to the functions and devices used to obtain information from social media.

[1033] "Data cleansing means" refers to a function or device for organizing collected information and removing unnecessary elements.

[1034] "Data analysis means" refers to a function or device for analyzing cleansed information and extracting specific patterns or features.

[1035] "Disinformation" refers to information that is intentionally or falsely stated and that is not factually correct.

[1036] "Propaganda" refers to information that is spread with a specific intention or purpose, and that is not necessarily factual.

[1037] "Sentiment analysis means" refers to a function or device for assessing the emotional state (positive, negative, neutral) of text data.

[1038] "Warning display means" means a feature or device that provides a visual or other warning when disinformation or propaganda is detected.

[1039] "Notification sending means" refers to a function or device for sending a notification to the outside when a specific condition is met.

[1040] The present invention relates to a social media monitoring system for identifying and combating false information and propaganda. Specifically, the system provides a server that collects social media data, cleanses the collected data, analyzes the cleansed data, detects false information and propaganda, and displays warnings or notifications based on the detection results.

[1041] First, the server collects the latest post data through the social media API. For example, when using the Twitter API, the server sends an HTTP request to the appropriate API endpoint and analyzes the JSON data received as a response. This data includes metadata such as the post content (text), poster information, and the post date and time.

[1042] Next, the server uses regular expressions to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data. This process extracts only the text and formats it into a form suitable for analysis. A specific example of cleansing is a method of removing links and tags using Python's regular expression library.

[1043] The server then performs sentiment analysis on the cleansed text data. For sentiment analysis, it uses a natural language processing library such as TextBlob. Using TextBlob, it is possible to calculate a positive, negative, or neutral sentiment score for the text. For example, it calculates that the negative score for the text "Breaking news! Government is lying! Do not trust anyone! FakeNews" is -0.8.

[1044] Next, the server makes a judgment to detect misinformation or propaganda based on the sentiment score. If the negative score exceeds a certain threshold, the post is judged to be misinformation or propaganda. Based on this judgment result, the server issues a warning for posts in which misinformation or propaganda has been detected. The warning is notified to the user with the post ID and the reason for the detection. For example, a warning message such as "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" can be generated and a notification email can be sent to the administrator.

[1045] As a concrete example, a server collects the post "Breaking news! Government is lying! Do not trust anyone! FakeNews" from the Twitter API, cleanses it, and formats it as "Breaking news! Government is lying! Do not trust anyone! FakeNews." Next, it performs sentiment analysis using TextBlob and detects a high negative score, determining that the post is misinformation or propaganda. As a result, the server generates a warning "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and notifies the user.

[1046] In this way, the present invention realizes a system that suppresses the spread of false information and propaganda and provides accurate information.

[1047] Example prompts for generative AI models

[1048] "Use Twitter's API to remove links and hashtags from posts collected, and use TextBlob to perform sentiment analysis. Generate warning messages for posts with high negative scores."

[1049] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1050] Step 1:

[1051] The server collects post data from social media APIs. The server sends an HTTP request to the API endpoint and receives JSON format data as a response. The input is a request that includes an API key and query parameters, and the output is an array of the received post data. As a specific example, the Twitter API endpoint "statuses / user_timeline" is used to obtain the latest posts from a specific user.

[1052] Step 2:

[1053] The server cleanses the received post data. It uses regular expressions to remove links (URLs), user tags (@username), and hashtags. This cleansing process passes the received JSON data post content as input, and outputs text data with unnecessary elements removed. As a specific example of how it works, the Python regular expression library is used to format "Breaking news! Government is lying! Do not trust anyone! FakeNews" to "Breaking news! Government is lying! Do not trust anyone! FakeNews".

[1054] Step 3:

[1055] The server performs sentiment analysis on the cleansed text data. It uses a natural language processing library such as TextBlob to calculate a sentiment score. It receives the cleansed text as input and outputs a positive, negative, or neutral sentiment score. As a specific example, TextBlob is used to calculate a negative score of -0.8 for "Breaking news! Government is lying! Do not trust anyone! FakeNews."

[1056] Step 4:

[1057] The server detects false information and propaganda based on the sentiment score. The server compares the sentiment score with a set negative score threshold to identify posts that are likely to be false information or propaganda. The server receives the sentiment score as input and obtains posts determined to be false information as output. As a specific example of operation, if the negative score is -0.5 or less, the server determines that the post is false information.

