System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting information on the Internet; means for pre-processing the collected information; means for analyzing the pre-processed information and performing sentiment analysis and credibility determination; means for performing an action on unreliable and negative information; and means for notifying a user of a result of the action.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] There is a huge amount of false and misleading information on social media and the internet, and its spread can cause social unrest. This information poses a major problem, especially in situations requiring rapid and accurate responses during natural disasters and emergencies. Many social media users, including young people and the elderly, may be affected by this false information. Furthermore, false information is sometimes generated automatically by AI, making it extremely difficult for humans to monitor and remove it by themselves. Given this background, there is a growing need for systems that can detect false information on social media and the internet in real time and take effective countermeasures. [Means for solving the problem]

[0005] The present invention provides a system including a means for collecting information on the Internet, a means for preprocessing the collected information, a means for analyzing the preprocessed information and performing sentiment analysis and reliability determination, a means for taking action on unreliable and negative information, and a means for notifying a user of the results of the action. The system further includes a filtering means for removing unnecessary data from the collected information and a preprocessing means for removing special characters and tokenizing text. The analysis means is configured to perform sentiment analysis and reliability determination using natural language processing technology. This enables real-time detection of false information and misinformation and rapid and effective response.

[0006] "Means of collecting information on the Internet" refers to technologies and devices that obtain information based on specified keywords or conditions from the Internet or social media platforms.

[0007] "Means for preprocessing collected information" refers to techniques and processes for removing unnecessary data from acquired information and converting it into a format that is easy to analyze.

[0008] "Means for analyzing preprocessed information" refers to technologies and algorithms that use natural language processing technology to analyze the content of preprocessed information and evaluate sentiment and reliability.

[0009] "Sentiment analysis" is the process of extracting emotional elements such as positive, negative, and neutral from the context of text data and analyzing their distribution and trends.

[0010] "Credibility determination" is the process of assessing whether information is trustworthy, based on reliable sources and other criteria.

[0011] "Low-reliability, negative information" is information that is judged to be low-reliability based on the evaluation criteria and that is judged to have a strong negative element as a result of sentiment analysis.

[0012] "Means for taking action" refers to the technology or process that takes specific action based on the analysis results, such as sending a request to remove false or misleading information or displaying a warning message.

[0013] "Means for notifying the user of the results of the action" refers to a notification function or interface that notifies the user that an action has been performed.

[0014] "Filtering means" is a process that removes unnecessary information and noise from collected data and extracts only the necessary information.

[0015] "Special character removal" is the process of removing special characters and symbols from text data that are not required for analysis.

[0016] "Text tokenization" is the process of breaking down text into units such as words and phrases, making it easier to analyze.

[0017] "Natural language processing technology" is a general term for algorithms and technologies that help computers understand, interpret, and generate human language. [Brief explanation of the drawings]

[0018] [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

[0019] 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.

[0020] First, the terms used in the following description will be explained.

[0021] 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).

[0022] 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.

[0023] 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.

[0024] 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.

[0025] 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."

[0026] [First embodiment]

[0027] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0028] 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.

[0029] 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).

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0035] 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.

[0036] 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.

[0037] 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.

[0038] 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."

[0039] This invention is a system for detecting false and misleading information on the Internet in real time and issuing warnings to users. The system includes a server that collects information on the Internet, preprocesses the collected information, analyzes the preprocessed information, takes action against unreliable and negative information, and notifies users of the results.

[0040] Specifically, the program of this system performs the following processes.

[0041] Data collection

[0042] The server collects public information from social media and the internet based on pre-specified keywords, using the API of the social media platform to automatically retrieve new posts.

[0043] Data Preprocessing

[0044] The server then removes unnecessary information from the collected data and converts it into a format that is easier to analyze, by performing preprocessing such as removing HTML tags, eliminating special characters, and tokenizing the text.

[0045] Data analysis

[0046] The server analyzes the preprocessed data using natural language processing technology, extracting important keywords, analyzing sentiment, and assessing trustworthiness. Sentiment analysis determines and scores positive, negative, or neutral sentiment from the context of the text. Credibility assessment verifies whether the collected information source is trustworthy and assigns a trustworthiness score.

[0047] Reliability determination

[0048] The server determines the reliability of the information based on the results of the data analysis. This includes comparing it with reliable sources, checking against past data, etc. If the information is deemed unreliable and negative, it proceeds to the next step.

[0049] Action Execution

[0050] The server sends a request to the social media platform to remove unreliable and negative information, and also requests that a warning message be added to suspicious information. Based on this request, the social media platform will delete the post or display a warning message.

[0051] User Notification

[0052] The device will notify the user that the deletion request has been carried out and will display a warning message, allowing the user to review the notification and reconfirm the authenticity of the information.

[0053] Specific examples

[0054] For example, suppose a server collects posts about the "Noto Peninsula Earthquake." After removing HTML tags and special characters from the collected data, the text is tokenized. Next, the collected data is analyzed using natural language processing technology to perform sentiment analysis and determine trustworthiness. For posts that are determined to be unreliable and contain negative information based on comparison with past reliable sources, a request is sent to the social media platform to delete them. In addition, for suspicious information, a request is sent to display a warning message. This allows the device to notify the user of the warning and reconfirm the authenticity of the information.

[0055] This system will effectively prevent the spread of false and misinformation and provide users with accurate information.

[0056] The processing flow will be explained below.

[0057] Step 1:

[0058] The server collects public information from social media and the internet. This information collection is done by using the API of the social media platform to obtain new posts containing specified keywords (e.g., "Noto Peninsula Earthquake"). The collected data is temporarily stored in a database.

[0059] Step 2:

[0060] The server pre-processes the collected data to remove unnecessary information, such as removing HTML tags, eliminating special characters, and tokenizing the text, converting the data into a format that is easier to analyze.

[0061] Step 3:

[0062] The server then analyzes the preprocessed data using natural language processing (NLP) techniques. The main analysis items include keyword frequency analysis, sentiment analysis, and information source verification. Sentiment analysis produces a positive, negative, or neutral sentiment score.

[0063] Step 4:

[0064] The server evaluates the reliability of the information based on the data analysis results. It compares the data with reliable sources (e.g., official announcements, well-known news sites) and calculates a reliability score. If the reliability is low, it proceeds to the next step.

[0065] Step 5:

[0066] The server takes action against unreliable and negative information, specifically by sending a deletion request to the social media platform for the relevant social media post, and by sending a request to display a warning message for sensitive information.

[0067] Step 6:

[0068] The terminal executes deletion requests and displays warning messages based on requests from the server. When a post is deleted, the poster is notified of the deletion, and when a post has a warning message attached, a warning message is displayed to warn viewers.

[0069] Step 7:

[0070] Users can check the warning messages and deletion notices displayed on their devices, reconfirm the reliability of the information provided, and make appropriate decisions.

[0071] In this way, a system will be built that can detect fake and misleading information on the Internet in real time and respond quickly.

[0072] Example 1

[0073] 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."

[0074] In today's internet environment, the rapid spread of false and misleading information has become a social problem. Accurate information and false information are often mixed together on social media platforms, making it difficult for users to determine the reliability of the information. Therefore, there is a need for methods to quickly detect and respond to unreliable information.

[0075] 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.

[0076] In this invention, the server includes a means for collecting information from the Internet, a means for preprocessing the collected information, a means for analyzing the preprocessed information using natural language processing technology to perform sentiment analysis and reliability determination, a means for deleting unreliable and negative information or adding a warning message to the information, and a means for notifying the user of the results of the action, thereby enabling the rapid detection and response to false information and misinformation.

[0077] "Means for collecting information on the Internet" refers to a mechanism for obtaining data based on specified keywords from websites, social networking services (SNS), and other public information on the Internet.

[0078] The "means for preprocessing collected information" refers to a mechanism for performing a process to remove unnecessary information from the collected data and convert it into an analyzable format.

[0079] "Means of analysis using natural language processing technology" refers to a system that utilizes algorithms and models to automatically perform sentiment analysis and reliability assessment on collected text data.

[0080] "Sentiment analysis" is a technique that determines positive, negative, or neutral sentiment from the context of text and assigns a score to each post.

[0081] "Credibility assessment" is the process of assessing the reliability of collected information sources and calculating a reliability score by matching and comparing them with past data.

[0082] "Means for taking action to delete or add a warning message" refers to a mechanism for sending a request to a social media platform to delete or display a warning message for unreliable and negative information.

[0083] "Means for notifying the user of the results of an action" refers to a mechanism that notifies the user of the results of an action taken on their device, allowing the user to verify the reliability and authenticity of the information.

[0084] This invention is a system that detects false information and misinformation on the Internet in real time and warns users. This system includes a series of processes that collect, process, and notify data between servers, terminals, and users.

[0085] The server first uses the API of a social media platform as a means of collecting information on the Internet. For example, it can use the Twitter API to collect tweets related to the "Great Noto Peninsula Earthquake." This allows the server to obtain new posts in real time. As a specific example, when a server collects posts related to the "Great Noto Peninsula Earthquake," it uses the Twitter API to collect tweets containing the relevant keyword.

[0086] Next, the server preprocesses the collected information. To remove unnecessary information from the collected data and convert it into a format that is easier to analyze, it uses the BeautifulSoup library to remove HTML tags and special characters, and then uses the NLTK library to tokenize the text. For example, it preprocesses the tweet "Information about the Noto Peninsula earthquake can be found here → [link]" by removing the link and separating it into words.

[0087] The preprocessed data is then analyzed using natural language processing techniques. The server performs sentiment analysis using Hugging Face Transformers to determine positive, negative, or neutral sentiment. It also evaluates the reliability of the collected information sources and assigns a credibility score. For example, the server might perform sentiment analysis on a tweet such as "This earthquake is having a major impact on people" and assign it a negative score.

[0088] The server then combines these analysis results to determine the reliability of the information. For unreliable and negative information, it sends a deletion request to the social media platform, and for suspicious information, it adds a warning message. Users can request the deletion of specific tweets or the display of warning messages via the Twitter API. For example, they can send a request with a warning that "This is a hoax about the Noto Peninsula earthquake."

[0089] Finally, the server notifies the user of the results of these actions. The device displays a notification on the user's smartphone or other device, allowing the user to check the content of the notification. For example, the device may send the user a notification message such as, "Unreliable information has been detected and deleted. Please check for details."

[0090] This system enables the rapid detection and response of false information and misinformation, and provides accurate information to users. For example, keywords can be specified by inputting the following prompt sentences into the generative AI model:

[0091] Prompt Sentence Examples

[0092] Collect the latest information on the "Noto Peninsula Earthquake" and be alerted to unreliable and negative information.

[0093] Based on this prompt, the system will detect fake and misinformation related to the specified keywords in real time and notify the user.

[0094] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0095] Step 1:

[0096] Data Collection Settings

[0097] The server loads a pre-specified list of keywords to be used for data collection. For example, the keyword "Noto Peninsula Earthquake" is added to the list. This list is used for subsequent API queries.

[0098] Input: A specified list of keywords

[0099] Output: Keyword list loaded

[0100] Specific behavior:

[0101] The server retrieves "Noto Peninsula Earthquake" from the keyword list and sets up data collection.

[0102] Step 2:

[0103] Using SNS API

[0104] The server uses the API of the social media platform to retrieve posts based on the target keywords. For example, the Twitter API is used to collect tweets related to the "Noto Peninsula Earthquake." The server uses the Streaming API to retrieve new posts in real time.

[0105] Input: Keywords to be collected

[0106] Output: Collected SNS post data

[0107] Specific behavior:

[0108] The server uses Twitter's Streaming API to retrieve tweets related to the keyword "Noto Peninsula Earthquake" in real time.

[0109] Step 3:

[0110] Data Preprocessing

[0111] The server cleans the collected data and converts it into a format that is easy to analyze, using BeautifulSoup to remove HTML tags and special characters, and the NLTK library to tokenize the text.

[0112] Input: Collected social media posting data

[0113] Output: Preprocessed text data

[0114] Specific behavior:

[0115] The server removes the link from the tweet "Information about the Noto Peninsula earthquake can be found here → [link]" and separates the text.

[0116] Step 4:

[0117] Conducting sentiment analysis

[0118] The server analyzes the preprocessed text data using natural language processing techniques. Specifically, it uses Hugging Face Transformers to perform sentiment analysis on the text data and determine positive, negative, or neutral sentiment. It assigns a sentiment score to each piece of text.

[0119] Input: Preprocessed text data

[0120] Output: Text data with sentiment scores

[0121] Specific behavior:

[0122] The server performs sentiment analysis on tweets such as "This earthquake is having a big impact on people" and assigns them a negative score.

[0123] Step 5:

[0124] Conducting reliability evaluation

[0125] The server compares the collected information with a historical database and calculates a reliability score to assess the reliability of the information source.

[0126] Input: Text data with sentiment scores

[0127] Output: Text data with confidence scores

[0128] Specific behavior:

[0129] The server references the poster's past reliability data and calculates a reliability score.

[0130] Step 6:

[0131] Integration of reliability judgments

[0132] The server combines the sentiment score and the reliability score to determine the overall reliability of the information. For information that is low in reliability and negative, it proceeds to the next step.

[0133] Input: Text data with confidence scores

[0134] Output: Judgment result

[0135] Specific behavior:

[0136] The server selects posts with a "negative" sentiment score and a low credibility score.

[0137] Step 7:

[0138] Execute Action

[0139] The server sends a request to the social media platform to remove unreliable and negative information or to add a warning message to the information, for example, through the Twitter API.

[0140] Input: Judgment result

[0141] Output: Request sent to the social media platform

[0142] Specific behavior:

[0143] The server sends a request via the Twitter API that includes a warning: "This is false information about the Noto Peninsula earthquake."

[0144] Step 8:

[0145] User Notification

[0146] The device notifies the user of the result of the action received from the server, for example by displaying a notification on the user's smartphone so that the user can check the content.

[0147] Input: Action result

[0148] Output: Notification message sent to the user

[0149] Specific behavior:

[0150] The device will display a notification on the user's smartphone saying, "Unreliable information has been detected and removed. Please check for details."

[0151] (Application example 1)

[0152] 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."

[0153] The speed at which information spreads on the Internet has created an environment in which false and misinformation can easily spread. It is extremely important to detect such false and misinformation in real time and to warn users appropriately. However, conventional technology has made it difficult to do this efficiently and quickly. There is a need to provide a system that can solve these issues, more reliably evaluate the reliability of information, and provide appropriate notifications.

[0154] 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.

[0155] In this invention, the server includes means for collecting information on the Internet, means for preprocessing the collected information, means for analyzing the preprocessed information and performing sentiment analysis and reliability determination, means for requesting deletion and adding a warning message to unreliable and negative information, and means for notifying the user of the results of the action, thereby enabling the effective detection of false information and misinformation in real time and notifying the user.

[0156] "Means for collecting information on the Internet" refers to methods and devices for obtaining the latest posted information from social networking sites, news sites, etc.

[0157] "Means for preprocessing collected information" refers to a processing method and device for removing unnecessary data from the acquired information and converting it into a format that is easy to analyze.

[0158] "Means for analyzing pre-processed information and performing sentiment analysis and credibility determination" refers to methods and apparatus for analyzing tokenized text data using natural language processing techniques to determine sentiment from the context of the text and assess the credibility of the source.

[0159] "Means for executing requests to delete and add warning messages to unreliable and negative information" refers to a method and device for sending a request to delete information determined to be unreliable or a request to add a warning message to a social media platform.

[0160] "Means for notifying users of the results of actions" refers to methods and devices for notifying users in an appropriate format of the results of actions taken to remove or warn against false or misleading information.

[0161] This invention is a system for detecting false and misleading information on the Internet in real time and warning users, and is composed of the following components:

[0162] Data collection

[0163] The server collects the latest posting information from social media platforms, news sites, etc. To collect the data, it uses the API of the social media platform to automatically obtain new posting information.

[0164] Data Preprocessing

[0165] The server filters unnecessary data (HTML tags and special characters) from the retrieved information and tokenizes the text, using the Python libraries BeautifulSoup and NLTK (Natural Language Toolkit) for this process.

[0166] Data analysis

[0167] The server analyzes the tokenized text data using natural language processing technology. Sentiment analysis and credibility assessment are performed here. Sentiment analysis determines and scores positive, negative, or neutral sentiment from the context of the text. Credibility assessment verifies whether the collected information source is trustworthy and assigns a credibility score. This analysis uses Python NLP libraries (spaCy, NLTK).

[0168] Reliability determination

[0169] The server uses natural language processing technology to analyze the data and then determines the reliability of the information. This includes comparing it with trusted sources and checking against past data. If the information is deemed unreliable and negative, it proceeds to the next step.

[0170] Action Execution

[0171] A request to delete or add a warning message to unreliable and negative information is sent to the SNS platform. The server uses the SNS platform API to send the request to delete or add a warning message.

[0172] User Notification

[0173] The device will notify the user that the deletion request has been carried out and will display a warning message, allowing the user to reconfirm the authenticity of the information.

[0174] Specific examples

[0175] For example, if a user sets the keyword "novel coronavirus vaccine hoax," the server will collect posts related to "coronavirus vaccine." The collected data will be tokenized after removing HTML tags and special characters. The collected data will then be analyzed using natural language processing technology to perform sentiment analysis and determine trustworthiness. For posts that are deemed untrustworthy and negative, a request will be sent to the social media platform to delete them, and a request will be sent to display a warning message for suspicious information. The device will then send a warning notification to the user, allowing them to reconfirm the trustworthiness of the information.

[0176] Prompt Sentence Examples

[0177] Collect the latest posts about "COVID-19 vaccines" from social media and news sites. Remove HTML tags and special characters from the collected information and tokenize the text. Next, use natural language processing technology to perform sentiment analysis and credibility assessment to identify unreliable and negative information. Then, send a request to the social media platform to remove the information and add a warning message, and finally send a push notification to the user.

[0178] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0179] Step 1:

[0180] The server uses the APIs of social media and news sites to collect the latest post information based on specified keywords. The data retrieved by the server includes metadata such as the post text, poster information, and posting date and time. The input for this process is keywords related to "novel coronavirus vaccine hoaxes," and the output is the raw post data.

[0181] Step 2:

[0182] The server filters unnecessary information from the collected data. Specifically, it removes HTML tags and special characters to generate clean text data. This process uses the Python library BeautifulSoup. The input is the raw post data collected in step 1, and the output is the filtered text data.

