system

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting published posts; means for analyzing the collected published posts and extracting posts including a specific keyword; means for collating contents of the extracted posts with official information of a company; means for identifying erroneous information based on a collation result and setting a flag to the erroneous information; means for generating correction information for a post determined to be erroneous information; and means for posting the generated correction information to an SNS.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] When false information about a company or product name is spread on social media, it can have a negative impact on the company's credibility and stock price. However, there are limits to what a company can do to fact-check information in-house, and corrections made by a company are not necessarily accepted as legitimate by society as a whole. For this reason, there is a need for an efficient and accurate system to verify the veracity of information spread on social media and quickly provide corrections, but currently no such system exists. The objective of this invention is to improve this situation, protect the credibility of companies, and widely provide accurate information to society. [Means for solving the problem]

[0005] The present invention provides a system including a means for collecting public posts, a means for analyzing the collected public posts to extract posts containing specific keywords, a means for comparing the content of the extracted posts with official company information, a means for identifying and flagging misinformation based on the comparison results, a means for generating corrections for posts determined to be misinformation, and a means for posting the generated corrections to a social networking site (SNS). This system enables rapid detection of misinformation spreading on SNS and widespread provision of corrections to society. The corrections include a citation for the official source, ensuring accuracy and maintaining neutrality. Furthermore, the system includes a means for providing a report of the collected data, the history of corrections, and the correction results, allowing companies to accurately grasp the situation and take appropriate measures.

[0006] "Public posts" refers to all posted data that is publicly available on social networking services (SNS).

[0007] "Means of collection" refers to the function of automatically crawling public posts on social media and storing posts containing specific keywords in a database.

[0008] The "means of analyzing and extracting posts containing specific keywords" is a function that analyzes collected public posts using text analysis technology to identify and extract posts that contain specific keywords such as company names or product names.

[0009] "Means for matching" refers to a function that compares the content of extracted posts with official company information to determine whether they match or mismatch.

[0010] "Means to identify and flag misinformation" refers to the ability to identify posts as misinformation when their content contradicts official information and add an identifying marker for management within the system.

[0011] The "means of generating correction information" is a function that automatically creates a correction based on official, correct information for posts that are determined to be incorrect information.

[0012] "Means of posting on social media" refers to the function of making the generated correction information public through official social media accounts, etc., and spreading corrections to misinformation.

[0013] "Source" refers to the source of official information used in the correction, and is a reference for verifying accuracy.

[0014] The "means of providing in report format" refers to a function that aggregates collected data and generated correction information, documents it in a form that includes past history and impact, and provides it to companies. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0023] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0036] The present invention provides a system for preventing the spread of false information about company and product names on social media, and provides a process for automatically collecting, analyzing, and collating public posts and correcting the false information. Specific embodiments of the system are described below.

[0037] System Configuration

[0038] The system consists of the following main components:

[0039] 1. Server: A central location responsible for collecting, analyzing, collating data, identifying misinformation, and generating and distributing corrections.

[0040] 2. Device: A device on which a user uses SNS and posts publicly. If incorrect information is detected, you may receive a correction notice.

[0041] 3. Users: People who post and share information on social media.

[0042] Program processing

[0043] 1. Collection of information

[0044] The server periodically uses the social networking service's API to crawl public posts containing specific keywords, such as company names or product names.

[0045] Example: The server uses the keyword "Company X" to collect relevant tweets from Twitter's API. At the same time, metadata such as the post content, user information, posting date and time, number of retweets, and number of likes are also obtained.

[0046] 2. Analysis and extraction of post content

[0047] The server then runs the collected public posts through a text analysis algorithm, using natural language processing (NLP) techniques, to extract posts containing specific keywords.

[0048] Example: Analyze 1,000 public posts collected by the server to find posts containing the specified keyword "Company X" and extract 100 posts.

[0049] 3. Comparison with official information

[0050] The server then compares the content of the extracted public posts with official company information, often obtained through web scraping or RSS feeds, to determine whether the post matches existing official information.