[1058] Step 5:

[1059] The server displays a warning if misinformation or propaganda is detected. The server generates a warning message including the detected post ID, username, and reason for detection, and notifies the user. The input is the detected post information, and the output is the warning message. As a specific example of operation, the server generates a warning message saying "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and sends a notification email to the user.

[1060] (Application example 1)

[1061] 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."

[1062] This invention aims to solve the technical challenges of detecting and quickly addressing disinformation and propaganda spreading on social media. In particular, there is a need to collect and analyze information in real time, visualize it, and provide prompt warnings. However, current systems have difficulty integrating multiple data collection methods and complex processes such as speech recognition, text analysis, and sentiment analysis.

[1063] 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.

[1064] In this invention, the server includes means for collecting social media information to identify false information and propaganda, means for cleansing the collected information, means for analyzing the cleansed information to detect false information and propaganda, means for displaying a warning when false information and propaganda are detected, means for collecting the latest media information through speech recognition and text analysis, means for performing sentiment analysis based on the speech recognition and text analysis, and means for displaying a warning on the display of the smart device. This makes it possible to quickly mitigate the impact of false information and propaganda by evaluating the authenticity of information in real time and displaying a warning.

[1065] "Disinformation" is information that is not based on facts and is spread intentionally or unintentionally with the intent to mislead or confuse.

[1066] "Propaganda" is a method of selectively providing or manipulating information with a specific political, social, or religious intent, and is used to influence people's opinions and actions.

[1067] "Social media" refers to a platform that allows users to share, exchange, and disseminate information via the Internet, and a typical example is SNS (social networking service).

[1068] "Information collection means" refers to the means for collecting post data and metadata sent on social media, and is realized using technologies such as APIs.

[1069] "Cleansing methods" are methods used to remove unnecessary data (e.g., links, user tags, hashtags) from collected information and prepare it in a format suitable for analysis.

[1070] "Analysis methods" are means for detecting false information and propaganda by conducting sentiment analysis and content verification on cleansed information.

[1071] "Warning display means" refers to means for displaying or notifying users of false information or propaganda as a warning when such information is detected.

[1072] "Speech recognition and text analysis means" refers to means for converting speech data into text data and analyzing the text data, and extracting information using natural language processing techniques.

[1073] "Sentiment analysis means" is a means for evaluating the emotional state (positive, negative, neutral) of text data, and analyzes the credibility of information by assigning an emotional score.

[1074] "Smart device display means" means a means for visually displaying important information or warnings to a user on smart glasses or other wearable devices.

[1075] This invention relates to a system for identifying and countering false information and propaganda, and specifically includes the following means:

[1076] The server first uses social media APIs to collect post data and related metadata, then uses a cleansing method to remove unnecessary information such as links, user tags, and hashtags from the collected data, converting it into text suitable for analysis.

[1077] Next, speech recognition and text analysis tools are used to capture audio and video data from smart devices (e.g., smart glasses) and convert them into text data. OpenAI's speech recognition API is used here.

[1078] The cleansed text data is then evaluated using sentiment analysis tools. Natural language processing libraries such as the TextBlob library and Hugging Face's Transformers are used to calculate a sentiment score for the text. If the sentiment score exceeds a certain threshold, the text is marked as misinformation or propaganda.

[1079] When disinformation or propaganda is detected, a warning will be displayed on the smart device's display, including the ID of the detected information and the reason for the detection, and the user will be visually notified. For example, a warning message will be displayed in real time on the display of smart glasses.

[1080] Consider the following scenario: Smart glasses recognize the speech phrase "Breaking news! The government is lying!" and convert it into text. Sentiment analysis of the text detects a high negative score, identifying it as disinformation or propaganda, and a warning is displayed on the glasses' display.

[1081] In addition to cleansing and sentiment analysis of the generated data, the following example prompts can be used to feed the generative AI model:

[1082] Example prompt sentence:

[1083] Transcribe the speech "Breaking news! The government is lying! Trust no one!" and perform sentiment analysis on this text. If the negative score is high, show a warning to the user.

[1084] This allows users to make informed decisions in real time, quickly reducing the impact of disinformation and propaganda.