[0183] Step 3:

[0184] The server tokenizes the filtered data, splitting the text into words and converting it into a format that is easy to use with natural language processing. This process uses NLTK (Natural Language Toolkit). The input is the clean text data generated in step 2, and the output is the tokenized text data.

[0185] Step 4:

[0186] The server analyzes the tokenized data using natural language processing technology. This involves sentiment analysis and credibility assessment. Sentiment analysis involves scoring positive, negative, or neutral sentiment from the context of the text. Credibility assessment verifies whether the collected information source is trustworthy and assigns a credibility score. This process uses Python NLP libraries (spaCy, NLTK). The input is the tokenized text data generated in step 3, and the output is a sentiment score and a credibility score.

[0187] Step 5:

[0188] The server determines the reliability of the information based on the analysis results. If the reliability is low and the information is determined to be negative, it proceeds to the next step. The input is the emotion score and reliability score generated in step 4, and the output is the information whose reliability and emotion have been determined.

[0189] Step 6:

[0190] The server sends a request to the SNS platform to delete and add a warning message to unreliable and negative information. This process uses the SNS platform's API. The input is the information determined in step 5, and the output is the result of sending the request to the SNS platform.

[0191] Step 7:

[0192] The device notifies the user of the results of the deletion request or warning message addition request received from the server. The notification is performed using a push notification. The input is the request result returned from the SNS platform in step 6, and the output is a push notification sent to the user's device.

[0193] 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.

[0194] This invention is a system that detects fake and misleading information on the Internet in real time and warns users. By combining this system with an emotion engine that recognizes user emotions, it can comprehensively consider the reliability of information and user emotions.

[0195] The system of the present invention comprises the following means:

[0196] Data collection

[0197] The server collects public information from social media and the internet based on pre-specified keywords. Data collection is done by automatically retrieving new posts using the API of the social media platform. The retrieved data is temporarily stored in a database.

[0198] Data Preprocessing

[0199] The server pre-processes the collected data to remove unnecessary information, including removing HTML tags, removing special characters, and tokenizing the text, converting the data into a format that is easier to analyze.

[0200] Data analysis

[0201] The server then analyzes the preprocessed data using natural language processing (NLP) techniques. Analysis items include keyword frequency analysis, sentiment analysis, and information source verification. Sentiment analysis calculates a positive, negative, or neutral sentiment score based on the context of the text.

[0202] Reliability determination

[0203] The server evaluates the reliability of the information based on the data analysis results. It compares the data with reliable sources (e.g., official announcements, well-known news sites) and calculates a reliability score. Information with low reliability and negative results proceeds to the next step.

[0204] Using the Emotion Engine

[0205] The server recognizes the user's feelings toward the information using an emotion engine, which analyzes emotions from the user's browsing history and posted content to understand the potential emotional response to specific information.

[0206] Action Execution

[0207] The server sends a request to the social media platform to remove unreliable and negative information. It also sends a request to add a warning message to suspicious information, taking into account user sentiment. Based on this request, the social media platform deletes the post or displays a warning message.

[0208] User Notification

[0209] The terminal executes deletion requests and displays warning messages based on requests from the server. When a post is deleted, the poster is notified of the deletion, and when a post has a warning message attached, a warning message is displayed to warn viewers.

[0210] Specific examples

[0211] For example, suppose a server collects posts related to the "Noto Peninsula Earthquake." The collected data is stripped of HTML tags and special characters and the text is tokenized. Next, natural language processing technology is used to analyze the collected data, performing sentiment analysis and determining reliability. If the information is determined to be unreliable and negative, a request to delete the post is sent to the social media platform. A request is also sent to add a warning message, taking into account user sentiment, if necessary. Finally, the device displays a deletion or warning notice to the user, allowing them to confirm and take appropriate action.

[0212] This system will prevent the spread of false and misinformation and issue warnings that take user sentiment into account, allowing users to obtain accurate and reliable information and improving the information environment on the Internet.

[0213] The processing flow will be explained below.

[0214] Step 1:

[0215] The server collects public information from social media and the internet. This information collection is done by using the API of the social media platform to obtain new posts containing specified keywords (e.g., "Noto Peninsula Earthquake"). The collected data is temporarily stored in a database.

[0216] Step 2:

[0217] The server pre-processes the collected data to remove unnecessary information, such as removing HTML tags, eliminating special characters, and tokenizing the text, converting the data into a format that is easier to analyze.

[0218] Step 3:

[0219] The server then analyzes the preprocessed data using natural language processing (NLP) techniques. Key analysis items include keyword frequency analysis, context analysis, and sentiment analysis. Sentiment analysis calculates a positive, negative, or neutral sentiment score.

[0220] Step 4:

[0221] The server performs a credibility assessment: it compares the preprocessed data with trusted sources (e.g., official announcements, well-known news sites) and calculates a credibility score. This assessment determines whether the information is trustworthy.

[0222] Step 5:

[0223] The server uses an emotion engine to recognize the user's emotions toward the information. This emotion engine analyzes emotions from the user's browsing history and posted content to understand the potential emotional response to the information. This step is particularly effective when the authenticity of the information is unknown.

[0224] Step 6:

[0225] The server takes action against unreliable and negative information, specifically by sending a request to the social media platform to delete the relevant social media post, and also by sending a request to add a warning message to any suspicious information.

[0226] Step 7:

[0227] The terminal executes deletion requests and displays warning messages based on requests from the server. When a post is deleted, the poster is notified of the deletion, and when a post has a warning message attached, a warning message is displayed to warn viewers.

[0228] Step 8:

[0229] Users can check the warning messages and removal notices displayed on their devices, which allows them to reconfirm the reliability of the information provided and make an appropriate decision.

[0230] In this way, we can build a system that can detect fake and misleading information on the Internet in real time and respond quickly. In addition, by using an emotion engine, it is possible to take user emotions into account and provide more appropriate warnings and notifications.

[0231] Example 2

[0232] 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."

[0233] In recent years, a lot of false and misleading information has been spread on the Internet, resulting in an increasing number of cases where users believe the incorrect information. Furthermore, the impact of this information on users' emotions is also a problem that cannot be ignored. Conventional technologies lacked sufficient mechanisms to prevent the spread of false information and lacked measures that took user emotions into consideration.

[0234] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0235] In this invention, the server includes means for collecting information on the Internet, means for preprocessing the collected information, means for analyzing the preprocessed information and performing sentiment analysis and reliability determination, means for taking action on unreliable and negative information, means for recognizing user sentiment and grasping emotional reactions to the information, and means for notifying the user of the results of the action. This makes it possible to prevent the spread of false and misleading information on the Internet and to take warnings and countermeasures that take user sentiment into consideration.

[0236] "Information on the Internet" refers to data such as text, images, audio, and video that is publicly available on the Internet, including websites, social media platforms, blogs, and news sites.

[0237] "Means of collection" refers to methods and tools for obtaining information on the Internet, such as collecting data using APIs or web scraping technology.

[0238] "Preprocessing means" refers to processing methods and tools used to remove unnecessary data from collected information and convert it into a format that is easy to analyze.

[0239] "Means of analysis" refers to techniques for analyzing preprocessed data and extracting useful information. Specifically, this includes natural language processing, sentiment analysis, keyword frequency analysis, etc.

[0240] "Sentiment analysis" is a technique for identifying and scoring positive, negative, and neutral emotions from text data.

[0241] "Credibility assessment" is a method of evaluating how trustworthy information is based on its source and content.

[0242] "Means for taking action" refers to measures to take against unreliable and negative information, including sending a removal request or adding a warning message.

[0243] "Means for recognizing user emotions" refers to technology that analyzes users' browsing history and posted content to understand their emotional reactions to specific information.

[0244] "Means for notifying the user of the outcome of the action" means a method for informing the user about the action taken, including, for example, displaying a removal notice or a warning message.

[0245] "Filtering means" refers to processing methods or tools for removing unnecessary data from collected information.

[0246] "Special character removal" is the process of removing unnecessary symbols and special characters from text data.

[0247] "Text tokenization" is the process of dividing text data into units of words or phrases.

[0248] "Natural language processing technology" refers to technology that enables computers to understand, interpret, and generate human language. Examples include text analysis, context understanding, and machine translation.

[0249] A "trustworthiness score" is a numerical representation of the reliability of information based on an evaluation of its source and content.

[0250] An "emotion engine" is a tool or algorithm that analyzes a user's emotions and identifies their emotional response to specific information.

[0251] A "removal request" is an action requesting that a social media platform remove a specific post.

[0252] A "warning message" is a message that alerts the user to specific information.

[0253] This invention is a system that detects fake and misleading information on the Internet in real time and warns users. By combining this system with an emotion engine that recognizes user emotions, it can comprehensively consider the reliability of information and user emotions.

[0254] First, the server collects information from the Internet. Specifically, it uses the API of a social networking platform (for example, Twitter API or general web scraping methods) to automatically retrieve new posts based on specified keywords. This collected data is temporarily stored in a database (for example, MySQL or MongoDB).

[0255] The server then preprocesses the collected information, which includes removing HTML tags, stripping special characters, and tokenizing the text using Python's BeautifulSoup, regular expressions, and NLTK libraries, converting the data into a format that is easier to parse.

[0256] The server then analyzes the preprocessed data. Specifically, it uses natural language processing techniques (e.g., Spacy or BERT) to perform sentiment analysis and keyword frequency analysis of the text. It also verifies the source of the information and calculates a credibility score by comparing it with highly reliable sources (e.g., official announcements or well-known news sites).

[0257] If the information is judged to be unreliable and negative, the server uses an emotion engine to recognize the user's emotions. This engine analyzes emotions from the user's browsing history and posted content to understand the potential emotional reaction to specific information. This process uses Google Analytics and emotion analysis APIs (e.g., IBM Watson Tone Analyzer).

[0258] The server then takes action against unreliable and negative information, sending a request to the social media platform to remove it and also sending a request to add a warning message to any suspicious information.

[0259] Finally, the device notifies the user of deletion requests or warning messages based on requests from the server. For deleted posts, the device notifies the poster, and for posts with attached warning messages, the device displays a warning message to the viewer.

[0260] Specific examples

[0261] For example, consider a case where a server collects posts related to the "Noto Peninsula Earthquake." The collected data is preprocessed by removing HTML tags and special characters and tokenizing the text. Next, natural language processing techniques (e.g., Spacy or BERT) are used to analyze sentiment and determine reliability. If the information is determined to be unreliable and negative, a deletion request for the post is sent to the social media platform. A request with a warning message attached is also sent, taking into account the user's sentiment. The device displays the deletion or warning notification to the user, allowing the user to review it and take appropriate action.

[0262] Examples of prompt statements

[0263] Collect and preprocess the latest posts related to the "Noto Peninsula Earthquake." Next, use Spacy and BERT to perform sentiment analysis and determine trustworthiness, and create a deletion request and warning message for posts with low trustworthiness and negative views. Finally, notify users.

[0264] This system will prevent the spread of false and misinformation and will be able to issue warnings and take measures that take user sentiment into account, allowing users to obtain accurate and reliable information and improving the information environment on the Internet.

[0265] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0266] Step 1: Data collection

[0267] The server collects information from the Internet. Specifically, it uses the API of social media platforms and web scraping technology to automatically retrieve new posts based on specified keywords. The input is a specific keyword (e.g., "Noto Peninsula Earthquake"). The output is the collected post data (e.g., text, user name, posting date and time, etc.).

[0268] Step 2: Data Preprocessing

[0269] The server preprocesses the collected information. Specifically, it uses BeautifulSoup to remove HTML tags and regular expressions to remove special characters. It also uses NLTK to tokenize the text. The input for this step is the collected post data. The output is the preprocessed, clean text data.

[0270] Step 3: Data analysis

[0271] The server analyzes the preprocessed data. Specifically, it performs sentiment and keyword frequency analysis on the text using Spacy and BERT. It also compares it with established news sites and official announcements to verify the source of the information. The input for this step is the preprocessed, clean text data. The output is a sentiment score, keyword frequency, and confidence score.

[0272] Step 4: Reliability determination

[0273] The server evaluates the reliability of the information based on the data analysis results. Specifically, it checks whether the information matches a reliable source based on the obtained reliability score. The inputs to this step are the sentiment score, keyword frequency, and reliability score. The output is a judgment result of whether the information is highly or low reliable.

[0274] Step 5: Use the Emotion Engine

[0275] The server uses an emotion engine to recognize the user's emotions toward the information. It uses Google Analytics and an emotion analysis API to analyze emotions from the user's browsing history and posted content, and understands the emotional reaction to specific information. The inputs to this step are the credibility judgment results and the user's past browsing history and posted data. The output is the user's emotional reaction pattern.

[0276] Step 6: Take Action

[0277] The server takes action against unreliable and negative information by sending a deletion request to the social media platform and adding a warning message to suspicious information. The inputs to this step are the reliability judgment results and the user's emotional reaction patterns. As outputs, a deletion request and a warning message request are sent to the social media platform.

[0278] Step 7: Notify users

[0279] The device executes the deletion request or displays the warning message based on the request from the server. Specifically, the notification is displayed using JavaScript or HTML. The input to this step is the deletion request and warning message request from the server. The output is the deletion notification or warning notification displayed to the user.

[0280] This processing flow makes it possible to evaluate the reliability of information on the Internet and provide appropriate measures and notifications to users.

[0281] (Application example 2)

[0282] 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."

[0283] In the electronic payment process, users often make transactions based on unreliable information or fake reviews, which can lead to unreliable transactions and fraud. To solve this problem, a system is needed that can evaluate the reliability of information related to transactions in real time and warn users in a timely manner.

[0284] 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.

[0285] In this invention, the server includes means for collecting information from the Internet, means for preprocessing the collected information, means for analyzing the preprocessed information and performing sentiment analysis and reliability determination, means for taking action on information that is low in reliability and negative, means for notifying the user of the results of the action, and means for evaluating the reliability of transaction information in the payment system and displaying a warning message. This makes it possible to evaluate the reliability of transaction information during electronic payment and issue an appropriate warning to the user.

[0286] The "Internet" is a system of interconnected computer networks around the world that enables the exchange of information.

[0287] "Information collection means" is a system that automatically acquires data from the Internet and uses APIs to collect the necessary information.

[0288] "Preprocessing" refers to the process of converting collected information into a format that is easier to analyze, by removing unnecessary data and special characters and tokenizing the text.

[0289] The "analysis method" is a system that uses preprocessed data to evaluate the reliability of information and emotions, and is carried out using natural language processing technology.

[0290] "Sentiment analysis" is a technique for reading emotions from text data and identifying positive, negative, and neutral emotions.

[0291] "Credibility determination" is the process of assessing the trustworthiness of a source and calculating a credibility score, which is then compared to other trusted sources.

[0292] "Negative information" refers to information that may have a negative impact on users and is identified through the results of sentiment analysis.

[0293] The "action execution means" is a mechanism for performing specific processing on information that is determined to be unreliable and negative, such as adding a warning message or sending a deletion request.

[0294] "Notification means" refers to the function of informing the user of the results of an action taken, and is performed through an application or device.

[0295] A "payment system" is a system for electronically conducting monetary transactions and allowing users to purchase goods and services online.

[0296] A "warning message" is a message that informs users that the information is unreliable and helps them avoid inappropriate transactions.

[0297] The system for realizing this invention is composed of three main elements: a server, a terminal, and a user. The specific configuration and operation of the system are explained in detail below.

[0298] The server collects information from the Internet, preprocesses it, and performs sentiment analysis and reliability assessment. The data is collected using APIs from social media and review sites. For example, the Twitter API and Facebook Graph API are used to collect reviews and posts based on specific keywords. The collected data is then temporarily stored in a MySQL database.

[0299] The server then pre-processes the collected information, which includes removing HTML tags and special characters, and tokenizing the text to convert the data into a format that is easier to parse. This process is performed using the Python libraries NLTK and SpaCy.

[0300] The preprocessed data is then analyzed using natural language processing (NLP) techniques. Specifically, the text is subjected to sentiment analysis and credibility assessment. Sentiment analysis involves identifying positive, negative, and neutral sentiment from the text and calculating a sentiment score. TextBlob and IBM Watson Emotion Analysis are used for this task. Credibility assessment involves comparing the data with official sources (e.g., well-known news sites or official announcements) and calculating a credibility score.

[0301] Actions are taken against information that is deemed unreliable and negative. These actions include sending a deletion request to the social media platform or displaying a warning message to the user. After the action is taken, the result is notified to the user's device. This notification allows the user to receive appropriate information and avoid inappropriate transactions.

[0302] As an example of how this system can be used, consider the electronic payment app "SecurePay Warning System." When a user browses a specific product or store through this app, the server collects reviews related to that product or store, performs sentiment analysis, and determines its trustworthiness. If it is determined to be untrustworthy, a warning message is displayed on the user's device. This warning allows the user to make transactions safely.

[0303] As an example, use the following prompt:

[0304] Please analyze this review body: "{review body}" and provide a rating and sentiment score. User ID: {user ID}, Purchase history: {purchase history}"

[0305] The system will protect users from false and misinformation online and allow them to make decisions based on more reliable information.

[0306] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0307] Step 1:

[0308] The server collects information from the Internet. Specifically, it uses the APIs of social media and review sites to collect reviews and posts based on specified keywords. For example, it uses the Twitter API or Facebook Graph API to obtain posts related to a "specific product name." The collected data is then temporarily stored in a MySQL database.

[0309] Input: Keywords, SNS API request

[0310] Output: Collected reviews and submission data (JSON format)

[0311] Step 2:

[0312] The server preprocesses the collected information, removing HTML tags and special characters, and tokenizing the text. It uses the Python libraries NLTK and SpaCy to convert the collected data into a format that is easy to analyze.

[0313] Input: Collected reviews and submission data

[0314] Output: Preprocessed text data (cleaned)

[0315] Step 3:

[0316] The server analyzes the preprocessed data. Based on the preprocessed text data, natural language processing techniques are used to perform sentiment analysis and reliability determination. For sentiment analysis, TextBlob and IBM Watson Emotion Analysis are used to calculate positive, negative, or neutral sentiment scores from the text. For reliability determination, a reliability score is calculated by comparing it with official sources.

[0317] Input: Preprocessed text data

[0318] Output: sentiment score, confidence score

[0319] Step 4:

[0320] The server takes action against unreliable and negative information. Specifically, it sends a deletion request to the social media platform or a request to add a warning message. If the deletion request is accepted, the information is deleted from the platform. In addition, information with a warning message is displayed to warn viewers.