[0051] Example: A server compares a post that says "Company X's new product is being offered for free" with information on the official website that says "The new product costs $100" and recognizes a discrepancy.

[0052] 4. Identifying and flagging misinformation

[0053] The server will then flag any posts that are inconsistent based on the results of the match, and posts identified as misinformation will be moderated within the system.

[0054] Example: A post containing the misinformation that servers are "provided for free" is flagged and added to a misinformation list.

[0055] 5. Generating Corrections

[0056] The server automatically generates corrections based on official information to correct misinformation, including the correct information and the source.

[0057] Example: The server creates a correction that reads, "Company X's new product costs $100. More information here."

[0058] 6. Distribution of Corrections

[0059] The server posts the generated correction information to the social networking site, and if necessary, notifies the user who posted the incorrect information and the user who retweeted it.

[0060] Example: In response to a false claim that "Company X's new product will be provided for free," the server tweets a correction from its official account and also sends a notification to users who have retweeted the false information.

[0061] Operational Management and Reporting

[0062] The server periodically generates reports on the collected data and the results of detecting and correcting misinformation, and provides them to companies and relevant public institutions. The reports include a list of detected misinformation, the spread of the correction information, and its impact.

[0063] Example: A server generates a monthly report and sends it to a company's marketing department, detailing the list of misinformation detected and the extent to which corrections to those misinformation have been disseminated.

[0064] In this way, this system efficiently detects misinformation on social media and quickly distributes corrections, thereby protecting the credibility of companies and providing accurate information.

[0065] The processing flow will be explained below.

[0066] Step 1:

[0067] The server accesses the social networking service's API and crawls public posts containing specific keywords (company names or product names). It sets the API endpoint and request parameters and periodically retrieves data.

[0068] Example: A server periodically crawls tweets containing the keyword "Company X" and retrieves metadata including the post content, user information, posting date and time, number of retweets, and number of likes.

[0069] Step 2:

[0070] The server crawls and retrieves the posted data, which is then passed through a text analysis algorithm. Natural language processing (NLP) techniques are used to analyze and extract posts containing specific keywords.

[0071] Example: The server analyzes 1,000 tweets collected and finds posts containing "Company X" and extracts 100 of them.

[0072] Step 3:

[0073] The server uses web scraping or RSS feeds to obtain official company information, and stores the information in a database.

[0074] Example: A server web scrapes new product information and product pricing information from the official website of "Company X" and stores it in a database.

[0075] Step 4:

[0076] The server compares the extracted social media posts with official information, using natural language processing technology to determine whether the posts match the official information.

[0077] Example: A server analyzes a social media post that says, "Company X's new product is being offered for free," and compares it with information on the official website that says, "The price of the new product is $100."

[0078] Step 5:

[0079] If the server does not match, it will flag the post as misinformation and keep it in a database.

[0080] Example: A post claiming that a server is "provided for free" is detected as contradicting official information, and the information is flagged as misinformation and added to a moderation list.

[0081] Step 6:

[0082] The server automatically generates a correction based on official information to correct misinformation. The correction includes the source of the official information and accurate information.

[0083] Example: The server automatically generates a correction: "Company X's new product costs $100. Official information is here."

[0084] Step 7:

[0085] The server generates a correction and posts it to the social networking site. If necessary, notifications are also sent to the user who posted the incorrect information and to users who retweeted it.

[0086] Example: In response to a false claim that "Company X's new product will be provided for free," the server tweets a correction from its official account and also sends a notification to users who have retweeted the false information.

[0087] Step 8:

[0088] The server generates a report of the collected data, the history of correction information, and the correction results, and provides it to companies and relevant public institutions.

[0089] Example: A server generates a monthly report and sends it to a company's marketing department, detailing the list of misinformation detected and the extent to which corrections to those misinformation have been disseminated.

[0090] In this way, this system efficiently detects misinformation on social media and quickly distributes corrections, thereby protecting the credibility of companies and providing accurate information.

[0091] Example 1

[0092] 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."