[1085] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1086] Step 1:

[1087] The server uses the social media API to collect the latest post data and related metadata. Specifically, it sends an HTTP request to the API endpoint and parses the JSON data received as a response. The input is data from the API, and the output is the parsed post data and metadata.

[1088] Step 2:

[1089] The server cleanses the collected post data by using regular expressions to remove links (URLs), user tags (@username), hashtags, etc. The input is the analyzed post data, and the output is cleansed text data with unnecessary information removed.

[1090] Step 3:

[1091] The server evaluates the cleansed text data using speech recognition and text analysis. First, the voice data is converted into text data using OpenAI's speech recognition API, and then the converted text data is analyzed. The input is the voice data and the cleansed text data, and the output is the analyzed text data.

[1092] Step 4:

[1093] The server performs sentiment analysis on the parsed text data, using the TextBlob library and Hugging Face Transformers to calculate a positive, negative, or neutral sentiment score for the text. The input is the parsed text data, and the output is the sentiment score.

[1094] Step 5:

[1095] The server determines the likelihood of misinformation or propaganda based on the sentiment score. If the sentiment score exceeds a certain threshold, the post is identified as misinformation or propaganda. The input is the sentiment score, and the output is the results of identifying the post as misinformation.

[1096] Step 6:

[1097] The server displays a warning when false information or propaganda is identified. Specifically, it sends and displays a warning message in real time on the display of a smart device (such as smart glasses). The input is the result of identifying false information, and the output is the display of a warning message.

[1098] Step 7:

[1099] Users can check the warnings displayed on their smart devices and take measures to prevent the effects of disinformation and propaganda. Specific actions include receiving the alert and checking reliable sources of information. The input is the warning message, and the output is the user's response.

[1100] 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.

[1101] The present invention relates to a social media monitoring system for identifying false information and propaganda and implementing countermeasures. Furthermore, by combining an emotion engine that recognizes user emotions, the system is equipped with the function of assessing the impact of false information and propaganda and adjusting the warning content. The system operates mainly on a server and includes the following main means:

[1102] System Overview

[1103] 1. Data collection methods:

[1104] The server collects post data from the social media API, including metadata such as the post content (text), poster information, and post date and time.

[1105] 2. Data cleansing methods:

[1106] The server cleanses the collected data by removing unnecessary elements such as links, user tags, and hashtags, resulting in text data suitable for analysis.

[1107] 3. Data analysis methods:

[1108] The server uses the cleansed data to perform sentiment analysis and content validation, using libraries such as TextBlob to assess the text's emotional state as positive, negative, or neutral.

[1109] 4. Detecting Disinformation:

[1110] Based on the analysis results, the server sets thresholds for positive and negative scores to identify posts that are likely to contain false information or propaganda, and displays a warning if the negative score exceeds a certain level.

[1111] 5. Warning display means:

[1112] If the server detects false information or propaganda, it will display or notify the user of the post's ID and reason as a warning.

[1113] 6. Emotion Engine:

[1114] The server uses an emotion engine to recognize users' emotions and assess the impact of misinformation and propaganda, analyzing text inputs and behavioral data when users interact with the system.

[1115] 7. Adjustment of warning content:

[1116] The server adjusts the warning content based on the results of the emotion engine, for example, strengthening the warning if the user is already emotionally affected, and appropriately adjusting the warning content to avoid excessively negative impact.

[1117] 8. Recording and statistical information generation means:

[1118] The server will record the results of detecting false information and propaganda and generate statistical information based on this information, which will be useful for analyzing trends in the spread of false information and developing effective countermeasures.

[1119] Explaining program processing in natural language

[1120] 1. Data Collection:

[1121] The server collects the latest post data from the social media API, sends an HTTP request to the API endpoint, and parses the JSON data received as a response.

[1122] 2. Data cleansing:

[1123] The server uses regular expressions to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data. By cleansing, only the text is extracted and formatted for analysis.

[1124] 3. Data Analysis:

[1125] The server performs sentiment analysis on the cleansed text data, using libraries such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[1126] 4. Detecting Disinformation:

[1127] Based on the sentiment score, the server determines that there is a high possibility of misinformation or propaganda if the negative score exceeds a predetermined threshold.

[1128] 5. User emotion recognition using emotion engine:

[1129] The server analyzes the user's input and behavioral data and uses an emotion engine to recognize the user's emotional state, which can include joy, sadness, anger, surprise, etc.