[0321] Input: sentiment score, confidence score

[0322] Output: Delete request, warning message request

[0323] Step 5:

[0324] The terminal will inform the user of the results of the action. Warning messages will be displayed for unreliable information, allowing the user to avoid inappropriate transactions. Warning messages will be delivered to the user through the mobile application, helping them to carry out safe transactions.

[0325] Input: Delete request result, Warning message request result

[0326] Output: User notification (warning message)

[0327] Step 6:

[0328] The user will review the notification and take appropriate action based on its content, for example, choosing to avoid the transaction if a warning message is displayed, or confirming that the deletion request has been accepted and the information has been deleted.

[0329] Input: User notification (warning message)

[0330] Output: User decision

[0331] 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.

[0332] 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.

[0333] 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.

[0334] [Second embodiment]

[0335] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0336] 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.

[0337] 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).

[0338] 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.

[0339] 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.

[0340] 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).

[0341] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0342] 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.

[0343] 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.

[0344] 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.

[0345] 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.

[0346] 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."

[0347] This invention is a system for detecting false and misleading information on the Internet in real time and issuing warnings to users. The system includes a server that collects information on the Internet, preprocesses the collected information, analyzes the preprocessed information, takes action against unreliable and negative information, and notifies users of the results.

[0348] Specifically, the program of this system performs the following processes.

[0349] Data collection

[0350] The server collects public information from social media and the internet based on pre-specified keywords, using the API of the social media platform to automatically retrieve new posts.

[0351] Data Preprocessing

[0352] The server then removes unnecessary information from the collected data and converts it into a format that is easier to analyze, by performing preprocessing such as removing HTML tags, eliminating special characters, and tokenizing the text.

[0353] Data analysis

[0354] The server analyzes the preprocessed data using natural language processing technology, extracting important keywords, analyzing sentiment, and assessing trustworthiness. Sentiment analysis determines and scores positive, negative, or neutral sentiment from the context of the text. Credibility assessment verifies whether the collected information source is trustworthy and assigns a trustworthiness score.

[0355] Reliability determination

[0356] The server determines the reliability of the information based on the results of the data analysis. This includes comparing it with reliable sources, checking against past data, etc. If the information is deemed unreliable and negative, it proceeds to the next step.

[0357] Action Execution

[0358] The server sends a request to the social media platform to remove unreliable and negative information, and also requests that a warning message be added to suspicious information. Based on this request, the social media platform will delete the post or display a warning message.

[0359] User Notification

[0360] The device will notify the user that the deletion request has been carried out and will display a warning message, allowing the user to review the notification and reconfirm the authenticity of the information.

[0361] Specific examples

[0362] For example, suppose a server collects posts about the "Noto Peninsula Earthquake." After removing HTML tags and special characters from the collected data, the text is tokenized. Next, the collected data is analyzed using natural language processing technology to perform sentiment analysis and determine trustworthiness. For posts that are determined to be unreliable and contain negative information based on comparison with past reliable sources, a request is sent to the social media platform to delete them. In addition, for suspicious information, a request is sent to display a warning message. This allows the device to notify the user of the warning and reconfirm the authenticity of the information.

[0363] This system will effectively prevent the spread of false and misinformation and provide users with accurate information.

[0364] The processing flow will be explained below.

[0365] Step 1:

[0366] The server collects public information from social media and the internet. This information collection is done by using the API of the social media platform to obtain new posts containing specified keywords (e.g., "Noto Peninsula Earthquake"). The collected data is temporarily stored in a database.

[0367] Step 2:

[0368] The server pre-processes the collected data to remove unnecessary information, such as removing HTML tags, eliminating special characters, and tokenizing the text, converting the data into a format that is easier to analyze.

[0369] Step 3:

[0370] The server then analyzes the preprocessed data using natural language processing (NLP) techniques. The main analysis items include keyword frequency analysis, sentiment analysis, and information source verification. Sentiment analysis produces a positive, negative, or neutral sentiment score.

[0371] Step 4:

[0372] The server evaluates the reliability of the information based on the data analysis results. It compares the data with reliable sources (e.g., official announcements, well-known news sites) and calculates a reliability score. If the reliability is low, it proceeds to the next step.

[0373] Step 5:

[0374] The server takes action against unreliable and negative information, specifically by sending a deletion request to the social media platform for the relevant social media post, and by sending a request to display a warning message for sensitive information.

[0375] Step 6:

[0376] The terminal executes deletion requests and displays warning messages based on requests from the server. When a post is deleted, the poster is notified of the deletion, and when a post has a warning message attached, a warning message is displayed to warn viewers.

[0377] Step 7:

[0378] Users can check the warning messages and deletion notices displayed on their devices, reconfirm the reliability of the information provided, and make appropriate decisions.

[0379] In this way, a system will be built that can detect fake and misleading information on the Internet in real time and respond quickly.

[0380] Example 1

[0381] 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."

[0382] In today's internet environment, the rapid spread of false and misleading information has become a social problem. Accurate information and false information are often mixed together on social media platforms, making it difficult for users to determine the reliability of the information. Therefore, there is a need for methods to quickly detect and respond to unreliable information.

[0383] 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.

[0384] In this invention, the server includes a means for collecting information from the Internet, a means for preprocessing the collected information, a means for analyzing the preprocessed information using natural language processing technology to perform sentiment analysis and reliability determination, a means for deleting unreliable and negative information or adding a warning message to the information, and a means for notifying the user of the results of the action, thereby enabling the rapid detection and response to false information and misinformation.

[0385] "Means for collecting information on the Internet" refers to a mechanism for obtaining data based on specified keywords from websites, social networking services (SNS), and other public information on the Internet.

[0386] The "means for preprocessing collected information" refers to a mechanism for performing a process to remove unnecessary information from the collected data and convert it into an analyzable format.

[0387] "Means of analysis using natural language processing technology" refers to a system that utilizes algorithms and models to automatically perform sentiment analysis and reliability assessment on collected text data.

[0388] "Sentiment analysis" is a technique that determines positive, negative, or neutral sentiment from the context of text and assigns a score to each post.

[0389] "Credibility assessment" is the process of assessing the reliability of collected information sources and calculating a reliability score by matching and comparing them with past data.

[0390] "Means for taking action to delete or add a warning message" refers to a mechanism for sending a request to a social media platform to delete or display a warning message for unreliable and negative information.

[0391] "Means for notifying the user of the results of an action" refers to a mechanism that notifies the user of the results of an action taken on their device, allowing the user to verify the reliability and authenticity of the information.

[0392] This invention is a system that detects false information and misinformation on the Internet in real time and warns users. This system includes a series of processes that collect, process, and notify data between servers, terminals, and users.

[0393] The server first uses the API of a social media platform as a means of collecting information on the Internet. For example, it can use the Twitter API to collect tweets related to the "Great Noto Peninsula Earthquake." This allows the server to obtain new posts in real time. As a specific example, when a server collects posts related to the "Great Noto Peninsula Earthquake," it uses the Twitter API to collect tweets containing the relevant keyword.

[0394] Next, the server preprocesses the collected information. To remove unnecessary information from the collected data and convert it into a format that is easier to analyze, it uses the BeautifulSoup library to remove HTML tags and special characters, and then uses the NLTK library to tokenize the text. For example, it preprocesses the tweet "Information about the Noto Peninsula earthquake can be found here → [link]" by removing the link and separating it into words.

[0395] The preprocessed data is then analyzed using natural language processing techniques. The server performs sentiment analysis using Hugging Face Transformers to determine positive, negative, or neutral sentiment. It also evaluates the reliability of the collected information sources and assigns a credibility score. For example, the server might perform sentiment analysis on a tweet such as "This earthquake is having a major impact on people" and assign it a negative score.

[0396] The server then combines these analysis results to determine the reliability of the information. For unreliable and negative information, it sends a deletion request to the social media platform, and for suspicious information, it adds a warning message. Users can request the deletion of specific tweets or the display of warning messages via the Twitter API. For example, they can send a request with a warning that "This is a hoax about the Noto Peninsula earthquake."

[0397] Finally, the server notifies the user of the results of these actions. The device displays a notification on the user's smartphone or other device, allowing the user to check the content of the notification. For example, the device may send the user a notification message such as, "Unreliable information has been detected and deleted. Please check for details."

[0398] This system enables the rapid detection and response of false information and misinformation, and provides accurate information to users. For example, keywords can be specified by inputting the following prompt sentences into the generative AI model:

[0399] Prompt Sentence Examples

[0400] Collect the latest information on the "Noto Peninsula Earthquake" and be alerted to unreliable and negative information.

[0401] Based on this prompt, the system will detect fake and misinformation related to the specified keywords in real time and notify the user.

[0402] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0403] Step 1:

[0404] Data Collection Settings

[0405] The server loads a pre-specified list of keywords to be used for data collection. For example, the keyword "Noto Peninsula Earthquake" is added to the list. This list is used for subsequent API queries.

[0406] Input: A specified list of keywords

[0407] Output: Keyword list loaded

[0408] Specific behavior:

[0409] The server retrieves "Noto Peninsula Earthquake" from the keyword list and sets up data collection.

[0410] Step 2:

[0411] Using SNS API

[0412] The server uses the API of the social media platform to retrieve posts based on the target keywords. For example, the Twitter API is used to collect tweets related to the "Noto Peninsula Earthquake." The server uses the Streaming API to retrieve new posts in real time.

[0413] Input: Keywords to be collected

[0414] Output: Collected SNS post data

[0415] Specific behavior:

[0416] The server uses Twitter's Streaming API to retrieve tweets related to the keyword "Noto Peninsula Earthquake" in real time.

[0417] Step 3:

[0418] Data Preprocessing

[0419] The server cleans the collected data and converts it into a format that is easy to analyze, using BeautifulSoup to remove HTML tags and special characters, and the NLTK library to tokenize the text.

[0420] Input: Collected social media posting data

[0421] Output: Preprocessed text data

[0422] Specific behavior:

[0423] The server removes the link from the tweet "Information about the Noto Peninsula earthquake can be found here → [link]" and separates the text.

[0424] Step 4:

[0425] Conducting sentiment analysis

[0426] The server analyzes the preprocessed text data using natural language processing techniques. Specifically, it uses Hugging Face Transformers to perform sentiment analysis on the text data and determine positive, negative, or neutral sentiment. It assigns a sentiment score to each piece of text.

[0427] Input: Preprocessed text data

[0428] Output: Text data with sentiment scores

[0429] Specific behavior:

[0430] The server performs sentiment analysis on tweets such as "This earthquake is having a big impact on people" and assigns them a negative score.

[0431] Step 5:

[0432] Conducting reliability evaluation

[0433] The server compares the collected information with a historical database and calculates a reliability score to assess the reliability of the information source.

[0434] Input: Text data with sentiment scores

[0435] Output: Text data with confidence scores

[0436] Specific behavior:

[0437] The server references the poster's past reliability data and calculates a reliability score.

[0438] Step 6:

[0439] Integration of reliability judgments

[0440] The server combines the sentiment score and the reliability score to determine the overall reliability of the information. For information that is low in reliability and negative, it proceeds to the next step.

[0441] Input: Text data with confidence scores

[0442] Output: Judgment result

[0443] Specific behavior:

[0444] The server selects posts with a "negative" sentiment score and a low credibility score.

[0445] Step 7:

[0446] Execute Action

[0447] The server sends a request to the social media platform to remove unreliable and negative information or to add a warning message to the information, for example, through the Twitter API.

[0448] Input: Judgment result

[0449] Output: Request sent to the social media platform

[0450] Specific behavior:

[0451] The server sends a request via the Twitter API that includes a warning: "This is false information about the Noto Peninsula earthquake."

[0452] Step 8:

[0453] User Notification

[0454] The device notifies the user of the result of the action received from the server, for example by displaying a notification on the user's smartphone so that the user can check the content.

[0455] Input: Action result

[0456] Output: Notification message sent to the user

[0457] Specific behavior:

[0458] The device will display a notification on the user's smartphone saying, "Unreliable information has been detected and removed. Please check for details."

[0459] (Application example 1)

[0460] 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."

[0461] The speed at which information spreads on the Internet has created an environment in which false and misinformation can easily spread. It is extremely important to detect such false and misinformation in real time and to warn users appropriately. However, conventional technology has made it difficult to do this efficiently and quickly. There is a need to provide a system that can solve these issues, more reliably evaluate the reliability of information, and provide appropriate notifications.

[0462] 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.

[0463] In this invention, the server includes means for collecting information on the Internet, means for preprocessing the collected information, means for analyzing the preprocessed information and performing sentiment analysis and reliability determination, means for requesting deletion and adding a warning message to unreliable and negative information, and means for notifying the user of the results of the action, thereby enabling the effective detection of false information and misinformation in real time and notifying the user.

[0464] "Means for collecting information on the Internet" refers to methods and devices for obtaining the latest posted information from social networking sites, news sites, etc.

[0465] "Means for preprocessing collected information" refers to a processing method and device for removing unnecessary data from the acquired information and converting it into a format that is easy to analyze.

[0466] "Means for analyzing pre-processed information and performing sentiment analysis and credibility determination" refers to methods and apparatus for analyzing tokenized text data using natural language processing techniques to determine sentiment from the context of the text and assess the credibility of the source.

[0467] "Means for executing requests to delete and add warning messages to unreliable and negative information" refers to a method and device for sending a request to delete information determined to be unreliable or a request to add a warning message to a social media platform.

[0468] "Means for notifying users of the results of actions" refers to methods and devices for notifying users in an appropriate format of the results of actions taken to remove or warn against false or misleading information.

[0469] This invention is a system for detecting false and misleading information on the Internet in real time and warning users, and is composed of the following components:

[0470] Data collection

[0471] The server collects the latest posting information from social media platforms, news sites, etc. To collect the data, it uses the API of the social media platform to automatically obtain new posting information.

[0472] Data Preprocessing

[0473] The server filters unnecessary data (HTML tags and special characters) from the retrieved information and tokenizes the text, using the Python libraries BeautifulSoup and NLTK (Natural Language Toolkit) for this process.

[0474] Data analysis

[0475] The server analyzes the tokenized text data using natural language processing technology. Sentiment analysis and credibility assessment are performed here. Sentiment analysis determines and scores positive, negative, or neutral sentiment from the context of the text. Credibility assessment verifies whether the collected information source is trustworthy and assigns a credibility score. This analysis uses Python NLP libraries (spaCy, NLTK).

[0476] Reliability determination

[0477] The server uses natural language processing technology to analyze the data and then determines the reliability of the information. This includes comparing it with trusted sources and checking against past data. If the information is deemed unreliable and negative, it proceeds to the next step.

[0478] Action Execution

[0479] A request to delete or add a warning message to unreliable and negative information is sent to the SNS platform. The server uses the SNS platform API to send the request to delete or add a warning message.

[0480] User Notification

[0481] The device will notify the user that the deletion request has been carried out and will display a warning message, allowing the user to reconfirm the authenticity of the information.

[0482] Specific examples

[0483] For example, if a user sets the keyword "novel coronavirus vaccine hoax," the server will collect posts related to "coronavirus vaccine." The collected data will be tokenized after removing HTML tags and special characters. The collected data will then be analyzed using natural language processing technology to perform sentiment analysis and determine trustworthiness. For posts that are deemed untrustworthy and negative, a request will be sent to the social media platform to delete them, and a request will be sent to display a warning message for suspicious information. The device will then send a warning notification to the user, allowing them to reconfirm the trustworthiness of the information.

[0484] Prompt Sentence Examples

[0485] Collect the latest posts about "COVID-19 vaccines" from social media and news sites. Remove HTML tags and special characters from the collected information and tokenize the text. Next, use natural language processing technology to perform sentiment analysis and credibility assessment to identify unreliable and negative information. Then, send a request to the social media platform to remove the information and add a warning message, and finally send a push notification to the user.

[0486] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0487] Step 1:

[0488] The server uses the APIs of social media and news sites to collect the latest post information based on specified keywords. The data retrieved by the server includes metadata such as the post text, poster information, and posting date and time. The input for this process is keywords related to "novel coronavirus vaccine hoaxes," and the output is the raw post data.

[0489] Step 2:

[0490] The server filters unnecessary information from the collected data. Specifically, it removes HTML tags and special characters to generate clean text data. This process uses the Python library BeautifulSoup. The input is the raw post data collected in step 1, and the output is the filtered text data.

[0491] Step 3:

[0492] The server tokenizes the filtered data, splitting the text into words and converting it into a format that is easy to use with natural language processing. This process uses NLTK (Natural Language Toolkit). The input is the clean text data generated in step 2, and the output is the tokenized text data.

[0493] Step 4:

[0494] The server analyzes the tokenized data using natural language processing technology. This involves sentiment analysis and credibility assessment. Sentiment analysis involves scoring positive, negative, or neutral sentiment from the context of the text. Credibility assessment verifies whether the collected information source is trustworthy and assigns a credibility score. This process uses Python NLP libraries (spaCy, NLTK). The input is the tokenized text data generated in step 3, and the output is a sentiment score and a credibility score.

[0495] Step 5:

[0496] The server determines the reliability of the information based on the analysis results. If the reliability is low and the information is determined to be negative, it proceeds to the next step. The input is the emotion score and reliability score generated in step 4, and the output is the information whose reliability and emotion have been determined.

[0497] Step 6:

[0498] The server sends a request to the SNS platform to delete and add a warning message to unreliable and negative information. This process uses the SNS platform's API. The input is the information determined in step 5, and the output is the result of sending the request to the SNS platform.

[0499] Step 7:

[0500] The device notifies the user of the results of the deletion request or warning message addition request received from the server. The notification is performed using a push notification. The input is the request result returned from the SNS platform in step 6, and the output is a push notification sent to the user's device.

[0501] 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.

[0502] This invention is a system that detects fake and misleading information on the Internet in real time and warns users. By combining this system with an emotion engine that recognizes user emotions, it can comprehensively consider the reliability of information and user emotions.

[0503] The system of the present invention comprises the following means:

[0504] Data collection

[0505] The server collects public information from social media and the internet based on pre-specified keywords. Data collection is done by automatically retrieving new posts using the API of the social media platform. The retrieved data is temporarily stored in a database.

[0506] Data Preprocessing

[0507] The server pre-processes the collected data to remove unnecessary information, including removing HTML tags, removing special characters, and tokenizing the text, converting the data into a format that is easier to analyze.