[0093] Currently, inaccurate and false information can spread quickly on online social networking services (SNS), causing problems that undermine the credibility of companies and products. Furthermore, delays in responding to false information can lead to further confusion, which could damage brand image and cause consumer confusion.

[0094] 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.

[0095] In this invention, the server includes means for collecting public posts, means for applying a text analysis algorithm to the collected public posts to extract posts containing specific keywords, means for comparing the content of the extracted posts with official information, means for identifying and flagging misinformation based on the comparison results, means for generating corrections containing correct information for posts determined to be misinformation, means for posting the generated corrections to a social networking service, and means for sending notifications to users who posted and retweeted the misinformation as needed, thereby enabling inaccurate and misinformation to be detected quickly and efficiently and corresponding corrections to be provided.

[0096] "Public Post" means a user's post that is publicly available on a social networking service.

[0097] "Means of collection" refers to algorithms and programs that automatically obtain public posts using APIs of social networking services, etc.

[0098] A "text analysis algorithm" refers to a program that uses natural language processing technology to extract specific keywords and meanings from text data.

[0099] "Means of extraction" refers to the process of using a text analysis algorithm to select data that meets certain conditions.

[0100] "Verification measures" refer to programs that compare the collected public posts with official information to check for matches or inconsistencies.

[0101] "Misinformation" refers to information that does not match official information or contains factually incorrect information.

[0102] "Flagging" refers to the process of assigning an identifying mark to information that meets certain conditions.

[0103] "Correction information" refers to content that corrects misinformation and includes correct information and sources of official information.

[0104] "Means for generating" refers to a program for automatically creating correct correction information for erroneous information.

[0105] "Means of posting" refers to the algorithms or programs used to disseminate the generated correction information on social networking services.

[0106] "Means for sending notifications" refers to a program for sending corrections to users who posted false information or spread it.

[0107] "Social networking service" refers to a service that allows users to share information with each other over the Internet.

[0108] "Report format" refers to a document format that visually presents the history of collected data and correction information in an easy-to-understand manner.

[0109] The present invention provides a system for preventing the spread of false information about company and product names on social networking services (SNS), and provides a process for automatically collecting, analyzing, and collating public posts and correcting the false information. A specific embodiment of this system will be described.

[0110] System Configuration

[0111] The system consists of the following main components:

[0112] 1. Server: A central location responsible for collecting, analyzing, collating data, identifying misinformation, and generating and distributing corrections.

[0113] 2. Device: A device on which a user uses SNS and posts publicly. If incorrect information is detected, you may receive a correction notice.

[0114] 3. Users: People who post and share information on social media.

[0115] Hardware and Software Use

[0116] The server should be a computer with powerful computing capabilities. Recommended hardware includes a fast CPU, large memory capacity, and an SSD. Software used includes Python, NLTK, spaCy, BeautifulSoup, and the Twitter API.

[0117] For example, a server periodically uses the Twitter API to collect tweets containing the keyword "Company X." It then uses a Python script to perform text analysis using NLTK and spaCy to extract posts containing specific keywords. The server then uses BeautifulSoup to scrape the latest information from the company's official website and compares the extracted posts with official information.

[0118] Generation and distribution of corrections

[0119] If a contradiction is detected, the server automatically generates a correction based on the official information. The correction includes the correct information and its source. The server then reposts the correction to the social networking site, and notifies the user who posted the misinformation and those who retweeted it. This quickly prevents the spread of misinformation.

[0120] Operational Management and Reporting

[0121] The server periodically generates reports on the collected data, misinformation detection, and correction results, and provides them to companies and relevant public institutions. The reports include a list of detected misinformation, the spread of correction information, and its impact. For example, the server generates a monthly report, converts it to PDF format, and emails it to the company's marketing department.

[0122] Prompt Sentence Examples

[0123] Please describe a system that detects misinformation about Company X on social media and automatically generates and distributes corrections. Please provide details, including examples, of how this system works.