[1130] 6. Adjustment of warning content:

[1131] The server adjusts the warning content appropriately based on the analysis results of the emotion engine. For example, if the user is already in a negative emotional state, the server displays a warning message that is careful not to further exacerbate the negative emotional state.

[1132] 7. Warning display:

[1133] The server issues a warning to posts that contain false information or propaganda, including the post's ID and the reason for the detection, and reflects the evaluation results of the emotion engine.

[1134] 8. Recording and Statistics Generation:

[1135] The server records the results of detecting false information and propaganda and generates statistical information that includes the results of the emotion engine, which allows for analysis of the spread of false information and countermeasures based on the emotional state of users.

[1136] Specific examples

[1137] For example, suppose your server collects the following post data from an API:

[1138] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[1139] The server cleanses this submission and formats it as follows:

[1140] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[1141] The server then performs sentiment analysis on this cleansed text data to detect high negative scores, and if the negative score exceeds a certain threshold, the server identifies the post as misinformation or propaganda.

[1142] The server uses an emotion engine to recognize the user's emotional state based on the user's behavioral data. For example, if the user has seen many negative posts in the past 24 hours, the emotion engine will recognize that the user is in a negative emotional state.

[1143] Based on the results, the server generates a warning message "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and notifies the user after adjusting the warning content.

[1144] In this way, the present invention is a system that curbs the spread of disinformation and propaganda, taking into account the emotional state of the user and taking appropriate measures.

[1145] The processing flow will be explained below.

[1146] Step 1:

[1147] The server sends an HTTP GET request to the specified social media API URL, collects post data from the API, and receives JSON formatted data as a response.

[1148] Step 2:

[1149] The server parses the JSON data obtained from the response and extracts metadata such as the content (text) of each post, poster information, and posting date and time.

[1150] Step 3:

[1151] The server cleanses the text data of each extracted post. Specifically, it removes links (URLs) using regular expressions, and also removes user tags (@username) and hashtags using regular expressions. By cleansing, the text data becomes suitable for analysis.

[1152] Step 4:

[1153] The server performs sentiment analysis on the cleansed text data, using libraries such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[1154] Step 5:

[1155] The server evaluates the resulting sentiment score and determines that a post is likely to be misinformation or propaganda, especially if the negative score exceeds a predefined threshold, specifically if the negative score is less than -0.5.

[1156] Step 6:

[1157] The server analyzes user input and behavioral data and uses an emotion engine to recognize the user's emotional state. For example, if a user has viewed many negative posts in the past 24 hours, the emotion engine will determine that the user is in a negative emotional state.

[1158] Step 7:

[1159] The server adjusts the warning content based on the emotion recognition results. If the user is already in a negative emotional state, the warning will be strengthened. The server also adjusts the warning content appropriately to avoid excessively negative impact.

[1160] Step 8:

[1161] If misinformation or propaganda is detected, the server generates a warning message containing the post's ID and the reason for the detection (that it was determined to have a high negative score).

[1162] Step 9:

[1163] The server notifies the user of the generated warning message by displaying the warning message on the console or by sending a notification email to the administrator.

[1164] Step 10:

[1165] The server records the results of detecting false information and propaganda and generates statistical information that includes the results of the emotion engine. This statistical information is useful for taking measures based on the spread of false information and the emotional state of users.

[1166] As a concrete example, suppose a server collects a post with the title "Breaking news! Government is lying! Do not trust anyone! FakeNews." After cleansing this post, the resulting text is formatted as "Breaking news! Government is lying! Do not trust anyone! FakeNews." Sentiment analysis determines that this text has a high negative score and is misinformation or propaganda. If the emotion engine subsequently determines that the user is in a negative emotional state, the server generates a warning message such as "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" to notify the user.

[1167] Example 2

[1168] 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."

[1169] In today's social media, false information and propaganda often spread quickly and have a significant impact on society. However, current systems are insufficient in detecting and dealing with false information, and do not take appropriate measures that take into account the emotional state of users. Therefore, it is necessary to solve the problem of quickly detecting false information and propaganda and minimizing their emotional impact on users.

[1170] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting social media data, a means for cleansing the collected data, a means for analyzing the cleansed data and detecting false information or propaganda, a means for displaying a warning when false information or propaganda is detected, a means for analyzing user behavior data and recognizing the user's emotional state using an emotion engine, and a means for adjusting the content of the warning based on the results of the emotion engine. This enables the rapid detection of false information and propaganda and the display of an appropriate warning that takes the user's emotional state into consideration.