[0508] Data analysis

[0509] The server then analyzes the preprocessed data using natural language processing (NLP) techniques. Analysis items include keyword frequency analysis, sentiment analysis, and information source verification. Sentiment analysis calculates a positive, negative, or neutral sentiment score based on the context of the text.

[0510] Reliability determination

[0511] The server evaluates the reliability of the information based on the data analysis results. It compares the data with reliable sources (e.g., official announcements, well-known news sites) and calculates a reliability score. Information with low reliability and negative results proceeds to the next step.

[0512] Using the Emotion Engine

[0513] The server recognizes the user's feelings toward the information using an emotion engine, which analyzes emotions from the user's browsing history and posted content to understand the potential emotional response to specific information.

[0514] Action Execution

[0515] The server sends a request to the social media platform to remove unreliable and negative information. It also sends a request to add a warning message to suspicious information, taking into account user sentiment. Based on this request, the social media platform deletes the post or displays a warning message.

[0516] User Notification

[0517] The terminal executes deletion requests and displays warning messages based on requests from the server. When a post is deleted, the poster is notified of the deletion, and when a post has a warning message attached, a warning message is displayed to warn viewers.

[0518] Specific examples

[0519] For example, suppose a server collects posts related to the "Noto Peninsula Earthquake." The collected data is stripped of HTML tags and special characters and the text is tokenized. Next, natural language processing technology is used to analyze the collected data, performing sentiment analysis and determining reliability. If the information is determined to be unreliable and negative, a request to delete the post is sent to the social media platform. A request is also sent to add a warning message, taking into account user sentiment, if necessary. Finally, the device displays a deletion or warning notice to the user, allowing them to confirm and take appropriate action.

[0520] This system will prevent the spread of false and misinformation and issue warnings that take user sentiment into account, allowing users to obtain accurate and reliable information and improving the information environment on the Internet.

[0521] The processing flow will be explained below.

[0522] Step 1:

[0523] The server collects public information from social media and the internet. This information collection is done by using the API of the social media platform to obtain new posts containing specified keywords (e.g., "Noto Peninsula Earthquake"). The collected data is temporarily stored in a database.

[0524] Step 2:

[0525] The server pre-processes the collected data to remove unnecessary information, such as removing HTML tags, eliminating special characters, and tokenizing the text, converting the data into a format that is easier to analyze.

[0526] Step 3:

[0527] The server then analyzes the preprocessed data using natural language processing (NLP) techniques. Key analysis items include keyword frequency analysis, context analysis, and sentiment analysis. Sentiment analysis calculates a positive, negative, or neutral sentiment score.

[0528] Step 4:

[0529] The server performs a credibility assessment: it compares the preprocessed data with trusted sources (e.g., official announcements, well-known news sites) and calculates a credibility score. This assessment determines whether the information is trustworthy.

[0530] Step 5:

[0531] The server uses an emotion engine to recognize the user's emotions toward the information. This emotion engine analyzes emotions from the user's browsing history and posted content to understand the potential emotional response to the information. This step is particularly effective when the authenticity of the information is unknown.

[0532] Step 6:

[0533] The server takes action against unreliable and negative information, specifically by sending a request to the social media platform to delete the relevant social media post, and also by sending a request to add a warning message to any suspicious information.

[0534] Step 7:

[0535] The terminal executes deletion requests and displays warning messages based on requests from the server. When a post is deleted, the poster is notified of the deletion, and when a post has a warning message attached, a warning message is displayed to warn viewers.

[0536] Step 8:

[0537] Users can check the warning messages and removal notices displayed on their devices, which allows them to reconfirm the reliability of the information provided and make an appropriate decision.

[0538] In this way, we can build a system that can detect fake and misleading information on the Internet in real time and respond quickly. In addition, by using an emotion engine, it is possible to take user emotions into account and provide more appropriate warnings and notifications.

[0539] Example 2

[0540] 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."

[0541] In recent years, a lot of false and misleading information has been spread on the Internet, resulting in an increasing number of cases where users believe the incorrect information. Furthermore, the impact of this information on users' emotions is also a problem that cannot be ignored. Conventional technologies lacked sufficient mechanisms to prevent the spread of false information and lacked measures that took user emotions into consideration.

[0542] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0543] In this invention, the server includes means for collecting information on the Internet, means for preprocessing the collected information, means for analyzing the preprocessed information and performing sentiment analysis and reliability determination, means for taking action on unreliable and negative information, means for recognizing user sentiment and grasping emotional reactions to the information, and means for notifying the user of the results of the action. This makes it possible to prevent the spread of false and misleading information on the Internet and to take warnings and countermeasures that take user sentiment into consideration.

[0544] "Information on the Internet" refers to data such as text, images, audio, and video that is publicly available on the Internet, including websites, social media platforms, blogs, and news sites.

[0545] "Means of collection" refers to methods and tools for obtaining information on the Internet, such as collecting data using APIs or web scraping technology.

[0546] "Preprocessing means" refers to processing methods and tools used to remove unnecessary data from collected information and convert it into a format that is easy to analyze.

[0547] "Means of analysis" refers to techniques for analyzing preprocessed data and extracting useful information. Specifically, this includes natural language processing, sentiment analysis, keyword frequency analysis, etc.

[0548] "Sentiment analysis" is a technique for identifying and scoring positive, negative, and neutral emotions from text data.

[0549] "Credibility assessment" is a method of evaluating how trustworthy information is based on its source and content.

[0550] "Means for taking action" refers to measures to take against unreliable and negative information, including sending a removal request or adding a warning message.

[0551] "Means for recognizing user emotions" refers to technology that analyzes users' browsing history and posted content to understand their emotional reactions to specific information.

[0552] "Means for notifying the user of the outcome of the action" means a method for informing the user about the action taken, including, for example, displaying a removal notice or a warning message.

[0553] "Filtering means" refers to processing methods or tools for removing unnecessary data from collected information.

[0554] "Special character removal" is the process of removing unnecessary symbols and special characters from text data.

[0555] "Text tokenization" is the process of dividing text data into units of words or phrases.

[0556] "Natural language processing technology" refers to technology that enables computers to understand, interpret, and generate human language. Examples include text analysis, context understanding, and machine translation.

[0557] A "trustworthiness score" is a numerical representation of the reliability of information based on an evaluation of its source and content.

[0558] An "emotion engine" is a tool or algorithm that analyzes a user's emotions and identifies their emotional response to specific information.

[0559] A "removal request" is an action requesting that a social media platform remove a specific post.

[0560] A "warning message" is a message that alerts the user to specific information.

[0561] This invention is a system that detects fake and misleading information on the Internet in real time and warns users. By combining this system with an emotion engine that recognizes user emotions, it can comprehensively consider the reliability of information and user emotions.

[0562] First, the server collects information from the Internet. Specifically, it uses the API of a social networking platform (for example, Twitter API or general web scraping methods) to automatically retrieve new posts based on specified keywords. This collected data is temporarily stored in a database (for example, MySQL or MongoDB).

[0563] The server then preprocesses the collected information, which includes removing HTML tags, stripping special characters, and tokenizing the text using Python's BeautifulSoup, regular expressions, and NLTK libraries, converting the data into a format that is easier to parse.

[0564] The server then analyzes the preprocessed data. Specifically, it uses natural language processing techniques (e.g., Spacy or BERT) to perform sentiment analysis and keyword frequency analysis of the text. It also verifies the source of the information and calculates a credibility score by comparing it with highly reliable sources (e.g., official announcements or well-known news sites).

[0565] If the information is judged to be unreliable and negative, the server uses an emotion engine to recognize the user's emotions. This engine analyzes emotions from the user's browsing history and posted content to understand the potential emotional reaction to specific information. This process uses Google Analytics and emotion analysis APIs (e.g., IBM Watson Tone Analyzer).

[0566] The server then takes action against unreliable and negative information, sending a request to the social media platform to remove it and also sending a request to add a warning message to any suspicious information.

[0567] Finally, the device notifies the user of deletion requests or warning messages based on requests from the server. For deleted posts, the device notifies the poster, and for posts with attached warning messages, the device displays a warning message to the viewer.

[0568] Specific examples

[0569] For example, consider a case where a server collects posts related to the "Noto Peninsula Earthquake." The collected data is preprocessed by removing HTML tags and special characters and tokenizing the text. Next, natural language processing techniques (e.g., Spacy or BERT) are used to analyze sentiment and determine reliability. If the information is determined to be unreliable and negative, a deletion request for the post is sent to the social media platform. A request with a warning message attached is also sent, taking into account the user's sentiment. The device displays the deletion or warning notification to the user, allowing the user to review it and take appropriate action.

[0570] Examples of prompt statements

[0571] Collect and preprocess the latest posts related to the "Noto Peninsula Earthquake." Next, use Spacy and BERT to perform sentiment analysis and determine trustworthiness, and create a deletion request and warning message for posts with low trustworthiness and negative views. Finally, notify users.

[0572] This system will prevent the spread of false and misinformation and will be able to issue warnings and take measures that take user sentiment into account, allowing users to obtain accurate and reliable information and improving the information environment on the Internet.

[0573] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0574] Step 1: Data collection

[0575] The server collects information from the Internet. Specifically, it uses the API of social media platforms and web scraping technology to automatically retrieve new posts based on specified keywords. The input is a specific keyword (e.g., "Noto Peninsula Earthquake"). The output is the collected post data (e.g., text, user name, posting date and time, etc.).

[0576] Step 2: Data Preprocessing

[0577] The server preprocesses the collected information. Specifically, it uses BeautifulSoup to remove HTML tags and regular expressions to remove special characters. It also uses NLTK to tokenize the text. The input for this step is the collected post data. The output is the preprocessed, clean text data.

[0578] Step 3: Data analysis

[0579] The server analyzes the preprocessed data. Specifically, it performs sentiment and keyword frequency analysis on the text using Spacy and BERT. It also compares it with established news sites and official announcements to verify the source of the information. The input for this step is the preprocessed, clean text data. The output is a sentiment score, keyword frequency, and confidence score.

[0580] Step 4: Reliability determination

[0581] The server evaluates the reliability of the information based on the data analysis results. Specifically, it checks whether the information matches a reliable source based on the obtained reliability score. The inputs to this step are the sentiment score, keyword frequency, and reliability score. The output is a judgment result of whether the information is highly or low reliable.

[0582] Step 5: Use the Emotion Engine

[0583] The server uses an emotion engine to recognize the user's emotions toward the information. It uses Google Analytics and an emotion analysis API to analyze emotions from the user's browsing history and posted content, and understands the emotional reaction to specific information. The inputs to this step are the credibility judgment results and the user's past browsing history and posted data. The output is the user's emotional reaction pattern.

[0584] Step 6: Take Action

[0585] The server takes action against unreliable and negative information by sending a deletion request to the social media platform and adding a warning message to suspicious information. The inputs to this step are the reliability judgment results and the user's emotional reaction patterns. As outputs, a deletion request and a warning message request are sent to the social media platform.

[0586] Step 7: Notify users

[0587] The device executes the deletion request or displays the warning message based on the request from the server. Specifically, the notification is displayed using JavaScript or HTML. The input to this step is the deletion request and warning message request from the server. The output is the deletion notification or warning notification displayed to the user.

[0588] This processing flow makes it possible to evaluate the reliability of information on the Internet and provide appropriate measures and notifications to users.

[0589] (Application example 2)

[0590] 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."

[0591] In the electronic payment process, users often make transactions based on unreliable information or fake reviews, which can lead to unreliable transactions and fraud. To solve this problem, a system is needed that can evaluate the reliability of information related to transactions in real time and warn users in a timely manner.

[0592] 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.

[0593] In this invention, the server includes means for collecting information from the Internet, means for preprocessing the collected information, means for analyzing the preprocessed information and performing sentiment analysis and reliability determination, means for taking action on information that is low in reliability and negative, means for notifying the user of the results of the action, and means for evaluating the reliability of transaction information in the payment system and displaying a warning message. This makes it possible to evaluate the reliability of transaction information during electronic payment and issue an appropriate warning to the user.

[0594] The "Internet" is a system of interconnected computer networks around the world that enables the exchange of information.

[0595] "Information collection means" is a system that automatically acquires data from the Internet and uses APIs to collect the necessary information.

[0596] "Preprocessing" refers to the process of converting collected information into a format that is easier to analyze, by removing unnecessary data and special characters and tokenizing the text.

[0597] The "analysis method" is a system that uses preprocessed data to evaluate the reliability of information and emotions, and is carried out using natural language processing technology.

[0598] "Sentiment analysis" is a technique for reading emotions from text data and identifying positive, negative, and neutral emotions.

[0599] "Credibility determination" is the process of assessing the trustworthiness of a source and calculating a credibility score, which is then compared to other trusted sources.

[0600] "Negative information" refers to information that may have a negative impact on users and is identified through the results of sentiment analysis.

[0601] The "action execution means" is a mechanism for performing specific processing on information that is determined to be unreliable and negative, such as adding a warning message or sending a deletion request.

[0602] "Notification means" refers to the function of informing the user of the results of an action taken, and is performed through an application or device.

[0603] A "payment system" is a system for electronically conducting monetary transactions and allowing users to purchase goods and services online.

[0604] A "warning message" is a message that informs users that the information is unreliable and helps them avoid inappropriate transactions.

[0605] The system for realizing this invention is composed of three main elements: a server, a terminal, and a user. The specific configuration and operation of the system are explained in detail below.

[0606] The server collects information from the Internet, preprocesses it, and performs sentiment analysis and reliability assessment. The data is collected using APIs from social media and review sites. For example, the Twitter API and Facebook Graph API are used to collect reviews and posts based on specific keywords. The collected data is then temporarily stored in a MySQL database.

[0607] The server then pre-processes the collected information, which includes removing HTML tags and special characters, and tokenizing the text to convert the data into a format that is easier to parse. This process is performed using the Python libraries NLTK and SpaCy.

[0608] The preprocessed data is then analyzed using natural language processing (NLP) techniques. Specifically, the text is subjected to sentiment analysis and credibility assessment. Sentiment analysis involves identifying positive, negative, and neutral sentiment from the text and calculating a sentiment score. TextBlob and IBM Watson Emotion Analysis are used for this task. Credibility assessment involves comparing the data with official sources (e.g., well-known news sites or official announcements) and calculating a credibility score.

[0609] Actions are taken against information that is deemed unreliable and negative. These actions include sending a deletion request to the social media platform or displaying a warning message to the user. After the action is taken, the result is notified to the user's device. This notification allows the user to receive appropriate information and avoid inappropriate transactions.

[0610] As an example of how this system can be used, consider the electronic payment app "SecurePay Warning System." When a user browses a specific product or store through this app, the server collects reviews related to that product or store, performs sentiment analysis, and determines its trustworthiness. If it is determined to be untrustworthy, a warning message is displayed on the user's device. This warning allows the user to make transactions safely.

[0611] As an example, use the following prompt:

[0612] Please analyze this review body: "{review body}" and provide a rating and sentiment score. User ID: {user ID}, Purchase history: {purchase history}"

[0613] The system will protect users from false and misinformation online and allow them to make decisions based on more reliable information.

[0614] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0615] Step 1:

[0616] The server collects information from the Internet. Specifically, it uses the APIs of social media and review sites to collect reviews and posts based on specified keywords. For example, it uses the Twitter API or Facebook Graph API to obtain posts related to a "specific product name." The collected data is then temporarily stored in a MySQL database.

[0617] Input: Keywords, SNS API request

[0618] Output: Collected reviews and submission data (JSON format)

[0619] Step 2:

[0620] The server preprocesses the collected information, removing HTML tags and special characters, and tokenizing the text. It uses the Python libraries NLTK and SpaCy to convert the collected data into a format that is easy to analyze.

[0621] Input: Collected reviews and submission data

[0622] Output: Preprocessed text data (cleaned)

[0623] Step 3:

[0624] The server analyzes the preprocessed data. Based on the preprocessed text data, natural language processing techniques are used to perform sentiment analysis and reliability determination. For sentiment analysis, TextBlob and IBM Watson Emotion Analysis are used to calculate positive, negative, or neutral sentiment scores from the text. For reliability determination, a reliability score is calculated by comparing it with official sources.

[0625] Input: Preprocessed text data

[0626] Output: sentiment score, confidence score

[0627] Step 4:

[0628] The server takes action against unreliable and negative information. Specifically, it sends a deletion request to the social media platform or a request to add a warning message. If the deletion request is accepted, the information is deleted from the platform. In addition, information with a warning message is displayed to warn viewers.

[0629] Input: sentiment score, confidence score

[0630] Output: Delete request, warning message request

[0631] Step 5:

[0632] The terminal will inform the user of the results of the action. Warning messages will be displayed for unreliable information, allowing the user to avoid inappropriate transactions. Warning messages will be delivered to the user through the mobile application, helping them to carry out safe transactions.

[0633] Input: Delete request result, Warning message request result

[0634] Output: User notification (warning message)

[0635] Step 6:

[0636] The user will review the notification and take appropriate action based on its content, for example, choosing to avoid the transaction if a warning message is displayed, or confirming that the deletion request has been accepted and the information has been deleted.

[0637] Input: User notification (warning message)

[0638] Output: User decision

[0639] 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.

[0640] 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.

[0641] 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.

[0642] [Third embodiment]

[0643] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0644] 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.

[0645] 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).

[0646] 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.

[0647] 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.

[0648] 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).

[0649] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0650] 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.

[0651] 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.

[0652] 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.

[0653] 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.

[0654] 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."

[0655] This invention is a system for detecting false and misleading information on the Internet in real time and issuing warnings to users. The system includes a server that collects information on the Internet, preprocesses the collected information, analyzes the preprocessed information, takes action against unreliable and negative information, and notifies users of the results.

[0656] Specifically, the program of this system performs the following processes.

[0657] Data collection

[0658] The server collects public information from social media and the internet based on pre-specified keywords, using the API of the social media platform to automatically retrieve new posts.

[0659] Data Preprocessing

[0660] The server then removes unnecessary information from the collected data and converts it into a format that is easier to analyze, by performing preprocessing such as removing HTML tags, eliminating special characters, and tokenizing the text.

[0661] Data analysis

[0662] The server analyzes the preprocessed data using natural language processing technology, extracting important keywords, analyzing sentiment, and assessing trustworthiness. Sentiment analysis determines and scores positive, negative, or neutral sentiment from the context of the text. Credibility assessment verifies whether the collected information source is trustworthy and assigns a trustworthiness score.