[0124] In this way, this system efficiently detects misinformation on social media and quickly distributes corrections, thereby protecting the credibility of companies and providing accurate information.

[0125] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0126] Step 1:

[0127] The server collects public posts. The server periodically crawls public posts containing specific keywords, such as "Company X," using the API of the social networking site. Specifically, the server uses Python to access the Twitter API and collect tweets containing the keyword "Company X." The data obtained from the API of the social networking site includes metadata such as the post content, user information, posting date and time, number of retweets, and number of likes. This provides the public posts to be collected as input data.

[0128] Step 2:

[0129] The server runs the collected public posts through a text analysis algorithm to extract posts containing specific keywords. Specifically, the server uses natural language processing libraries such as NLTK and spaCy to analyze collected tweets for posts containing the keyword "Company X" and extract the relevant posts. At this stage, the collected post data is input and posts that match the keywords are extracted.

[0130] Step 3:

[0131] The server compares the extracted public post content with official information. Specifically, the server uses a web scraping tool such as BeautifulSoup to retrieve the latest product information and news from the company's official website. It then compares the extracted post content with the retrieved official information to check for matches and inconsistencies. The input data is the extracted post and official information, and the comparison results are output based on this.

[0132] Step 4:

[0133] The server identifies and flags misinformation based on the matching results. Specifically, the server flags posts containing the misinformation "provided for free" as misinformation and stores it in a database. The input data is the matching results, and based on the results, misinformation is identified and flagged data is output.

[0134] Step 5:

[0135] The server generates correction information based on official information. Specifically, the server automatically generates a correction statement that includes the correct information and its source. For example, it generates a correction statement such as, "Company X's new product costs $100. Click here for more information." The input data is official information and incorrect information, and the output is correction information.

[0136] Step 6:

[0137] The server posts the generated correction information to SNS, and if necessary, sends notifications to the user who posted the misinformation and to users who retweeted it.Specifically, the server tweets the correction information on its official account and also sends individual notifications to users who retweeted the misinformation.The input data is the generated correction information, and the output is posted to SNS, and notifications are sent.

[0138] Step 7:

[0139] The server generates and provides the collected data, correction information history, and correction results in report format.Specifically, the server creates a monthly report, converts it to PDF format, and emails it to the company's marketing department.The input data are the collected data and correction results, and the report is generated and provided as output.

[0140] (Application example 1)

[0141] 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."

[0142] The spread of misinformation on social media is a serious problem that damages the trust of companies and individuals. In particular, with the rapid increase in the number of social media users, misinformation spreads quickly, creating a need for systems that provide fast and accurate corrections. There is also a growing need for automated solutions that can save users the trouble of monitoring and correcting relevant misinformation themselves.

[0143] 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.

[0144] In this invention, the server includes means for collecting public posts, means for analyzing the collected public posts to extract posts containing specific keywords, means for comparing the content of the extracted posts with official information, means for identifying misinformation based on the comparison results and flagging the misinformation, means for generating corrections to posts determined to be misinformation, means for posting the generated corrections to an online environment, means for setting specific keywords and notifying users when misinformation is detected, means for automatically generating corrections using a generative AI model, and means for setting prompt text for the generative AI model. This enables quick detection and correction of misinformation spreading on SNS and quick notification to users.

[0145] A "means for collecting public posts" is a device or program that has the functionality to automatically collect posts that are publicly available on social media or online platforms.

[0146] "Means for analyzing collected public posts and extracting posts containing specific keywords" refers to a device or program that has the function of analyzing collected post data and identifying and extracting posts containing specified keywords.

[0147] "Means for comparing the content of extracted posts with official information" refers to a device or program that has the function of comparing and matching the content of extracted posts with official information provided by companies or official institutions.

[0148] "Means for identifying false information based on the matching results and flagging false information" refers to a device or program that has the function of detecting false information from the matching results and flagging posts that are determined to be false information.

[0149] "Means for generating correction information for posts determined to contain false information" refers to a device or program that has the function of automatically generating a correction statement to provide accurate information for posts identified as false information.