[1171] "Data collection means" refers to a device or program that uses a social media API to obtain data such as the content of posts, poster information, and posting date and time.

[1172] A "data cleansing tool" is a device or program that removes unnecessary elements such as links, user tags, and hashtags from collected data and formats it into a form suitable for analysis.

[1173] The "data analysis means" is a device or program that performs sentiment analysis and content verification on the cleansed data and calculates a positive, negative, or neutral sentiment score.

[1174] A "fake news detection tool" is a device or program that, based on the results of data analysis, identifies posts that may be fake or propaganda based on a set threshold.

[1175] A "warning display means" is a device or program that generates and displays a message or alert to notify the user when false information or propaganda is detected.

[1176] An "emotion engine" is an algorithm or device that analyzes a user's text input and behavioral data to recognize the user's emotional state.

[1177] The "alert content adjustment means" is a device or program that optimizes the alert content based on the results of the emotion engine and taking into account the emotional state of the user.

[1178] The present invention relates to a social media monitoring system for identifying disinformation and propaganda and taking countermeasures. The system is server-based and includes the following main means:

[1179] The server first collects post data from the social media API. Specifically, it sends an HTTP request to the API endpoint and analyzes the JSON data received as a response. The collected data includes the post content (text), poster information, and posting date and time. This allows the server to store the most recent post data in a local database.

[1180] Next, the server uses a data cleansing method to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data using regular expressions, extracting only the text and preparing it in a form suitable for analysis.

[1181] The server then performs sentiment analysis on the cleansed data using a data analysis tool. For sentiment analysis, it uses a library such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text, which is then used to evaluate and record the emotional state of the post.

[1182] The server then uses a false information detection method based on the sentiment score to determine that a message is likely to be false information or propaganda if the negative score exceeds a predetermined threshold. If false information or propaganda is detected, the result is recorded.

[1183] The system also analyzes user behavior data and uses an emotion engine to recognize the user's emotional state. Based on the user's input and past behavioral data, the system identifies the user's emotional state. For example, if a user has seen many negative posts, the system will recognize the user's emotional state as negative.

[1184] Based on the analysis results of the emotion engine, the server uses the warning content adjustment means to appropriately adjust the warning content. For example, if the user is already in a negative emotional state, the server may take measures such as changing the warning text to a softer one.

[1185] Finally, for posts that contain false information or propaganda, the server issues a warning using a warning display method. The warning includes the post's ID and the reason for the detection, and reflects the evaluation results of the emotion engine. This allows the server to notify users of an appropriate warning.

[1186] To illustrate this system, here is an example prompt:

[1187] Please explain how your system would perform sentiment analysis on the following post data, and if the negative score exceeds a certain threshold, display a warning about misinformation or propaganda and adjust the warning appropriately. Post data: 'Breaking news! Government is lying! Do not trust anyone! FakeNews'

[1188] In this way, the present invention can effectively curb the spread of false information and propaganda, taking into account the emotional state of the user and taking appropriate measures.

[1189] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1190] Step 1: Data collection

[1191] The server collects post data from the social media API. Specifically, it sends an HTTP GET request to the API endpoint and analyzes the JSON data received as a response. This process obtains metadata such as the post content (text), poster information, and posting date and time as input, and stores them in a local database.

[1192] Step 2: Data cleansing

[1193] The server uses regular expressions to remove unnecessary elements such as links (URLs), user tags (@username), and hashtags from the collected JSON data. The input to this process is the collected post data, and the output is cleansed text data. Specifically, it applies regular expression patterns to remove unnecessary strings.

[1194] Step 3: Data analysis

[1195] The server performs sentiment analysis on the cleansed text data as input. It uses a library such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text. The output of this process is a sentiment score for each post. Specifically, it calls a library to analyze the text and calculate the sentiment score.

[1196] Step 4: Detecting misinformation

[1197] The server takes the sentiment score as input and checks whether the negative score exceeds a predetermined threshold. If the negative score exceeds the threshold, the post is deemed to be false information or propaganda. The output of this process is a list of posts that have been deemed to be false information. Specifically, the sentiment score is compared with the threshold, and posts that meet the criteria are added to the list.