[0663] Reliability determination

[0664] The server determines the reliability of the information based on the results of the data analysis. This includes comparing it with reliable sources, checking against past data, etc. If the information is deemed unreliable and negative, it proceeds to the next step.

[0665] Action Execution

[0666] The server sends a request to the social media platform to remove unreliable and negative information, and also requests that a warning message be added to suspicious information. Based on this request, the social media platform will delete the post or display a warning message.

[0667] User Notification

[0668] The device will notify the user that the deletion request has been carried out and will display a warning message, allowing the user to review the notification and reconfirm the authenticity of the information.

[0669] Specific examples

[0670] For example, suppose a server collects posts about the "Noto Peninsula Earthquake." After removing HTML tags and special characters from the collected data, the text is tokenized. Next, the collected data is analyzed using natural language processing technology to perform sentiment analysis and determine trustworthiness. For posts that are determined to be unreliable and contain negative information based on comparison with past reliable sources, a request is sent to the social media platform to delete them. In addition, for suspicious information, a request is sent to display a warning message. This allows the device to notify the user of the warning and reconfirm the authenticity of the information.

[0671] This system will effectively prevent the spread of false and misinformation and provide users with accurate information.

[0672] The processing flow will be explained below.

[0673] Step 1:

[0674] The server collects public information from social media and the internet. This information collection is done by using the API of the social media platform to obtain new posts containing specified keywords (e.g., "Noto Peninsula Earthquake"). The collected data is temporarily stored in a database.

[0675] Step 2:

[0676] The server pre-processes the collected data to remove unnecessary information, such as removing HTML tags, eliminating special characters, and tokenizing the text, converting the data into a format that is easier to analyze.

[0677] Step 3:

[0678] The server then analyzes the preprocessed data using natural language processing (NLP) techniques. The main analysis items include keyword frequency analysis, sentiment analysis, and information source verification. Sentiment analysis produces a positive, negative, or neutral sentiment score.

[0679] Step 4:

[0680] The server evaluates the reliability of the information based on the data analysis results. It compares the data with reliable sources (e.g., official announcements, well-known news sites) and calculates a reliability score. If the reliability is low, it proceeds to the next step.

[0681] Step 5:

[0682] The server takes action against unreliable and negative information, specifically by sending a deletion request to the social media platform for the relevant social media post, and by sending a request to display a warning message for sensitive information.

[0683] Step 6:

[0684] The terminal executes deletion requests and displays warning messages based on requests from the server. When a post is deleted, the poster is notified of the deletion, and when a post has a warning message attached, a warning message is displayed to warn viewers.

[0685] Step 7:

[0686] Users can check the warning messages and deletion notices displayed on their devices, reconfirm the reliability of the information provided, and make appropriate decisions.

[0687] In this way, a system will be built that can detect fake and misleading information on the Internet in real time and respond quickly.

[0688] Example 1

[0689] 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."

[0690] In today's internet environment, the rapid spread of false and misleading information has become a social problem. Accurate information and false information are often mixed together on social media platforms, making it difficult for users to determine the reliability of the information. Therefore, there is a need for methods to quickly detect and respond to unreliable information.

[0691] 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.

[0692] In this invention, the server includes a means for collecting information from the Internet, a means for preprocessing the collected information, a means for analyzing the preprocessed information using natural language processing technology to perform sentiment analysis and reliability determination, a means for deleting unreliable and negative information or adding a warning message to the information, and a means for notifying the user of the results of the action, thereby enabling the rapid detection and response to false information and misinformation.

[0693] "Means for collecting information on the Internet" refers to a mechanism for obtaining data based on specified keywords from websites, social networking services (SNS), and other public information on the Internet.

[0694] The "means for preprocessing collected information" refers to a mechanism for performing a process to remove unnecessary information from the collected data and convert it into an analyzable format.

[0695] "Means of analysis using natural language processing technology" refers to a system that utilizes algorithms and models to automatically perform sentiment analysis and reliability assessment on collected text data.

[0696] "Sentiment analysis" is a technique that determines positive, negative, or neutral sentiment from the context of text and assigns a score to each post.

[0697] "Credibility assessment" is the process of assessing the reliability of collected information sources and calculating a reliability score by matching and comparing them with past data.

[0698] "Means for taking action to delete or add a warning message" refers to a mechanism for sending a request to a social media platform to delete or display a warning message for unreliable and negative information.

[0699] "Means for notifying the user of the results of an action" refers to a mechanism that notifies the user of the results of an action taken on their device, allowing the user to verify the reliability and authenticity of the information.

[0700] This invention is a system that detects false information and misinformation on the Internet in real time and warns users. This system includes a series of processes that collect, process, and notify data between servers, terminals, and users.

[0701] The server first uses the API of a social media platform as a means of collecting information on the Internet. For example, it can use the Twitter API to collect tweets related to the "Great Noto Peninsula Earthquake." This allows the server to obtain new posts in real time. As a specific example, when a server collects posts related to the "Great Noto Peninsula Earthquake," it uses the Twitter API to collect tweets containing the relevant keyword.

[0702] Next, the server preprocesses the collected information. To remove unnecessary information from the collected data and convert it into a format that is easier to analyze, it uses the BeautifulSoup library to remove HTML tags and special characters, and then uses the NLTK library to tokenize the text. For example, it preprocesses the tweet "Information about the Noto Peninsula earthquake can be found here → [link]" by removing the link and separating it into words.

[0703] The preprocessed data is then analyzed using natural language processing techniques. The server performs sentiment analysis using Hugging Face Transformers to determine positive, negative, or neutral sentiment. It also evaluates the reliability of the collected information sources and assigns a credibility score. For example, the server might perform sentiment analysis on a tweet such as "This earthquake is having a major impact on people" and assign it a negative score.

[0704] The server then combines these analysis results to determine the reliability of the information. For unreliable and negative information, it sends a deletion request to the social media platform, and for suspicious information, it adds a warning message. Users can request the deletion of specific tweets or the display of warning messages via the Twitter API. For example, they can send a request with a warning that "This is a hoax about the Noto Peninsula earthquake."

[0705] Finally, the server notifies the user of the results of these actions. The device displays a notification on the user's smartphone or other device, allowing the user to check the content of the notification. For example, the device may send the user a notification message such as, "Unreliable information has been detected and deleted. Please check for details."

[0706] This system enables the rapid detection and response of false information and misinformation, and provides accurate information to users. For example, keywords can be specified by inputting the following prompt sentences into the generative AI model:

[0707] Prompt Sentence Examples

[0708] Collect the latest information on the "Noto Peninsula Earthquake" and be alerted to unreliable and negative information.

[0709] Based on this prompt, the system will detect fake and misinformation related to the specified keywords in real time and notify the user.

[0710] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0711] Step 1:

[0712] Data Collection Settings

[0713] The server loads a pre-specified list of keywords to be used for data collection. For example, the keyword "Noto Peninsula Earthquake" is added to the list. This list is used for subsequent API queries.

[0714] Input: A specified list of keywords

[0715] Output: Keyword list loaded

[0716] Specific behavior:

[0717] The server retrieves "Noto Peninsula Earthquake" from the keyword list and sets up data collection.

[0718] Step 2:

[0719] Using SNS API

[0720] The server uses the API of the social media platform to retrieve posts based on the target keywords. For example, the Twitter API is used to collect tweets related to the "Noto Peninsula Earthquake." The server uses the Streaming API to retrieve new posts in real time.

[0721] Input: Keywords to be collected

[0722] Output: Collected SNS post data

[0723] Specific behavior:

[0724] The server uses Twitter's Streaming API to retrieve tweets related to the keyword "Noto Peninsula Earthquake" in real time.

[0725] Step 3:

[0726] Data Preprocessing

[0727] The server cleans the collected data and converts it into a format that is easy to analyze, using BeautifulSoup to remove HTML tags and special characters, and the NLTK library to tokenize the text.

[0728] Input: Collected social media posting data

[0729] Output: Preprocessed text data

[0730] Specific behavior:

[0731] The server removes the link from the tweet "Information about the Noto Peninsula earthquake can be found here → [link]" and separates the text.

[0732] Step 4:

[0733] Conducting sentiment analysis

[0734] The server analyzes the preprocessed text data using natural language processing techniques. Specifically, it uses Hugging Face Transformers to perform sentiment analysis on the text data and determine positive, negative, or neutral sentiment. It assigns a sentiment score to each piece of text.

[0735] Input: Preprocessed text data

[0736] Output: Text data with sentiment scores

[0737] Specific behavior:

[0738] The server performs sentiment analysis on tweets such as "This earthquake is having a big impact on people" and assigns them a negative score.

[0739] Step 5:

[0740] Conducting reliability evaluation

[0741] The server compares the collected information with a historical database and calculates a reliability score to assess the reliability of the information source.

[0742] Input: Text data with sentiment scores

[0743] Output: Text data with confidence scores

[0744] Specific behavior:

[0745] The server references the poster's past reliability data and calculates a reliability score.

[0746] Step 6:

[0747] Integration of reliability judgments

[0748] The server combines the sentiment score and the reliability score to determine the overall reliability of the information. For information that is low in reliability and negative, it proceeds to the next step.

[0749] Input: Text data with confidence scores

[0750] Output: Judgment result

[0751] Specific behavior:

[0752] The server selects posts with a "negative" sentiment score and a low credibility score.

[0753] Step 7:

[0754] Execute Action

[0755] The server sends a request to the social media platform to remove unreliable and negative information or to add a warning message to the information, for example, through the Twitter API.

[0756] Input: Judgment result

[0757] Output: Request sent to the social media platform

[0758] Specific behavior:

[0759] The server sends a request via the Twitter API that includes a warning: "This is false information about the Noto Peninsula earthquake."

[0760] Step 8:

[0761] User Notification

[0762] The device notifies the user of the result of the action received from the server, for example by displaying a notification on the user's smartphone so that the user can check the content.

[0763] Input: Action result

[0764] Output: Notification message sent to the user

[0765] Specific behavior:

[0766] The device will display a notification on the user's smartphone saying, "Unreliable information has been detected and removed. Please check for details."

[0767] (Application example 1)

[0768] 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."

[0769] The speed at which information spreads on the Internet has created an environment in which false and misinformation can easily spread. It is extremely important to detect such false and misinformation in real time and to warn users appropriately. However, conventional technology has made it difficult to do this efficiently and quickly. There is a need to provide a system that can solve these issues, more reliably evaluate the reliability of information, and provide appropriate notifications.

[0770] 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.

[0771] In this invention, the server includes means for collecting information on the Internet, means for preprocessing the collected information, means for analyzing the preprocessed information and performing sentiment analysis and reliability determination, means for requesting deletion and adding a warning message to unreliable and negative information, and means for notifying the user of the results of the action, thereby enabling the effective detection of false information and misinformation in real time and notifying the user.

[0772] "Means for collecting information on the Internet" refers to methods and devices for obtaining the latest posted information from social networking sites, news sites, etc.

[0773] "Means for preprocessing collected information" refers to a processing method and device for removing unnecessary data from the acquired information and converting it into a format that is easy to analyze.

[0774] "Means for analyzing pre-processed information and performing sentiment analysis and credibility determination" refers to methods and apparatus for analyzing tokenized text data using natural language processing techniques to determine sentiment from the context of the text and assess the credibility of the source.

[0775] "Means for executing requests to delete and add warning messages to unreliable and negative information" refers to a method and device for sending a request to delete information determined to be unreliable or a request to add a warning message to a social media platform.

[0776] "Means for notifying users of the results of actions" refers to methods and devices for notifying users in an appropriate format of the results of actions taken to remove or warn against false or misleading information.

[0777] This invention is a system for detecting false and misleading information on the Internet in real time and warning users, and is composed of the following components:

[0778] Data collection

[0779] The server collects the latest posting information from social media platforms, news sites, etc. To collect the data, it uses the API of the social media platform to automatically obtain new posting information.

[0780] Data Preprocessing

[0781] The server filters unnecessary data (HTML tags and special characters) from the retrieved information and tokenizes the text, using the Python libraries BeautifulSoup and NLTK (Natural Language Toolkit) for this process.

[0782] Data analysis

[0783] The server analyzes the tokenized text data using natural language processing technology. Sentiment analysis and credibility assessment are performed here. Sentiment analysis determines and scores positive, negative, or neutral sentiment from the context of the text. Credibility assessment verifies whether the collected information source is trustworthy and assigns a credibility score. This analysis uses Python NLP libraries (spaCy, NLTK).

[0784] Reliability determination

[0785] The server uses natural language processing technology to analyze the data and then determines the reliability of the information. This includes comparing it with trusted sources and checking against past data. If the information is deemed unreliable and negative, it proceeds to the next step.

[0786] Action Execution

[0787] A request to delete or add a warning message to unreliable and negative information is sent to the SNS platform. The server uses the SNS platform API to send the request to delete or add a warning message.

[0788] User Notification

[0789] The device will notify the user that the deletion request has been carried out and will display a warning message, allowing the user to reconfirm the authenticity of the information.

[0790] Specific examples

[0791] For example, if a user sets the keyword "novel coronavirus vaccine hoax," the server will collect posts related to "coronavirus vaccine." The collected data will be tokenized after removing HTML tags and special characters. The collected data will then be analyzed using natural language processing technology to perform sentiment analysis and determine trustworthiness. For posts that are deemed untrustworthy and negative, a request will be sent to the social media platform to delete them, and a request will be sent to display a warning message for suspicious information. The device will then send a warning notification to the user, allowing them to reconfirm the trustworthiness of the information.

[0792] Prompt Sentence Examples

[0793] Collect the latest posts about "COVID-19 vaccines" from social media and news sites. Remove HTML tags and special characters from the collected information and tokenize the text. Next, use natural language processing technology to perform sentiment analysis and credibility assessment to identify unreliable and negative information. Then, send a request to the social media platform to remove the information and add a warning message, and finally send a push notification to the user.

[0794] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0795] Step 1:

[0796] The server uses the APIs of social media and news sites to collect the latest post information based on specified keywords. The data retrieved by the server includes metadata such as the post text, poster information, and posting date and time. The input for this process is keywords related to "novel coronavirus vaccine hoaxes," and the output is the raw post data.

[0797] Step 2:

[0798] The server filters unnecessary information from the collected data. Specifically, it removes HTML tags and special characters to generate clean text data. This process uses the Python library BeautifulSoup. The input is the raw post data collected in step 1, and the output is the filtered text data.

[0799] Step 3:

[0800] The server tokenizes the filtered data, splitting the text into words and converting it into a format that is easy to use with natural language processing. This process uses NLTK (Natural Language Toolkit). The input is the clean text data generated in step 2, and the output is the tokenized text data.

[0801] Step 4:

[0802] The server analyzes the tokenized data using natural language processing technology. This involves sentiment analysis and credibility assessment. Sentiment analysis involves scoring positive, negative, or neutral sentiment from the context of the text. Credibility assessment verifies whether the collected information source is trustworthy and assigns a credibility score. This process uses Python NLP libraries (spaCy, NLTK). The input is the tokenized text data generated in step 3, and the output is a sentiment score and a credibility score.

[0803] Step 5:

[0804] The server determines the reliability of the information based on the analysis results. If the reliability is low and the information is determined to be negative, it proceeds to the next step. The input is the emotion score and reliability score generated in step 4, and the output is the information whose reliability and emotion have been determined.

[0805] Step 6:

[0806] The server sends a request to the SNS platform to delete and add a warning message to unreliable and negative information. This process uses the SNS platform's API. The input is the information determined in step 5, and the output is the result of sending the request to the SNS platform.

[0807] Step 7:

[0808] The device notifies the user of the results of the deletion request or warning message addition request received from the server. The notification is performed using a push notification. The input is the request result returned from the SNS platform in step 6, and the output is a push notification sent to the user's device.

[0809] 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.

[0810] This invention is a system that detects fake and misleading information on the Internet in real time and warns users. By combining this system with an emotion engine that recognizes user emotions, it can comprehensively consider the reliability of information and user emotions.

[0811] The system of the present invention comprises the following means:

[0812] Data collection

[0813] The server collects public information from social media and the internet based on pre-specified keywords. Data collection is done by automatically retrieving new posts using the API of the social media platform. The retrieved data is temporarily stored in a database.

[0814] Data Preprocessing

[0815] The server pre-processes the collected data to remove unnecessary information, including removing HTML tags, removing special characters, and tokenizing the text, converting the data into a format that is easier to analyze.

[0816] Data analysis

[0817] The server then analyzes the preprocessed data using natural language processing (NLP) techniques. Analysis items include keyword frequency analysis, sentiment analysis, and information source verification. Sentiment analysis calculates a positive, negative, or neutral sentiment score based on the context of the text.

[0818] Reliability determination

[0819] The server evaluates the reliability of the information based on the data analysis results. It compares the data with reliable sources (e.g., official announcements, well-known news sites) and calculates a reliability score. Information with low reliability and negative results proceeds to the next step.

[0820] Using the Emotion Engine

[0821] The server recognizes the user's feelings toward the information using an emotion engine, which analyzes emotions from the user's browsing history and posted content to understand the potential emotional response to specific information.

[0822] Action Execution

[0823] The server sends a request to the social media platform to remove unreliable and negative information. It also sends a request to add a warning message to suspicious information, taking into account user sentiment. Based on this request, the social media platform deletes the post or displays a warning message.

[0824] User Notification

[0825] The terminal executes deletion requests and displays warning messages based on requests from the server. When a post is deleted, the poster is notified of the deletion, and when a post has a warning message attached, a warning message is displayed to warn viewers.

[0826] Specific examples

[0827] For example, suppose a server collects posts related to the "Noto Peninsula Earthquake." The collected data is stripped of HTML tags and special characters and the text is tokenized. Next, natural language processing technology is used to analyze the collected data, performing sentiment analysis and determining reliability. If the information is determined to be unreliable and negative, a request to delete the post is sent to the social media platform. A request is also sent to add a warning message, taking into account user sentiment, if necessary. Finally, the device displays a deletion or warning notice to the user, allowing them to confirm and take appropriate action.

[0828] This system will prevent the spread of false and misinformation and issue warnings that take user sentiment into account, allowing users to obtain accurate and reliable information and improving the information environment on the Internet.

[0829] The processing flow will be explained below.