[0150] "Means for posting the generated correction information online" refers to a device or program that has the function of automatically posting the generated correction information to social media or other online platforms.

[0151] "Means for setting specific keywords and notifying users when misinformation is detected" refers to a device or program that has the function of allowing users to set keywords of interest, collecting and analyzing information based on those keywords, and notifying users when misinformation is detected.

[0152] "Means for automatically generating correction information using a generative AI model" refers to a device or program that has the function of automatically generating correction information for erroneous information using an artificial intelligence model.

[0153] "Means for setting a prompt sentence to a generative AI model" refers to a device or program that has the function of setting a prompt sentence to be input to a generative AI model and performing information processing based on that prompt.

[0154] The present invention is a system that effectively detects misinformation spread on social media and quickly distributes corrections. The system includes the following main components:

[0155] System Configuration

[0156] 1. Server:

[0157] The server has the ability to collect public posts using the API of a social networking site. For example, it uses the Twython library to collect posts from Twitter that contain specific keywords.

[0158] The server uses the SpaCy library to process the collected data in natural language, analyzing and extracting posts containing specific keywords.

[0159] The server scrapes official information using BeautifulSoup and the requests library to match the extracted public posts with official information obtained from the websites of companies or official institutions.

[0160] Based on the results of the comparison, the server flags any posts that contradict official information as misinformation.

[0161] 2. Use of generative AI models:

[0162] For posts that are determined to contain false information, the server automatically generates corrections using a generative AI model. The generative AI model uses the latest natural language processing technology and generates corrections by setting prompt sentences.

[0163] For example, a prompt such as "We have detected a tweet that says, 'Company X's new product will be offered for free.' Please correct this information." can be input into the generative AI model to obtain appropriate correction information.

[0164] 3. User Notice:

[0165] If certain keywords are set, the server will notify users when misinformation is detected, allowing users to respond quickly.

[0166] 4. Distribution of Corrections:

[0167] The server has the function of posting the generated correction information online, for example, by posting the correction information from an official account on social media to spread it.

[0168] Specific examples

[0169] As a specific example, consider a case where misinformation related to Company X is spread on social media. In this case, the server collects related posts from Twitter using the keyword "Company X." By analyzing the posts, a tweet containing the misinformation "provided for free" is detected. The server compares it with official information, flags it as misinformation, and then uses a generative AI model to automatically generate a correction such as "Company X's new product costs $100. Click here for more information." The server then posts the generated correction on social media and sends a notification to the affected users.

[0170] This system enables the rapid detection and correction of misinformation, protecting the credibility of businesses and individuals. Furthermore, by utilizing generative AI models and prompts, corrections can be automatically generated, enabling efficient correction of information. Specific examples of prompts are as follows:

[0171] Example prompt sentence:

[0172] "We have detected a tweet that states 'Company X's new product will be offered for free.' Please correct this information."

[0173] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0174] Step 1:

[0175] The server uses the SNS API to collect public posts containing specific keywords. Specifically, it uses the Twython library to set keywords such as "Company X" and collects tweets that match the keywords. The input is the specific keyword, and the output is the collected post data. This collected data includes the post content, poster information, posting date and time, number of retweets, number of likes, etc.

[0176] Step 2:

[0177] The server analyzes the collected public posts using natural language processing (NLP) technology and extracts posts containing specific keywords. Specifically, it performs text analysis using the SpaCy library to extract posts containing the keyword "Company X." The input is the collected post data, and the output is the analyzed and extracted post data. The analysis includes data processing such as part-of-speech tagging and named entity extraction.

[0178] Step 3:

[0179] The server compares the post content extracted through analysis with official information. Specifically, it uses BeautifulSoup and the requests library to scrape official information from the company's official website and compares it with the extracted post content. The input is the extracted post data and official information, and the output is the comparison result. This comparison involves data calculations to determine whether the post content matches the official information.