[1198] Step 5: User Emotion Recognition

[1199] The server collects user behavior data and uses an emotion engine to recognize the user's emotional state. The input is the user's text input and behavior log, and the output is the user's emotional state (joy, sadness, anger, etc.). Specifically, the server analyzes the user's behavior data and identifies the user's emotional state through the emotion engine.

[1200] Step 6: Adjust the warning content

[1201] The server uses the results of the emotion engine as input and adjusts the warning content appropriately. For example, if the user is already in a negative emotional state, the warning text is changed to a softer expression. The output of this process is an adjusted warning message. Specifically, the content of the warning text is generated and changed based on the user's emotional state.

[1202] Step 7: Warning Display

[1203] The server issues a warning for posts that are determined to be false. The input is the post that needs a warning and the adjusted warning message, and the output is a warning that is displayed to the user. Specifically, the generated warning message is displayed through an interface to notify the user.

[1204] Step 8: Recording and generating statistics

[1205] The server records the results of the detection of false information and propaganda and the results of the emotion engine, and generates statistical information. The detection results and the emotion engine results are used as input, and the output is statistical information. Specifically, statistical analysis is performed based on this data, and a report is created to analyze the spread of false information and effective countermeasures.

[1206] (Application example 2)

[1207] 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."

[1208] In modern society, the spread of false information and propaganda through social media has become a major problem. This can lead to users being influenced by false information, causing anxiety and confusion. Furthermore, existing systems only detect false information and do not display warnings that take into account the user's emotional state, which can cause excessive stress. To solve this situation, a system that not only detects false information but also takes into account the user's emotional state and displays appropriate warnings is needed.

[1209] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1210] In this invention, the server includes a data collection means for identifying false information and propaganda, a means for cleansing the collected data, a means for analyzing the cleansed data and detecting false information and propaganda, a means for displaying a warning when false information and propaganda are detected, a means for processing the collected data in real time, a means for analyzing the emotional state of the user, and a means for adjusting the content of the warning based on the emotional state of the user, thereby enabling appropriate detection of false information and propaganda and display of a warning that takes into account the emotional state of the user.

[1211] "Data collection means" refers to the means used to collect posting data and metadata from social media.

[1212] A "cleansing method" is a method for removing unnecessary elements from collected data and preparing text data in a form suitable for analysis.

[1213] The "data analysis means" is a means for analyzing the cleansed text data and determining sentiment scores and false information.

[1214] "Disinformation and propaganda detection measures" are measures that identify posts that are likely to be disinformation or propaganda based on the results of data analysis.

[1215] "Warning display measures" are measures that display a warning on a post when false information or propaganda is detected.

[1216] "Real-time processing means" refers to means for updating and analyzing collected data in real time and responding immediately.

[1217] The "emotional state analysis means" is a means for analyzing the user's emotional state based on the user's past behavioral data and text input.

[1218] The "warning content adjustment means" is a means for adjusting the warning content according to the emotional state of the user based on the result of the emotional state analysis means.

[1219] This paper describes an embodiment of the present invention, which describes a system for identifying false information and propaganda and displaying warnings based on a user's emotional state. The system includes means for data collection, data cleansing, data analysis, false information detection, emotional state analysis, warning content adjustment, and warning display.

[1220] Data collection methods

[1221] The server collects post data from the social media API. Specifically, it sends an HTTP request to the API endpoint and analyzes the JSON data received as a response. This process uses a communication library such as the requests library.

[1222] Data Cleansing Methods

[1223] The server uses regular expressions to remove unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected data, thereby obtaining text data suitable for analysis.

[1224] Data Analysis Methods

[1225] The server performs sentiment analysis on the cleansed text data, using a natural language processing library such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[1226] Measures to detect disinformation and propaganda

[1227] The server determines the likelihood of misinformation or propaganda if the negative score exceeds a predetermined threshold based on the sentiment score. This determination is made using a threshold to determine whether a warning should be sent to the user.

[1228] Emotional state analysis means

[1229] The server analyzes user input and behavioral data and uses an emotion engine to recognize the user's emotional state. Based on the user's behavioral data within the past 24 hours, it counts the number of negative posts and evaluates the user's current emotional state.