[0830] Step 1:

[0831] The server collects public information from social media and the internet. This information collection is done by using the API of the social media platform to obtain new posts containing specified keywords (e.g., "Noto Peninsula Earthquake"). The collected data is temporarily stored in a database.

[0832] Step 2:

[0833] The server pre-processes the collected data to remove unnecessary information, such as removing HTML tags, eliminating special characters, and tokenizing the text, converting the data into a format that is easier to analyze.

[0834] Step 3:

[0835] The server then analyzes the preprocessed data using natural language processing (NLP) techniques. Key analysis items include keyword frequency analysis, context analysis, and sentiment analysis. Sentiment analysis calculates a positive, negative, or neutral sentiment score.

[0836] Step 4:

[0837] The server performs a credibility assessment: it compares the preprocessed data with trusted sources (e.g., official announcements, well-known news sites) and calculates a credibility score. This assessment determines whether the information is trustworthy.

[0838] Step 5:

[0839] The server uses an emotion engine to recognize the user's emotions toward the information. This emotion engine analyzes emotions from the user's browsing history and posted content to understand the potential emotional response to the information. This step is particularly effective when the authenticity of the information is unknown.

[0840] Step 6:

[0841] The server takes action against unreliable and negative information, specifically by sending a request to the social media platform to delete the relevant social media post, and also by sending a request to add a warning message to any suspicious information.

[0842] Step 7:

[0843] The terminal executes deletion requests and displays warning messages based on requests from the server. When a post is deleted, the poster is notified of the deletion, and when a post has a warning message attached, a warning message is displayed to warn viewers.

[0844] Step 8:

[0845] Users can check the warning messages and removal notices displayed on their devices, which allows them to reconfirm the reliability of the information provided and make an appropriate decision.

[0846] In this way, we can build a system that can detect fake and misleading information on the Internet in real time and respond quickly. In addition, by using an emotion engine, it is possible to take user emotions into account and provide more appropriate warnings and notifications.

[0847] Example 2

[0848] 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."

[0849] In recent years, a lot of false and misleading information has been spread on the Internet, resulting in an increasing number of cases where users believe the incorrect information. Furthermore, the impact of this information on users' emotions is also a problem that cannot be ignored. Conventional technologies lacked sufficient mechanisms to prevent the spread of false information and lacked measures that took user emotions into consideration.

[0850] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0851] In this invention, the server includes means for collecting information on the Internet, means for preprocessing the collected information, means for analyzing the preprocessed information and performing sentiment analysis and reliability determination, means for taking action on unreliable and negative information, means for recognizing user sentiment and grasping emotional reactions to the information, and means for notifying the user of the results of the action. This makes it possible to prevent the spread of false and misleading information on the Internet and to take warnings and countermeasures that take user sentiment into consideration.

[0852] "Information on the Internet" refers to data such as text, images, audio, and video that is publicly available on the Internet, including websites, social media platforms, blogs, and news sites.

[0853] "Means of collection" refers to methods and tools for obtaining information on the Internet, such as collecting data using APIs or web scraping technology.

[0854] "Preprocessing means" refers to processing methods and tools used to remove unnecessary data from collected information and convert it into a format that is easy to analyze.

[0855] "Means of analysis" refers to techniques for analyzing preprocessed data and extracting useful information. Specifically, this includes natural language processing, sentiment analysis, keyword frequency analysis, etc.

[0856] "Sentiment analysis" is a technique for identifying and scoring positive, negative, and neutral emotions from text data.

[0857] "Credibility assessment" is a method of evaluating how trustworthy information is based on its source and content.

[0858] "Means for taking action" refers to measures to take against unreliable and negative information, including sending a removal request or adding a warning message.

[0859] "Means for recognizing user emotions" refers to technology that analyzes users' browsing history and posted content to understand their emotional reactions to specific information.

[0860] "Means for notifying the user of the outcome of the action" means a method for informing the user about the action taken, including, for example, displaying a removal notice or a warning message.

[0861] "Filtering means" refers to processing methods or tools for removing unnecessary data from collected information.

[0862] "Special character removal" is the process of removing unnecessary symbols and special characters from text data.

[0863] "Text tokenization" is the process of dividing text data into units of words or phrases.

[0864] "Natural language processing technology" refers to technology that enables computers to understand, interpret, and generate human language. Examples include text analysis, context understanding, and machine translation.

[0865] A "trustworthiness score" is a numerical representation of the reliability of information based on an evaluation of its source and content.

[0866] An "emotion engine" is a tool or algorithm that analyzes a user's emotions and identifies their emotional response to specific information.

[0867] A "removal request" is an action requesting that a social media platform remove a specific post.

[0868] A "warning message" is a message that alerts the user to specific information.

[0869] This invention is a system that detects fake and misleading information on the Internet in real time and warns users. By combining this system with an emotion engine that recognizes user emotions, it can comprehensively consider the reliability of information and user emotions.

[0870] First, the server collects information from the Internet. Specifically, it uses the API of a social networking platform (for example, Twitter API or general web scraping methods) to automatically retrieve new posts based on specified keywords. This collected data is temporarily stored in a database (for example, MySQL or MongoDB).

[0871] The server then preprocesses the collected information, which includes removing HTML tags, stripping special characters, and tokenizing the text using Python's BeautifulSoup, regular expressions, and NLTK libraries, converting the data into a format that is easier to parse.

[0872] The server then analyzes the preprocessed data. Specifically, it uses natural language processing techniques (e.g., Spacy or BERT) to perform sentiment analysis and keyword frequency analysis of the text. It also verifies the source of the information and calculates a credibility score by comparing it with highly reliable sources (e.g., official announcements or well-known news sites).

[0873] If the information is judged to be unreliable and negative, the server uses an emotion engine to recognize the user's emotions. This engine analyzes emotions from the user's browsing history and posted content to understand the potential emotional reaction to specific information. This process uses Google Analytics and emotion analysis APIs (e.g., IBM Watson Tone Analyzer).

[0874] The server then takes action against unreliable and negative information, sending a request to the social media platform to remove it and also sending a request to add a warning message to any suspicious information.

[0875] Finally, the device notifies the user of deletion requests or warning messages based on requests from the server. For deleted posts, the device notifies the poster, and for posts with attached warning messages, the device displays a warning message to the viewer.

[0876] Specific examples

[0877] For example, consider a case where a server collects posts related to the "Noto Peninsula Earthquake." The collected data is preprocessed by removing HTML tags and special characters and tokenizing the text. Next, natural language processing techniques (e.g., Spacy or BERT) are used to analyze sentiment and determine reliability. If the information is determined to be unreliable and negative, a deletion request for the post is sent to the social media platform. A request with a warning message attached is also sent, taking into account the user's sentiment. The device displays the deletion or warning notification to the user, allowing the user to review it and take appropriate action.

[0878] Examples of prompt statements

[0879] Collect and preprocess the latest posts related to the "Noto Peninsula Earthquake." Next, use Spacy and BERT to perform sentiment analysis and determine trustworthiness, and create a deletion request and warning message for posts with low trustworthiness and negative views. Finally, notify users.

[0880] This system will prevent the spread of false and misinformation and will be able to issue warnings and take measures that take user sentiment into account, allowing users to obtain accurate and reliable information and improving the information environment on the Internet.

[0881] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0882] Step 1: Data collection

[0883] The server collects information from the Internet. Specifically, it uses the API of social media platforms and web scraping technology to automatically retrieve new posts based on specified keywords. The input is a specific keyword (e.g., "Noto Peninsula Earthquake"). The output is the collected post data (e.g., text, user name, posting date and time, etc.).

[0884] Step 2: Data Preprocessing

[0885] The server preprocesses the collected information. Specifically, it uses BeautifulSoup to remove HTML tags and regular expressions to remove special characters. It also uses NLTK to tokenize the text. The input for this step is the collected post data. The output is the preprocessed, clean text data.

[0886] Step 3: Data analysis

[0887] The server analyzes the preprocessed data. Specifically, it performs sentiment and keyword frequency analysis on the text using Spacy and BERT. It also compares it with established news sites and official announcements to verify the source of the information. The input for this step is the preprocessed, clean text data. The output is a sentiment score, keyword frequency, and confidence score.

[0888] Step 4: Reliability determination

[0889] The server evaluates the reliability of the information based on the data analysis results. Specifically, it checks whether the information matches a reliable source based on the obtained reliability score. The inputs to this step are the sentiment score, keyword frequency, and reliability score. The output is a judgment result of whether the information is highly or low reliable.

[0890] Step 5: Use the Emotion Engine

[0891] The server uses an emotion engine to recognize the user's emotions toward the information. It uses Google Analytics and an emotion analysis API to analyze emotions from the user's browsing history and posted content, and understands the emotional reaction to specific information. The inputs to this step are the credibility judgment results and the user's past browsing history and posted data. The output is the user's emotional reaction pattern.

[0892] Step 6: Take Action

[0893] The server takes action against unreliable and negative information by sending a deletion request to the social media platform and adding a warning message to suspicious information. The inputs to this step are the reliability judgment results and the user's emotional reaction patterns. As outputs, a deletion request and a warning message request are sent to the social media platform.

[0894] Step 7: Notify users

[0895] The device executes the deletion request or displays the warning message based on the request from the server. Specifically, the notification is displayed using JavaScript or HTML. The input to this step is the deletion request and warning message request from the server. The output is the deletion notification or warning notification displayed to the user.

[0896] This processing flow makes it possible to evaluate the reliability of information on the Internet and provide appropriate measures and notifications to users.

[0897] (Application example 2)

[0898] 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."

[0899] In the electronic payment process, users often make transactions based on unreliable information or fake reviews, which can lead to unreliable transactions and fraud. To solve this problem, a system is needed that can evaluate the reliability of information related to transactions in real time and warn users in a timely manner.

[0900] 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.

[0901] In this invention, the server includes means for collecting information from the Internet, means for preprocessing the collected information, means for analyzing the preprocessed information and performing sentiment analysis and reliability determination, means for taking action on information that is low in reliability and negative, means for notifying the user of the results of the action, and means for evaluating the reliability of transaction information in the payment system and displaying a warning message. This makes it possible to evaluate the reliability of transaction information during electronic payment and issue an appropriate warning to the user.

[0902] The "Internet" is a system of interconnected computer networks around the world that enables the exchange of information.

[0903] "Information collection means" is a system that automatically acquires data from the Internet and uses APIs to collect the necessary information.

[0904] "Preprocessing" refers to the process of converting collected information into a format that is easier to analyze, by removing unnecessary data and special characters and tokenizing the text.

[0905] The "analysis method" is a system that uses preprocessed data to evaluate the reliability of information and emotions, and is carried out using natural language processing technology.

[0906] "Sentiment analysis" is a technique for reading emotions from text data and identifying positive, negative, and neutral emotions.

[0907] "Credibility determination" is the process of assessing the trustworthiness of a source and calculating a credibility score, which is then compared to other trusted sources.

[0908] "Negative information" refers to information that may have a negative impact on users and is identified through the results of sentiment analysis.

[0909] The "action execution means" is a mechanism for performing specific processing on information that is determined to be unreliable and negative, such as adding a warning message or sending a deletion request.

[0910] "Notification means" refers to the function of informing the user of the results of an action taken, and is performed through an application or device.

[0911] A "payment system" is a system for electronically conducting monetary transactions and allowing users to purchase goods and services online.

[0912] A "warning message" is a message that informs users that the information is unreliable and helps them avoid inappropriate transactions.

[0913] The system for realizing this invention is composed of three main elements: a server, a terminal, and a user. The specific configuration and operation of the system are explained in detail below.

[0914] The server collects information from the Internet, preprocesses it, and performs sentiment analysis and reliability assessment. The data is collected using APIs from social media and review sites. For example, the Twitter API and Facebook Graph API are used to collect reviews and posts based on specific keywords. The collected data is then temporarily stored in a MySQL database.

[0915] The server then pre-processes the collected information, which includes removing HTML tags and special characters, and tokenizing the text to convert the data into a format that is easier to parse. This process is performed using the Python libraries NLTK and SpaCy.

[0916] The preprocessed data is then analyzed using natural language processing (NLP) techniques. Specifically, the text is subjected to sentiment analysis and credibility assessment. Sentiment analysis involves identifying positive, negative, and neutral sentiment from the text and calculating a sentiment score. TextBlob and IBM Watson Emotion Analysis are used for this task. Credibility assessment involves comparing the data with official sources (e.g., well-known news sites or official announcements) and calculating a credibility score.

[0917] Actions are taken against information that is deemed unreliable and negative. These actions include sending a deletion request to the social media platform or displaying a warning message to the user. After the action is taken, the result is notified to the user's device. This notification allows the user to receive appropriate information and avoid inappropriate transactions.

[0918] As an example of how this system can be used, consider the electronic payment app "SecurePay Warning System." When a user browses a specific product or store through this app, the server collects reviews related to that product or store, performs sentiment analysis, and determines its trustworthiness. If it is determined to be untrustworthy, a warning message is displayed on the user's device. This warning allows the user to make transactions safely.

[0919] As an example, use the following prompt:

[0920] Please analyze this review body: "{review body}" and provide a rating and sentiment score. User ID: {user ID}, Purchase history: {purchase history}"

[0921] The system will protect users from false and misinformation online and allow them to make decisions based on more reliable information.

[0922] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0923] Step 1:

[0924] The server collects information from the Internet. Specifically, it uses the APIs of social media and review sites to collect reviews and posts based on specified keywords. For example, it uses the Twitter API or Facebook Graph API to obtain posts related to a "specific product name." The collected data is then temporarily stored in a MySQL database.

[0925] Input: Keywords, SNS API request

[0926] Output: Collected reviews and submission data (JSON format)

[0927] Step 2:

[0928] The server preprocesses the collected information, removing HTML tags and special characters, and tokenizing the text. It uses the Python libraries NLTK and SpaCy to convert the collected data into a format that is easy to analyze.

[0929] Input: Collected reviews and submission data

[0930] Output: Preprocessed text data (cleaned)

[0931] Step 3:

[0932] The server analyzes the preprocessed data. Based on the preprocessed text data, natural language processing techniques are used to perform sentiment analysis and reliability determination. For sentiment analysis, TextBlob and IBM Watson Emotion Analysis are used to calculate positive, negative, or neutral sentiment scores from the text. For reliability determination, a reliability score is calculated by comparing it with official sources.

[0933] Input: Preprocessed text data

[0934] Output: sentiment score, confidence score

[0935] Step 4:

[0936] The server takes action against unreliable and negative information. Specifically, it sends a deletion request to the social media platform or a request to add a warning message. If the deletion request is accepted, the information is deleted from the platform. In addition, information with a warning message is displayed to warn viewers.

[0937] Input: sentiment score, confidence score

[0938] Output: Delete request, warning message request

[0939] Step 5:

[0940] The terminal will inform the user of the results of the action. Warning messages will be displayed for unreliable information, allowing the user to avoid inappropriate transactions. Warning messages will be delivered to the user through the mobile application, helping them to carry out safe transactions.

[0941] Input: Delete request result, Warning message request result

[0942] Output: User notification (warning message)

[0943] Step 6:

[0944] The user will review the notification and take appropriate action based on its content, for example, choosing to avoid the transaction if a warning message is displayed, or confirming that the deletion request has been accepted and the information has been deleted.

[0945] Input: User notification (warning message)

[0946] Output: User decision

[0947] 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.

[0948] 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.

[0949] 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.

[0950] [Fourth embodiment]

[0951] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0952] 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.

[0953] 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).

[0954] 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.

[0955] 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.

[0956] 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).

[0957] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0958] 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.

[0959] 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.

[0960] 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.

[0961] 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.

[0962] 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.

[0963] 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."

[0964] This invention is a system for detecting false and misleading information on the Internet in real time and issuing warnings to users. The system includes a server that collects information on the Internet, preprocesses the collected information, analyzes the preprocessed information, takes action against unreliable and negative information, and notifies users of the results.

[0965] Specifically, the program of this system performs the following processes.

[0966] Data collection

[0967] The server collects public information from social media and the internet based on pre-specified keywords, using the API of the social media platform to automatically retrieve new posts.

[0968] Data Preprocessing

[0969] The server then removes unnecessary information from the collected data and converts it into a format that is easier to analyze, by performing preprocessing such as removing HTML tags, eliminating special characters, and tokenizing the text.

[0970] Data analysis

[0971] The server analyzes the preprocessed data using natural language processing technology, extracting important keywords, analyzing sentiment, and assessing trustworthiness. Sentiment analysis determines and scores positive, negative, or neutral sentiment from the context of the text. Credibility assessment verifies whether the collected information source is trustworthy and assigns a trustworthiness score.

[0972] Reliability determination

[0973] The server determines the reliability of the information based on the results of the data analysis. This includes comparing it with reliable sources, checking against past data, etc. If the information is deemed unreliable and negative, it proceeds to the next step.

[0974] Action Execution

[0975] The server sends a request to the social media platform to remove unreliable and negative information, and also requests that a warning message be added to suspicious information. Based on this request, the social media platform will delete the post or display a warning message.

[0976] User Notification

[0977] The device will notify the user that the deletion request has been carried out and will display a warning message, allowing the user to review the notification and reconfirm the authenticity of the information.

[0978] Specific examples

[0979] For example, suppose a server collects posts about the "Noto Peninsula Earthquake." After removing HTML tags and special characters from the collected data, the text is tokenized. Next, the collected data is analyzed using natural language processing technology to perform sentiment analysis and determine trustworthiness. For posts that are determined to be unreliable and contain negative information based on comparison with past reliable sources, a request is sent to the social media platform to delete them. In addition, for suspicious information, a request is sent to display a warning message. This allows the device to notify the user of the warning and reconfirm the authenticity of the information.

[0980] This system will effectively prevent the spread of false and misinformation and provide users with accurate information.

[0981] The processing flow will be explained below.

[0982] Step 1:

[0983] The server collects public information from social media and the internet. This information collection is done by using the API of the social media platform to obtain new posts containing specified keywords (e.g., "Noto Peninsula Earthquake"). The collected data is temporarily stored in a database.

[0984] Step 2:

[0985] The server pre-processes the collected data to remove unnecessary information, such as removing HTML tags, eliminating special characters, and tokenizing the text, converting the data into a format that is easier to analyze.

[0986] Step 3:

[0987] The server then analyzes the preprocessed data using natural language processing (NLP) techniques. The main analysis items include keyword frequency analysis, sentiment analysis, and information source verification. Sentiment analysis produces a positive, negative, or neutral sentiment score.