[0180] Step 4:

[0181] Based on the matching results, the server identifies misinformation and flags the post. For example, if a post saying "Company X's new product will be offered for free" contradicts official information, it will be flagged as misinformation. The input is the matching results, and the output is the flagged post data. This flagging is a data processing process to identify misinformation.

[0182] Step 5:

[0183] The server uses a generative AI model to automatically generate corrections for posts determined to contain false information. Specifically, the prompt "We have detected a tweet stating that 'Company X's new product will be offered for free.' Please correct this information." is input into the generative AI model, which then outputs appropriate corrections. The input is the flagged post data and the prompt, and the output is the corrections. The AI ​​model performs data calculations based on this prompt, generating the corrections.

[0184] Step 6:

[0185] The server posts the generated correction information to the online environment. Specifically, it tweets the correction information from the official SNS account and, if necessary, sends a notification to users who retweeted the misinformation. The input is the generated correction information, and the output is the posting of the correction information on the SNS and a user notification. This posting and notification are specific actions that transmit data.

[0186] 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.

[0187] The present invention provides a system that prevents the spread of false information about company names and product names on social networking sites and also takes into consideration the feelings of users. Specific embodiments of the invention are described below.

[0188] System Configuration

[0189] The system consists of the following main components:

[0190] 1. Server:

[0191] Responsible for collecting, analyzing, collating data, identifying misinformation, and generating and distributing corrections.

[0192] Using an emotion engine, the content and context of users' posts are analyzed to identify emotions.

[0193] 2. Terminal:

[0194] A device that allows users to use social networking services and make public posts.

[0195] If incorrect information is detected, you will receive a notification of the correction.

[0196] 3. User:

[0197] People who post and share information on social media.

[0198] Program processing

[0199] 1. Collection of information

[0200] The server uses the social networking site's API to crawl public posts containing specific keywords (such as company or product names), and also collects metadata including the post content, user information, posting date and time, number of retweets, and number of likes.

[0201] Example: The server collects tweets containing the keyword "Company X" and obtains the post content, user information, posting date and time, number of retweets, and number of likes.

[0202] 2. Analysis and extraction of post content

[0203] The server runs a text analysis algorithm through the collected public posts to extract posts containing specific keywords, and in the process also uses an emotion engine to obtain sentiment data for the posts.

[0204] Example: A server collects 1,000 tweets, analyzes 100 posts containing "Company X," and the sentiment engine identifies that 50 of them have positive sentiment, 20 have negative sentiment, and 30 are neutral.

[0205] 3. Comparison with official information

[0206] The server compares the extracted social media posts with the company's official information, which is obtained using web scraping or RSS feeds and stored in a database. This comparison determines whether the posts match existing official information.

[0207] Example: A server compares a post that says "Company X's new product is being offered for free" with information on Company X's official website that says "The new product costs $100" and recognizes a contradiction.

[0208] 4. Identifying and flagging misinformation

[0209] Based on the results of the comparison, the server flags any posts that contain inconsistencies as misinformation and manages them in a database. The server also manages the sentiment information of the posts.

[0210] Example: A post that says the server will be "provided for free" is flagged as misinformation because it contradicts official information, and is managed along with "negative" sentiment.

[0211] 5. Generating Corrections

[0212] For posts that the server determines to be misinformation, the system automatically generates a correction based on official information. Using an emotion engine, the correction is adjusted to a tone that corresponds to the emotion, such as positive, negative, or neutral. The correction can also include a citation of the official information source.

[0213] Example: The server generates a correction message that reads, "Company X's new product will not be provided for free. It costs $100. Official information is here." If the user has expressed negative sentiment, the correction message will be written in a more polite tone.

[0214] 6. Distribution of Corrections

[0215] The server posts the generated correction information to social media, and notifies users who posted or retweeted the incorrect information as needed.

[0216] Example: The server posts a correction tweet through its official account saying, "Company X's new product will not be provided for free. It costs $100. Click here for details." It also notifies users who retweeted the misinformation of the correction.