[1230] Warning content adjustment means

[1231] The server adjusts the warning content based on the results of the emotional state analysis means. For example, if the user is already in a negative emotional state, the server displays a warning message that takes care not to further exacerbate the user's negative emotional state.

[1232] Warning display means

[1233] The server issues a warning to posts that contain false information or propaganda, including the post's ID and the reason for the detection, and reflects the evaluation results of the emotion engine.

[1234] Specific examples

[1235] For example, if your server collects the following post data from an API:

[1236] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[1237] The server cleanses this submission and formats it as follows:

[1238] "Breaking news! Government is lying! Do not trust anyone! FakeNews"

[1239] The server then performs sentiment analysis on this cleansed text data to detect high negative scores, and if the negative score exceeds a certain threshold, the server identifies the post as misinformation or propaganda.

[1240] The server uses an emotion engine to recognize the user's emotional state based on the user's behavioral data. For example, if the user has seen many negative posts in the past 24 hours, the emotion engine will recognize that the user is in a negative emotional state.

[1241] Based on the results, the server generates a warning message "Post ID: 12345 - Alert: Detected potential misinformation or propaganda" and notifies the user after adjusting the warning content.

[1242] Prompt Sentence Examples

[1243] Example prompts to apply to your API endpoints:

[1244] "Please provide the latest posts data from the social media API endpoint: 'https: / / api.socialmedia.com / latest_posts' additionally, gather the user's past 24-hour activity data for sentiment analysis: [ {"text": "I'm so sad about the news.", "timestamp": "2023-10-20T14:00:00Z"}, {"text": "Angry about the government's actions.", "timestamp": "2023-10-20T16:00:00Z"}]

[1245] In this way, the present invention makes it possible to curb the spread of false information and propaganda and to take appropriate measures taking into account the emotional state of the user.

[1246] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1247] Processing flow

[1248] Step 1: Data collection

[1249] The server sends an HTTP request to the social media API to collect the latest post data, and receives a JSON response from the API endpoint.

[1250] Input: API endpoint

[1251] Output: Post data (JSON format)

[1252] Step 2: Data cleansing

[1253] The server removes unnecessary information such as links (URLs), user tags (@username), and hashtags from the collected post data using regular expressions.

[1254] Input: Post data (JSON format)

[1255] Output: Cleansed text data

[1256] Step 3: Data analysis

[1257] The server performs sentiment analysis on the cleansed text data, using a natural language processing library such as TextBlob to calculate a positive, negative, or neutral sentiment score for the text.

[1258] Input: Cleansed text data

[1259] Output: Sentiment score (positive, negative, neutral)

[1260] Step 4: Detecting disinformation and propaganda

[1261] The server determines that the likelihood of the information being false information or propaganda is high if the negative score exceeds a predetermined threshold based on the emotion score.

[1262] Input: Sentiment score

[1263] Output: Disinformation and propaganda judgement result (True / False)

[1264] Step 5: Analyze emotional state

[1265] The server analyzes the user's behavioral data and uses an emotion engine to recognize the user's emotional state, for example, by counting the number of negative posts the user has made in the past 24 hours.

[1266] Input: User behavior data (post data from the past 24 hours)

[1267] Output: User's emotional state (number of negative comments)

[1268] Step 6: Adjust the warning content

[1269] The server adjusts the warning content based on the results of the emotional state analysis. If the user is already in a negative emotional state, the server generates a warning message that is careful not to further upset the user.

[1270] Input: User's emotional state, misinformation and propaganda judgment results

[1271] Output: Adjusted warning content

[1272] Step 7: View warnings

[1273] The server issues a warning to users when a post contains false information or propaganda, and notifies them. The warning includes the post's ID, the reason for the detection, and the evaluation result by the emotion engine.

[1274] Input: Adjusted warning content, post ID

[1275] Output: Display of warning message

[1276] This process allows the server to effectively detect false information and propaganda and display warnings in real time that take into account the user's emotional state.

[1277] 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.

[1278] 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.

[1279] 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.

[1280] 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.

[1281] 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.

[1282] 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.

[1283] 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).

[1284] 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.

[1285] 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."

[1286] 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.

[1287] 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).

[1288] 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.

[1289] 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.

[1290] 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.

[1291] 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.

[1292] 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.

[1293] 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.

[1294] 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.

[1295] 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.

[1296] 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.

[1297] 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.

[1298] The following is further disclosed regarding the above embodiment.