[0988] Step 4:

[0989] The server evaluates the reliability of the information based on the data analysis results. It compares the data with reliable sources (e.g., official announcements, well-known news sites) and calculates a reliability score. If the reliability is low, it proceeds to the next step.

[0990] Step 5:

[0991] The server takes action against unreliable and negative information, specifically by sending a deletion request to the social media platform for the relevant social media post, and by sending a request to display a warning message for sensitive information.

[0992] Step 6:

[0993] The terminal executes deletion requests and displays warning messages based on requests from the server. When a post is deleted, the poster is notified of the deletion, and when a post has a warning message attached, a warning message is displayed to warn viewers.

[0994] Step 7:

[0995] Users can check the warning messages and deletion notices displayed on their devices, reconfirm the reliability of the information provided, and make appropriate decisions.

[0996] In this way, a system will be built that can detect fake and misleading information on the Internet in real time and respond quickly.

[0997] Example 1

[0998] 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."

[0999] In today's internet environment, the rapid spread of false and misleading information has become a social problem. Accurate information and false information are often mixed together on social media platforms, making it difficult for users to determine the reliability of the information. Therefore, there is a need for methods to quickly detect and respond to unreliable information.

[1000] 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.

[1001] In this invention, the server includes a means for collecting information from the Internet, a means for preprocessing the collected information, a means for analyzing the preprocessed information using natural language processing technology to perform sentiment analysis and reliability determination, a means for deleting unreliable and negative information or adding a warning message to the information, and a means for notifying the user of the results of the action, thereby enabling the rapid detection and response to false information and misinformation.

[1002] "Means for collecting information on the Internet" refers to a mechanism for obtaining data based on specified keywords from websites, social networking services (SNS), and other public information on the Internet.

[1003] The "means for preprocessing collected information" refers to a mechanism for performing a process to remove unnecessary information from the collected data and convert it into an analyzable format.

[1004] "Means of analysis using natural language processing technology" refers to a system that utilizes algorithms and models to automatically perform sentiment analysis and reliability assessment on collected text data.

[1005] "Sentiment analysis" is a technique that determines positive, negative, or neutral sentiment from the context of text and assigns a score to each post.

[1006] "Credibility assessment" is the process of assessing the reliability of collected information sources and calculating a reliability score by matching and comparing them with past data.

[1007] "Means for taking action to delete or add a warning message" refers to a mechanism for sending a request to a social media platform to delete or display a warning message for unreliable and negative information.

[1008] "Means for notifying the user of the results of an action" refers to a mechanism that notifies the user of the results of an action taken on their device, allowing the user to verify the reliability and authenticity of the information.

[1009] This invention is a system that detects false information and misinformation on the Internet in real time and warns users. This system includes a series of processes that collect, process, and notify data between servers, terminals, and users.

[1010] The server first uses the API of a social media platform as a means of collecting information on the Internet. For example, it can use the Twitter API to collect tweets related to the "Great Noto Peninsula Earthquake." This allows the server to obtain new posts in real time. As a specific example, when a server collects posts related to the "Great Noto Peninsula Earthquake," it uses the Twitter API to collect tweets containing the relevant keyword.

[1011] Next, the server preprocesses the collected information. To remove unnecessary information from the collected data and convert it into a format that is easier to analyze, it uses the BeautifulSoup library to remove HTML tags and special characters, and then uses the NLTK library to tokenize the text. For example, it preprocesses the tweet "Information about the Noto Peninsula earthquake can be found here → [link]" by removing the link and separating it into words.

[1012] The preprocessed data is then analyzed using natural language processing techniques. The server performs sentiment analysis using Hugging Face Transformers to determine positive, negative, or neutral sentiment. It also evaluates the reliability of the collected information sources and assigns a credibility score. For example, the server might perform sentiment analysis on a tweet such as "This earthquake is having a major impact on people" and assign it a negative score.

[1013] The server then combines these analysis results to determine the reliability of the information. For unreliable and negative information, it sends a deletion request to the social media platform, and for suspicious information, it adds a warning message. Users can request the deletion of specific tweets or the display of warning messages via the Twitter API. For example, they can send a request with a warning that "This is a hoax about the Noto Peninsula earthquake."

[1014] Finally, the server notifies the user of the results of these actions. The device displays a notification on the user's smartphone or other device, allowing the user to check the content of the notification. For example, the device may send the user a notification message such as, "Unreliable information has been detected and deleted. Please check for details."

[1015] This system enables the rapid detection and response of false information and misinformation, and provides accurate information to users. For example, keywords can be specified by inputting the following prompt sentences into the generative AI model:

[1016] Prompt Sentence Examples

[1017] Collect the latest information on the "Noto Peninsula Earthquake" and be alerted to unreliable and negative information.

[1018] Based on this prompt, the system will detect fake and misinformation related to the specified keywords in real time and notify the user.

[1019] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1020] Step 1:

[1021] Data Collection Settings

[1022] The server loads a pre-specified list of keywords to be used for data collection. For example, the keyword "Noto Peninsula Earthquake" is added to the list. This list is used for subsequent API queries.

[1023] Input: A specified list of keywords

[1024] Output: Keyword list loaded

[1025] Specific behavior:

[1026] The server retrieves "Noto Peninsula Earthquake" from the keyword list and sets up data collection.

[1027] Step 2:

[1028] Using SNS API

[1029] The server uses the API of the social media platform to retrieve posts based on the target keywords. For example, the Twitter API is used to collect tweets related to the "Noto Peninsula Earthquake." The server uses the Streaming API to retrieve new posts in real time.

[1030] Input: Keywords to be collected

[1031] Output: Collected SNS post data

[1032] Specific behavior:

[1033] The server uses Twitter's Streaming API to retrieve tweets related to the keyword "Noto Peninsula Earthquake" in real time.

[1034] Step 3:

[1035] Data Preprocessing

[1036] The server cleans the collected data and converts it into a format that is easy to analyze, using BeautifulSoup to remove HTML tags and special characters, and the NLTK library to tokenize the text.

[1037] Input: Collected social media posting data

[1038] Output: Preprocessed text data

[1039] Specific behavior:

[1040] The server removes the link from the tweet "Information about the Noto Peninsula earthquake can be found here → [link]" and separates the text.

[1041] Step 4:

[1042] Conducting sentiment analysis

[1043] The server analyzes the preprocessed text data using natural language processing techniques. Specifically, it uses Hugging Face Transformers to perform sentiment analysis on the text data and determine positive, negative, or neutral sentiment. It assigns a sentiment score to each piece of text.

[1044] Input: Preprocessed text data

[1045] Output: Text data with sentiment scores

[1046] Specific behavior:

[1047] The server performs sentiment analysis on tweets such as "This earthquake is having a big impact on people" and assigns them a negative score.

[1048] Step 5:

[1049] Conducting reliability evaluation

[1050] The server compares the collected information with a historical database and calculates a reliability score to assess the reliability of the information source.

[1051] Input: Text data with sentiment scores

[1052] Output: Text data with confidence scores

[1053] Specific behavior:

[1054] The server references the poster's past reliability data and calculates a reliability score.

[1055] Step 6:

[1056] Integration of reliability judgments

[1057] The server combines the sentiment score and the reliability score to determine the overall reliability of the information. For information that is low in reliability and negative, it proceeds to the next step.

[1058] Input: Text data with confidence scores

[1059] Output: Judgment result

[1060] Specific behavior:

[1061] The server selects posts with a "negative" sentiment score and a low credibility score.

[1062] Step 7:

[1063] Execute Action

[1064] The server sends a request to the social media platform to remove unreliable and negative information or to add a warning message to the information, for example, through the Twitter API.

[1065] Input: Judgment result

[1066] Output: Request sent to the social media platform

[1067] Specific behavior:

[1068] The server sends a request via the Twitter API that includes a warning: "This is false information about the Noto Peninsula earthquake."

[1069] Step 8:

[1070] User Notification

[1071] The device notifies the user of the result of the action received from the server, for example by displaying a notification on the user's smartphone so that the user can check the content.

[1072] Input: Action result

[1073] Output: Notification message sent to the user

[1074] Specific behavior:

[1075] The device will display a notification on the user's smartphone saying, "Unreliable information has been detected and removed. Please check for details."

[1076] (Application example 1)

[1077] 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."

[1078] The speed at which information spreads on the Internet has created an environment in which false and misinformation can easily spread. It is extremely important to detect such false and misinformation in real time and to warn users appropriately. However, conventional technology has made it difficult to do this efficiently and quickly. There is a need to provide a system that can solve these issues, more reliably evaluate the reliability of information, and provide appropriate notifications.

[1079] 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.

[1080] In this invention, the server includes means for collecting information on the Internet, means for preprocessing the collected information, means for analyzing the preprocessed information and performing sentiment analysis and reliability determination, means for requesting deletion and adding a warning message to unreliable and negative information, and means for notifying the user of the results of the action, thereby enabling the effective detection of false information and misinformation in real time and notifying the user.

[1081] "Means for collecting information on the Internet" refers to methods and devices for obtaining the latest posted information from social networking sites, news sites, etc.

[1082] "Means for preprocessing collected information" refers to a processing method and device for removing unnecessary data from the acquired information and converting it into a format that is easy to analyze.

[1083] "Means for analyzing pre-processed information and performing sentiment analysis and credibility determination" refers to methods and apparatus for analyzing tokenized text data using natural language processing techniques to determine sentiment from the context of the text and assess the credibility of the source.

[1084] "Means for executing requests to delete and add warning messages to unreliable and negative information" refers to a method and device for sending a request to delete information determined to be unreliable or a request to add a warning message to a social media platform.

[1085] "Means for notifying users of the results of actions" refers to methods and devices for notifying users in an appropriate format of the results of actions taken to remove or warn against false or misleading information.

[1086] This invention is a system for detecting false and misleading information on the Internet in real time and warning users, and is composed of the following components:

[1087] Data collection

[1088] The server collects the latest posting information from social media platforms, news sites, etc. To collect the data, it uses the API of the social media platform to automatically obtain new posting information.

[1089] Data Preprocessing

[1090] The server filters unnecessary data (HTML tags and special characters) from the retrieved information and tokenizes the text, using the Python libraries BeautifulSoup and NLTK (Natural Language Toolkit) for this process.

[1091] Data analysis

[1092] The server analyzes the tokenized text data using natural language processing technology. Sentiment analysis and credibility assessment are performed here. Sentiment analysis determines and scores positive, negative, or neutral sentiment from the context of the text. Credibility assessment verifies whether the collected information source is trustworthy and assigns a credibility score. This analysis uses Python NLP libraries (spaCy, NLTK).

[1093] Reliability determination

[1094] The server uses natural language processing technology to analyze the data and then determines the reliability of the information. This includes comparing it with trusted sources and checking against past data. If the information is deemed unreliable and negative, it proceeds to the next step.

[1095] Action Execution

[1096] A request to delete or add a warning message to unreliable and negative information is sent to the SNS platform. The server uses the SNS platform API to send the request to delete or add a warning message.

[1097] User Notification

[1098] The device will notify the user that the deletion request has been carried out and will display a warning message, allowing the user to reconfirm the authenticity of the information.

[1099] Specific examples

[1100] For example, if a user sets the keyword "novel coronavirus vaccine hoax," the server will collect posts related to "coronavirus vaccine." The collected data will be tokenized after removing HTML tags and special characters. The collected data will then be analyzed using natural language processing technology to perform sentiment analysis and determine trustworthiness. For posts that are deemed untrustworthy and negative, a request will be sent to the social media platform to delete them, and a request will be sent to display a warning message for suspicious information. The device will then send a warning notification to the user, allowing them to reconfirm the trustworthiness of the information.

[1101] Prompt Sentence Examples

[1102] Collect the latest posts about "COVID-19 vaccines" from social media and news sites. Remove HTML tags and special characters from the collected information and tokenize the text. Next, use natural language processing technology to perform sentiment analysis and credibility assessment to identify unreliable and negative information. Then, send a request to the social media platform to remove the information and add a warning message, and finally send a push notification to the user.

[1103] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1104] Step 1:

[1105] The server uses the APIs of social media and news sites to collect the latest post information based on specified keywords. The data retrieved by the server includes metadata such as the post text, poster information, and posting date and time. The input for this process is keywords related to "novel coronavirus vaccine hoaxes," and the output is the raw post data.

[1106] Step 2:

[1107] The server filters unnecessary information from the collected data. Specifically, it removes HTML tags and special characters to generate clean text data. This process uses the Python library BeautifulSoup. The input is the raw post data collected in step 1, and the output is the filtered text data.

[1108] Step 3:

[1109] The server tokenizes the filtered data, splitting the text into words and converting it into a format that is easy to use with natural language processing. This process uses NLTK (Natural Language Toolkit). The input is the clean text data generated in step 2, and the output is the tokenized text data.

[1110] Step 4:

[1111] The server analyzes the tokenized data using natural language processing technology. This involves sentiment analysis and credibility assessment. Sentiment analysis involves scoring positive, negative, or neutral sentiment from the context of the text. Credibility assessment verifies whether the collected information source is trustworthy and assigns a credibility score. This process uses Python NLP libraries (spaCy, NLTK). The input is the tokenized text data generated in step 3, and the output is a sentiment score and a credibility score.

[1112] Step 5:

[1113] The server determines the reliability of the information based on the analysis results. If the reliability is low and the information is determined to be negative, it proceeds to the next step. The input is the emotion score and reliability score generated in step 4, and the output is the information whose reliability and emotion have been determined.

[1114] Step 6:

[1115] The server sends a request to the SNS platform to delete and add a warning message to unreliable and negative information. This process uses the SNS platform's API. The input is the information determined in step 5, and the output is the result of sending the request to the SNS platform.

[1116] Step 7:

[1117] The device notifies the user of the results of the deletion request or warning message addition request received from the server. The notification is performed using a push notification. The input is the request result returned from the SNS platform in step 6, and the output is a push notification sent to the user's device.

[1118] 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.

[1119] This invention is a system that detects fake and misleading information on the Internet in real time and warns users. By combining this system with an emotion engine that recognizes user emotions, it can comprehensively consider the reliability of information and user emotions.

[1120] The system of the present invention comprises the following means:

[1121] Data collection

[1122] The server collects public information from social media and the internet based on pre-specified keywords. Data collection is done by automatically retrieving new posts using the API of the social media platform. The retrieved data is temporarily stored in a database.

[1123] Data Preprocessing

[1124] The server pre-processes the collected data to remove unnecessary information, including removing HTML tags, removing special characters, and tokenizing the text, converting the data into a format that is easier to analyze.

[1125] Data analysis

[1126] The server then analyzes the preprocessed data using natural language processing (NLP) techniques. Analysis items include keyword frequency analysis, sentiment analysis, and information source verification. Sentiment analysis calculates a positive, negative, or neutral sentiment score based on the context of the text.

[1127] Reliability determination

[1128] The server evaluates the reliability of the information based on the data analysis results. It compares the data with reliable sources (e.g., official announcements, well-known news sites) and calculates a reliability score. Information with low reliability and negative results proceeds to the next step.

[1129] Using the Emotion Engine

[1130] The server recognizes the user's feelings toward the information using an emotion engine, which analyzes emotions from the user's browsing history and posted content to understand the potential emotional response to specific information.

[1131] Action Execution

[1132] The server sends a request to the social media platform to remove unreliable and negative information. It also sends a request to add a warning message to suspicious information, taking into account user sentiment. Based on this request, the social media platform deletes the post or displays a warning message.

[1133] User Notification

[1134] The terminal executes deletion requests and displays warning messages based on requests from the server. When a post is deleted, the poster is notified of the deletion, and when a post has a warning message attached, a warning message is displayed to warn viewers.

[1135] Specific examples

[1136] For example, suppose a server collects posts related to the "Noto Peninsula Earthquake." The collected data is stripped of HTML tags and special characters and the text is tokenized. Next, natural language processing technology is used to analyze the collected data, performing sentiment analysis and determining reliability. If the information is determined to be unreliable and negative, a request to delete the post is sent to the social media platform. A request is also sent to add a warning message, taking into account user sentiment, if necessary. Finally, the device displays a deletion or warning notice to the user, allowing them to confirm and take appropriate action.

[1137] This system will prevent the spread of false and misinformation and issue warnings that take user sentiment into account, allowing users to obtain accurate and reliable information and improving the information environment on the Internet.

[1138] The processing flow will be explained below.

[1139] Step 1:

[1140] The server collects public information from social media and the internet. This information collection is done by using the API of the social media platform to obtain new posts containing specified keywords (e.g., "Noto Peninsula Earthquake"). The collected data is temporarily stored in a database.

[1141] Step 2:

[1142] The server pre-processes the collected data to remove unnecessary information, such as removing HTML tags, eliminating special characters, and tokenizing the text, converting the data into a format that is easier to analyze.

[1143] Step 3:

[1144] The server then analyzes the preprocessed data using natural language processing (NLP) techniques. Key analysis items include keyword frequency analysis, context analysis, and sentiment analysis. Sentiment analysis calculates a positive, negative, or neutral sentiment score.

[1145] Step 4:

[1146] The server performs a credibility assessment: it compares the preprocessed data with trusted sources (e.g., official announcements, well-known news sites) and calculates a credibility score. This assessment determines whether the information is trustworthy.

[1147] Step 5:

[1148] The server uses an emotion engine to recognize the user's emotions toward the information. This emotion engine analyzes emotions from the user's browsing history and posted content to understand the potential emotional response to the information. This step is particularly effective when the authenticity of the information is unknown.

[1149] Step 6:

[1150] The server takes action against unreliable and negative information, specifically by sending a request to the social media platform to delete the relevant social media post, and also by sending a request to add a warning message to any suspicious information.

[1151] Step 7:

[1152] The terminal executes deletion requests and displays warning messages based on requests from the server. When a post is deleted, the poster is notified of the deletion, and when a post has a warning message attached, a warning message is displayed to warn viewers.

[1153] Step 8:

[1154] Users can check the warning messages and removal notices displayed on their devices, which allows them to reconfirm the reliability of the information provided and make an appropriate decision.

[1155] In this way, we can build a system that can detect fake and misleading information on the Internet in real time and respond quickly. In addition, by using an emotion engine, it is possible to take user emotions into account and provide more appropriate warnings and notifications.

[1156] Example 2

[1157] 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."