[0217] Operational Management and Reporting

[0218] The server periodically generates reports on the collected data, sentiment data, and misinformation detection and correction results, and provides them to companies and relevant public institutions. The reports include a list of detected misinformation, sentiment data, the spread of correction information, and its impact.

[0219] Example: A server generates a monthly report and sends it to a company's public relations department, which includes a list of misinformation about company X, a sentiment analysis of posts identified as misinformation, and the spread of corrections.

[0220] This system quickly detects misinformation on social media and delivers corrections in an appropriate tone based on the sentiment, thereby protecting the credibility of companies and providing accurate information.

[0221] The processing flow will be explained below.

[0222] Step 1:

[0223] The server accesses the social networking service's API and crawls public posts containing specific keywords (company names or product names). It sets the API endpoint and request parameters and periodically retrieves data.

[0224] Example: The server crawls tweets containing the keyword "Company X" every hour and retrieves the post content, user information, posting date and time, number of retweets, and number of likes.

[0225] Step 2:

[0226] The server crawls and collects post data, then runs it through a text analysis algorithm to extract posts containing specific keywords. Keyword analysis is performed using natural language processing (NLP) technology.

[0227] Example: Analyze and extract 100 posts containing "Company X" from 1,000 tweets collected by the server.

[0228] Step 3:

[0229] The server then runs the extracted public posts through an emotion engine to obtain the emotion data of the post content, which identifies emotion categories such as positive, negative, and neutral.

[0230] Example: A server analyzes the sentiment of 100 posts about "Company X" and identifies 50 as positive, 30 as negative, and 20 as neutral.

[0231] Step 4:

[0232] The server uses web scraping and RSS feeds to obtain official company information, which is then stored in a database.

[0233] Example: A server scrapes new product information and pricing information from the official website of "Company X" and stores it in a database.

[0234] Step 5:

[0235] The server compares the extracted social media posts with official information, using natural language processing technology to determine whether the posts match the official information.

[0236] Example: A server compares a social media post that says "Company X's new product is being offered for free" with information on Company X's official website that says "The new product costs $100" and recognizes a contradiction.

[0237] Step 6:

[0238] Based on the results of the comparison, the server flags any inconsistent posts as misinformation and manages them in a database, along with the sentiment information of the posts.

[0239] Example: A post that says the server will be "provided for free" contradicts official information, so we flag it as misinformation and manage it together with "negative" sentiment information.

[0240] Step 7:

[0241] For posts that the server determines to be misinformation, the system automatically generates a correction based on official information. Using an emotion engine, the correction is adjusted to a tone that corresponds to the emotion of the post, and includes a citation of the official source.

[0242] Example: The server generates a correction that reads, "Company X's new product is not free. It costs $100. Official information is here." In this case, the correction is written in a more polite tone for posts that express negative sentiment.

[0243] Step 8:

[0244] The server posts the generated correction information to the social networking site, and if necessary, sends notifications to users who posted or retweeted the incorrect information.

[0245] Example: The server tweets a correction through its official account: "Company X's new product will not be free. It will cost $100." It also notifies users who retweeted the false information.

[0246] Step 9:

[0247] The server generates reports on the collected data, sentiment data, misinformation detection, and correction results, and provides them to companies and relevant public institutions.

[0248] Example: A server generates a monthly report and sends it to a company's public relations department, which includes a list of misinformation about company X, a sentiment analysis of posts identified as misinformation, and the spread of corrections.

[0249] In this way, this system quickly detects misinformation on social media and delivers corrections in an appropriate tone based on the sentiment, thereby protecting the credibility of companies and providing accurate information.

[0250] Example 2

[0251] 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."

[0252] The rapid and widespread spread of misinformation about company and product names on the internet, particularly on social media platforms, has become a problem. This misinformation not only has a negative impact on a company's credibility and brand image, but also conveys incorrect information to consumers. It is also difficult to correct the misinformation, and it is also difficult to contain the impact of the misinformation that has spread. Furthermore, there is a lack of technology to generate appropriate corrections that take user sentiment into account. A system is needed to solve this issue and provide accurate information while maintaining a company's credibility.