[1299] (Claim 1)

[1300] Social media data collection measures to identify disinformation and propaganda;

[1301] a means for cleansing the collected data; and

[1302] A means of analyzing the cleansed data to detect disinformation and propaganda; and

[1303] A way to display warnings when disinformation or propaganda is detected;

[1304] A system including:

[1305] (Claim 2)

[1306] 10. The system of claim 1, further comprising means for performing sentiment analysis on the cleansed data.

[1307] (Claim 3)

[1308] 10. The system of claim 1, further comprising means for recording and generating statistical information on the detection of disinformation and propaganda.

[1309] "Example 1"

[1310] (Claim 1)

[1311] means of collecting social media data;

[1312] a means for cleansing the collected data; and

[1313] A means of analyzing the cleansed data to detect disinformation and propaganda; and

[1314] A way to display warnings when disinformation or propaganda is detected;

[1315] A system including:

[1316] (Claim 2)

[1317] 10. The system of claim 1, further comprising means for performing sentiment analysis on the cleansed data.

[1318] (Claim 3)

[1319] 10. The system of claim 1, further comprising means for recording and generating statistical information on the detection of disinformation and propaganda.

[1320] (Claim 4)

[1321] 3. The system of claim 2, further comprising means for utilizing natural language processing techniques as a computational resource for performing sentiment analysis.

[1322] (Claim 5)

[1323] 10. The system of claim 1, further comprising means for sending a notification if disinformation or propaganda is detected.

[1324] "Application Example 1"

[1325] (Claim 1)

[1326] Social media intelligence gathering tools to identify disinformation and propaganda;

[1327] a means for cleansing the collected information; and

[1328] A means of analyzing the cleansed information to detect disinformation and propaganda; and

[1329] A way to display warnings when disinformation or propaganda is detected;

[1330] A means for collecting the latest media information through speech recognition and text analysis;

[1331] means for performing sentiment analysis based on speech recognition and text analysis;

[1332] a means for displaying a warning on a display of the smart device;

[1333] A system including:

[1334] (Claim 2)

[1335] 10. The system of claim 1, further comprising means for performing a sentiment analysis of the cleansed information.

[1336] (Claim 3)

[1337] 10. The system of claim 1, further comprising means for recording and generating statistical information on the detection of disinformation and propaganda.

[1338] "Example 2: Combining Emotion Engines"

[1339] (Claim 1)

[1340] Social media data collection methods and

[1341] a means for cleansing the collected data; and

[1342] A means of analyzing the cleansed data to detect disinformation and propaganda; and

[1343] A way to display warnings when disinformation or propaganda is detected;

[1344] means for analyzing user behavior data and recognizing the user's emotional state using an emotion engine;

[1345] a means for adjusting the alert content based on the results of the emotion engine;

[1346] A system including:

[1347] (Claim 2)

[1348] 10. The system of claim 1, further comprising means for performing sentiment analysis on the cleansed data.

[1349] (Claim 3)

[1350] 10. The system of claim 1, further comprising means for recording and generating statistical information on the detection of disinformation and propaganda.

[1351] "Application example 2 when combining emotion engines"

[1352] (Claim 1)

[1353] Data collection measures to identify disinformation and propaganda;

[1354] a means for cleansing the collected data; and

[1355] A means of analyzing the cleansed data to detect disinformation and propaganda; and

[1356] A way to display warnings when disinformation or propaganda is detected;

[1357] a means for processing the collected data in real time;

[1358] means for analyzing the emotional state of a user;

[1359] means for tailoring alert content based on the user's emotional state;

[1360] A system including:

[1361] (Claim 2)

[1362] 10. The system of claim 1, further comprising means for performing sentiment analysis on the cleansed data.

[1363] (Claim 3)

[1364] 10. The system of claim 1, further comprising means for recording and generating statistical information on the detection of disinformation and propaganda. [Explanation of symbols]

[1365] 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. Social media data collection measures to identify disinformation and propaganda; a means for cleansing the collected data; and A means of analyzing the cleansed data to detect disinformation and propaganda; and A way to display warnings when disinformation or propaganda is detected; A system including:

2. 10. The system of claim 1, further comprising means for performing sentiment analysis on the cleansed data.

3. 10. The system of claim 1, further comprising means for recording and generating statistics on the detection of disinformation and propaganda.

Images