[1158] In recent years, a lot of false and misleading information has been spread on the Internet, resulting in an increasing number of cases where users believe the incorrect information. Furthermore, the impact of this information on users' emotions is also a problem that cannot be ignored. Conventional technologies lacked sufficient mechanisms to prevent the spread of false information and lacked measures that took user emotions into consideration.

[1159] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1160] In this invention, the server includes means for collecting information on the Internet, means for preprocessing the collected information, means for analyzing the preprocessed information and performing sentiment analysis and reliability determination, means for taking action on unreliable and negative information, means for recognizing user sentiment and grasping emotional reactions to the information, and means for notifying the user of the results of the action. This makes it possible to prevent the spread of false and misleading information on the Internet and to take warnings and countermeasures that take user sentiment into consideration.

[1161] "Information on the Internet" refers to data such as text, images, audio, and video that is publicly available on the Internet, including websites, social media platforms, blogs, and news sites.

[1162] "Means of collection" refers to methods and tools for obtaining information on the Internet, such as collecting data using APIs or web scraping technology.

[1163] "Preprocessing means" refers to processing methods and tools used to remove unnecessary data from collected information and convert it into a format that is easy to analyze.

[1164] "Means of analysis" refers to techniques for analyzing preprocessed data and extracting useful information. Specifically, this includes natural language processing, sentiment analysis, keyword frequency analysis, etc.

[1165] "Sentiment analysis" is a technique for identifying and scoring positive, negative, and neutral emotions from text data.

[1166] "Credibility assessment" is a method of evaluating how trustworthy information is based on its source and content.

[1167] "Means for taking action" refers to measures to take against unreliable and negative information, including sending a removal request or adding a warning message.

[1168] "Means for recognizing user emotions" refers to technology that analyzes users' browsing history and posted content to understand their emotional reactions to specific information.

[1169] "Means for notifying the user of the outcome of the action" means a method for informing the user about the action taken, including, for example, displaying a removal notice or a warning message.

[1170] "Filtering means" refers to processing methods or tools for removing unnecessary data from collected information.

[1171] "Special character removal" is the process of removing unnecessary symbols and special characters from text data.

[1172] "Text tokenization" is the process of dividing text data into units of words or phrases.

[1173] "Natural language processing technology" refers to technology that enables computers to understand, interpret, and generate human language. Examples include text analysis, context understanding, and machine translation.

[1174] A "trustworthiness score" is a numerical representation of the reliability of information based on an evaluation of its source and content.

[1175] An "emotion engine" is a tool or algorithm that analyzes a user's emotions and identifies their emotional response to specific information.

[1176] A "removal request" is an action requesting that a social media platform remove a specific post.

[1177] A "warning message" is a message that alerts the user to specific information.

[1178] This invention is a system that detects fake and misleading information on the Internet in real time and warns users. By combining this system with an emotion engine that recognizes user emotions, it can comprehensively consider the reliability of information and user emotions.

[1179] First, the server collects information from the Internet. Specifically, it uses the API of a social networking platform (for example, Twitter API or general web scraping methods) to automatically retrieve new posts based on specified keywords. This collected data is temporarily stored in a database (for example, MySQL or MongoDB).

[1180] The server then preprocesses the collected information, which includes removing HTML tags, stripping special characters, and tokenizing the text using Python's BeautifulSoup, regular expressions, and NLTK libraries, converting the data into a format that is easier to parse.

[1181] The server then analyzes the preprocessed data. Specifically, it uses natural language processing techniques (e.g., Spacy or BERT) to perform sentiment analysis and keyword frequency analysis of the text. It also verifies the source of the information and calculates a credibility score by comparing it with highly reliable sources (e.g., official announcements or well-known news sites).

[1182] If the information is judged to be unreliable and negative, the server uses an emotion engine to recognize the user's emotions. This engine analyzes emotions from the user's browsing history and posted content to understand the potential emotional reaction to specific information. This process uses Google Analytics and emotion analysis APIs (e.g., IBM Watson Tone Analyzer).

[1183] The server then takes action against unreliable and negative information, sending a request to the social media platform to remove it and also sending a request to add a warning message to any suspicious information.

[1184] Finally, the device notifies the user of deletion requests or warning messages based on requests from the server. For deleted posts, the device notifies the poster, and for posts with attached warning messages, the device displays a warning message to the viewer.

[1185] Specific examples

[1186] For example, consider a case where a server collects posts related to the "Noto Peninsula Earthquake." The collected data is preprocessed by removing HTML tags and special characters and tokenizing the text. Next, natural language processing techniques (e.g., Spacy or BERT) are used to analyze sentiment and determine reliability. If the information is determined to be unreliable and negative, a deletion request for the post is sent to the social media platform. A request with a warning message attached is also sent, taking into account the user's sentiment. The device displays the deletion or warning notification to the user, allowing the user to review it and take appropriate action.

[1187] Examples of prompt statements

[1188] Collect and preprocess the latest posts related to the "Noto Peninsula Earthquake." Next, use Spacy and BERT to perform sentiment analysis and determine trustworthiness, and create a deletion request and warning message for posts with low trustworthiness and negative views. Finally, notify users.

[1189] This system will prevent the spread of false and misinformation and will be able to issue warnings and take measures that take user sentiment into account, allowing users to obtain accurate and reliable information and improving the information environment on the Internet.

[1190] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1191] Step 1: Data collection

[1192] The server collects information from the Internet. Specifically, it uses the API of social media platforms and web scraping technology to automatically retrieve new posts based on specified keywords. The input is a specific keyword (e.g., "Noto Peninsula Earthquake"). The output is the collected post data (e.g., text, user name, posting date and time, etc.).

[1193] Step 2: Data Preprocessing

[1194] The server preprocesses the collected information. Specifically, it uses BeautifulSoup to remove HTML tags and regular expressions to remove special characters. It also uses NLTK to tokenize the text. The input for this step is the collected post data. The output is the preprocessed, clean text data.

[1195] Step 3: Data analysis

[1196] The server analyzes the preprocessed data. Specifically, it performs sentiment and keyword frequency analysis on the text using Spacy and BERT. It also compares it with established news sites and official announcements to verify the source of the information. The input for this step is the preprocessed, clean text data. The output is a sentiment score, keyword frequency, and confidence score.

[1197] Step 4: Reliability determination

[1198] The server evaluates the reliability of the information based on the data analysis results. Specifically, it checks whether the information matches a reliable source based on the obtained reliability score. The inputs to this step are the sentiment score, keyword frequency, and reliability score. The output is a judgment result of whether the information is highly or low reliable.

[1199] Step 5: Use the Emotion Engine

[1200] The server uses an emotion engine to recognize the user's emotions toward the information. It uses Google Analytics and an emotion analysis API to analyze emotions from the user's browsing history and posted content, and understands the emotional reaction to specific information. The inputs to this step are the credibility judgment results and the user's past browsing history and posted data. The output is the user's emotional reaction pattern.

[1201] Step 6: Take Action

[1202] The server takes action against unreliable and negative information by sending a deletion request to the social media platform and adding a warning message to suspicious information. The inputs to this step are the reliability judgment results and the user's emotional reaction patterns. As outputs, a deletion request and a warning message request are sent to the social media platform.

[1203] Step 7: Notify users

[1204] The device executes the deletion request or displays the warning message based on the request from the server. Specifically, the notification is displayed using JavaScript or HTML. The input to this step is the deletion request and warning message request from the server. The output is the deletion notification or warning notification displayed to the user.

[1205] This processing flow makes it possible to evaluate the reliability of information on the Internet and provide appropriate measures and notifications to users.

[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 the electronic payment process, users often make transactions based on unreliable information or fake reviews, which can lead to unreliable transactions and fraud. To solve this problem, a system is needed that can evaluate the reliability of information related to transactions in real time and warn users in a timely manner.

[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 means for collecting information from the Internet, means for preprocessing the collected information, means for analyzing the preprocessed information and performing sentiment analysis and reliability determination, means for taking action on information that is low in reliability and negative, means for notifying the user of the results of the action, and means for evaluating the reliability of transaction information in the payment system and displaying a warning message. This makes it possible to evaluate the reliability of transaction information during electronic payment and issue an appropriate warning to the user.

[1211] The "Internet" is a system of interconnected computer networks around the world that enables the exchange of information.

[1212] "Information collection means" is a system that automatically acquires data from the Internet and uses APIs to collect the necessary information.

[1213] "Preprocessing" refers to the process of converting collected information into a format that is easier to analyze, by removing unnecessary data and special characters and tokenizing the text.

[1214] The "analysis method" is a system that uses preprocessed data to evaluate the reliability of information and emotions, and is carried out using natural language processing technology.

[1215] "Sentiment analysis" is a technique for reading emotions from text data and identifying positive, negative, and neutral emotions.

[1216] "Credibility determination" is the process of assessing the trustworthiness of a source and calculating a credibility score, which is then compared to other trusted sources.

[1217] "Negative information" refers to information that may have a negative impact on users and is identified through the results of sentiment analysis.

[1218] The "action execution means" is a mechanism for performing specific processing on information that is determined to be unreliable and negative, such as adding a warning message or sending a deletion request.

[1219] "Notification means" refers to the function of informing the user of the results of an action taken, and is performed through an application or device.

[1220] A "payment system" is a system for electronically conducting monetary transactions and allowing users to purchase goods and services online.

[1221] A "warning message" is a message that informs users that the information is unreliable and helps them avoid inappropriate transactions.

[1222] The system for realizing this invention is composed of three main elements: a server, a terminal, and a user. The specific configuration and operation of the system are explained in detail below.

[1223] The server collects information from the Internet, preprocesses it, and performs sentiment analysis and reliability assessment. The data is collected using APIs from social media and review sites. For example, the Twitter API and Facebook Graph API are used to collect reviews and posts based on specific keywords. The collected data is then temporarily stored in a MySQL database.

[1224] The server then pre-processes the collected information, which includes removing HTML tags and special characters, and tokenizing the text to convert the data into a format that is easier to parse. This process is performed using the Python libraries NLTK and SpaCy.

[1225] The preprocessed data is then analyzed using natural language processing (NLP) techniques. Specifically, the text is subjected to sentiment analysis and credibility assessment. Sentiment analysis involves identifying positive, negative, and neutral sentiment from the text and calculating a sentiment score. TextBlob and IBM Watson Emotion Analysis are used for this task. Credibility assessment involves comparing the data with official sources (e.g., well-known news sites or official announcements) and calculating a credibility score.

[1226] Actions are taken against information that is deemed unreliable and negative. These actions include sending a deletion request to the social media platform or displaying a warning message to the user. After the action is taken, the result is notified to the user's device. This notification allows the user to receive appropriate information and avoid inappropriate transactions.

[1227] As an example of how this system can be used, consider the electronic payment app "SecurePay Warning System." When a user browses a specific product or store through this app, the server collects reviews related to that product or store, performs sentiment analysis, and determines its trustworthiness. If it is determined to be untrustworthy, a warning message is displayed on the user's device. This warning allows the user to make transactions safely.

[1228] As an example, use the following prompt:

[1229] Please analyze this review body: "{review body}" and provide a rating and sentiment score. User ID: {user ID}, Purchase history: {purchase history}"

[1230] The system will protect users from false and misinformation online and allow them to make decisions based on more reliable information.

[1231] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1232] Step 1:

[1233] The server collects information from the Internet. Specifically, it uses the APIs of social media and review sites to collect reviews and posts based on specified keywords. For example, it uses the Twitter API or Facebook Graph API to obtain posts related to a "specific product name." The collected data is then temporarily stored in a MySQL database.

[1234] Input: Keywords, SNS API request

[1235] Output: Collected reviews and submission data (JSON format)

[1236] Step 2:

[1237] The server preprocesses the collected information, removing HTML tags and special characters, and tokenizing the text. It uses the Python libraries NLTK and SpaCy to convert the collected data into a format that is easy to analyze.

[1238] Input: Collected reviews and submission data

[1239] Output: Preprocessed text data (cleaned)

[1240] Step 3:

[1241] The server analyzes the preprocessed data. Based on the preprocessed text data, natural language processing techniques are used to perform sentiment analysis and reliability determination. For sentiment analysis, TextBlob and IBM Watson Emotion Analysis are used to calculate positive, negative, or neutral sentiment scores from the text. For reliability determination, a reliability score is calculated by comparing it with official sources.

[1242] Input: Preprocessed text data

[1243] Output: sentiment score, confidence score

[1244] Step 4:

[1245] The server takes action against unreliable and negative information. Specifically, it sends a deletion request to the social media platform or a request to add a warning message. If the deletion request is accepted, the information is deleted from the platform. In addition, information with a warning message is displayed to warn viewers.

[1246] Input: sentiment score, confidence score

[1247] Output: Delete request, warning message request

[1248] Step 5:

[1249] The terminal will inform the user of the results of the action. Warning messages will be displayed for unreliable information, allowing the user to avoid inappropriate transactions. Warning messages will be delivered to the user through the mobile application, helping them to carry out safe transactions.

[1250] Input: Delete request result, Warning message request result

[1251] Output: User notification (warning message)

[1252] Step 6:

[1253] The user will review the notification and take appropriate action based on its content, for example, choosing to avoid the transaction if a warning message is displayed, or confirming that the deletion request has been accepted and the information has been deleted.

[1254] Input: User notification (warning message)

[1255] Output: User decision

[1256] 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.

[1257] 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.

[1258] 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.

[1259] 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.

[1260] 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.

[1261] 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.

[1262] 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).

[1263] 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.

[1264] 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."

[1265] 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.

[1266] 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).

[1267] 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.

[1268] 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.

[1269] 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.

[1270] 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.

[1271] 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.

[1272] 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.

[1273] 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.

[1274] 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.

[1275] 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.

[1276] 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.

[1277] The following is further disclosed regarding the above embodiment.

[1278] (Claim 1)

[1279] means of collecting information on the Internet;

[1280] means for preprocessing the collected information;

[1281] means for analyzing the pre-processed information and performing sentiment analysis and credibility determination;

[1282] a way to take action on unreliable and negative information;

[1283] A means of informing the user of the outcome of their action;

[1284] A system including:

[1285] (Claim 2)

[1286] 10. The system of claim 1, further comprising filtering means for removing unnecessary data from the collected information, and preprocessing means for removing special characters and tokenizing text.

[1287] (Claim 3)

[1288] 10. The system of claim 1, wherein the analysis means is configured to perform sentiment analysis and credibility determination using natural language processing techniques.

[1289] "Example 1"

[1290] (Claim 1)

[1291] means of collecting information on the Internet;

[1292] means for preprocessing the collected information;

[1293] means for analyzing the preprocessed information using natural language processing techniques to perform sentiment analysis and credibility determination;

[1294] A means to take action to delete or add a warning message to unreliable and negative information;

[1295] A means of informing the user of the outcome of their action;

[1296] A system including:

[1297] (Claim 2)

[1298] 10. The system of claim 1, further comprising filtering means for removing unnecessary data from the collected information, and preprocessing means for removing special characters and tokenizing text.

[1299] (Claim 3)

[1300] 10. The system of claim 1, wherein the analysis means is configured to utilize a generative AI model to perform sentiment analysis and credibility determination.

[1301] "Application Example 1"

[1302] (Claim 1)

[1303] means of collecting information on the Internet;

[1304] means for preprocessing the collected information;

[1305] means for analyzing the pre-processed information and performing sentiment analysis and credibility determination;

[1306] A means for executing a request to delete unreliable and negative information and a request to add a warning message;

[1307] A means of informing the user of the outcome of their action;

[1308] A system including:

[1309] (Claim 2)

[1310] 10. The system of claim 1, further comprising filtering means for removing unnecessary data from the collected information, and preprocessing means for removing special characters and tokenizing text.

[1311] (Claim 3)

[1312] 10. The system of claim 1, wherein the analysis means is configured to perform sentiment analysis and credibility determination using natural language processing techniques.

[1313] "Example 2: Combining Emotion Engines"

[1314] (Claim 1)

[1315] means of collecting information on the Internet;

[1316] means for preprocessing the collected information;

[1317] means for analyzing the pre-processed information and performing sentiment analysis and credibility determination;

[1318] a way to take action on unreliable and negative information;

[1319] A means of recognizing users' emotions and understanding their emotional responses to information;

[1320] A means of informing the user of the outcome of their action;

[1321] A system including:

[1322] (Claim 2)

[1323] 10. The system of claim 1, further comprising filtering means for removing unnecessary data from the collected information, and preprocessing means for removing special characters and tokenizing text.

[1324] (Claim 3)

[1325] 10. The system of claim 1, wherein the analysis means is configured to perform sentiment analysis and credibility determination using natural language processing techniques.

[1326] (Claim 4)

[1327] 10. The system of claim 1, further comprising means for calculating a reliability score for the collected information and matching it with a trusted source.

[1328] (Claim 5)

[1329] 10. The system of claim 1, further comprising means for using an emotion engine to analyze emotions from a user's browsing history and posted content to understand potential emotional responses to specific information.

[1330] (Claim 6)

[1331] 2. The system of claim 1, further comprising means for sending a request to the social media platform to remove unreliable and negative information and a request to add a warning message to suspicious information.

[1332] "Application example 2 when combining emotion engines"

[1333] (Claim 1)

[1334] means of collecting information on the Internet;

[1335] means for preprocessing the collected information;

[1336] means for analyzing the pre-processed information and performing sentiment analysis and credibility determination;

[1337] a way to take action on unreliable and negative information;

[1338] A means of informing the user of the outcome of their action;

[1339] means for evaluating the reliability of transaction information in the payment system and displaying a warning message;

[1340] A system including:

[1341] (Claim 2)

[1342] 10. The system of claim 1, further comprising filtering means for removing unnecessary data from the collected information, and preprocessing means for removing special characters and tokenizing text.

[1343] (Claim 3)

[1344] 10. The system of claim 1, wherein the analysis means is configured to perform sentiment analysis and credibility determination using natural language processing techniques. [Explanation of symbols]

[1345] 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. means of collecting information on the Internet; means for preprocessing the collected information; means for analyzing the pre-processed information and performing sentiment analysis and credibility determination; a way to take action on unreliable and negative information; A means of informing the user of the outcome of their action; A system including:

2. 10. The system of claim 1, further comprising filtering means for removing unnecessary data from the collected information, and preprocessing means for removing special characters and tokenizing text.

3. 10. The system of claim 1, wherein the analysis means is configured to perform sentiment analysis and credibility determination using natural language processing techniques.

Images