[0253] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting public posts, means for analyzing the collected public posts and extracting posts containing specific keywords, means for acquiring emotion data of the extracted posts, means for comparing the content of the extracted posts with official information, means for identifying false information based on the comparison result and flagging the false information, means for generating correction information in a tone corresponding to the emotion for posts determined to be false information, means for posting the generated correction information to an SNS, and means for notifying the generated correction information to users who posted the false information and users who retweeted the false information. This makes it possible to quickly detect false information and provide appropriate correction information while taking user emotions into consideration.

[0254] "Public posts" refer to information posted by users with public settings on social media, online forums, etc.

[0255] The "means of collection" refers to a part of the system that has the function of collecting public posts using social media APIs and web scraping technology.

[0256] The "means for analyzing" is a part of a system that has the function of analyzing collected public posts using text analysis algorithms and natural language processing techniques and extracting posts that contain specific keywords.

[0257] The "means for extracting" is a part of the system that has the function of selecting information that meets specific conditions from the analyzed data.

[0258] "Emotional data" is data that is classified based on an analysis of the content of posts to determine emotions such as positive, negative, or neutral.

[0259] "Official information" is reliable information officially released by a company or organization, such as from a website, press release, RSS feed, etc.

[0260] A "mechanism for matching" is a part of the system that has the ability to compare the collected content of public posts with official information and detect matches or inconsistencies.

[0261] "Misinformation" refers to public posts whose content contradicts official information.

[0262] "Flagging measures" are part of a system that has the ability to manage posts identified as misinformation by attaching specific marks or tags to them.

[0263] "Correction information" is content generated to provide correct information based on official information in response to posts identified as misinformation.

[0264] The "means for generating" is a part of the system that has the function of automatically generating correction information based on official information.

[0265] "Means of notification" refers to the part of the system that has the function of communicating the generated correction information to the user who posted the misinformation or retweeted it.

[0266] The present invention provides a system that prevents the spread of false information about company names and product names on social networking sites and also takes into consideration the feelings of users. A specific embodiment of the present invention will now be described.

[0267] System Configuration

[0268] The system consists of the following main components:

[0269] 1. Server:

[0270] Responsible for collecting, analyzing, collating data, identifying misinformation, and generating and distributing corrections.

[0271] Using an emotion engine, the content and context of users' posts are analyzed to identify emotions.

[0272] Specifically, we use Python natural language processing libraries (e.g., NLTK), sentiment analysis APIs, and web scraping libraries (e.g., Beautiful Soup).

[0273] 2. Terminal:

[0274] A device that users use to use SNS and post publicly. This includes smartphones and PCs.

[0275] If incorrect information is detected, you will receive a notification of the correction.

[0276] 3. User:

[0277] These are people who post and share information on social media.

[0278] Program processing

[0279] Collection of information

[0280] The server uses the social networking site's API to crawl public posts containing specific keywords (e.g., "Company X"), collecting metadata including the post content, user information, posting date and time, number of retweets, and number of likes.

[0281] Example: The server collects tweets containing the keyword "Company X" and obtains the post content, user information, posting date and time, number of retweets, and number of likes.

[0282] Analyzing and extracting post content

[0283] The server runs a text analysis algorithm through the collected public posts to extract posts containing specific keywords, and in the process also uses an emotion engine to obtain sentiment data for the posts.

[0284] Example: A server collects 1,000 tweets, analyzes 100 posts containing "Company X," and the sentiment engine identifies that 50 of them have positive sentiment, 20 have negative sentiment, and 30 are neutral.

[0285] Comparison with official information

[0286] The server compares the extracted social media posts with the company's official information, which is obtained using web scraping or RSS feeds and stored in a database. This comparison determines whether the posts match existing official information.

[0287] Example: A server compares a post that says "Company X's new product is being offered for free" with information on the company's official website that says "The new product costs $100" and recognizes a contradiction.

[0288] Identifying and flagging misinformation

[0289] Based on the results of the comparison, the server flags any posts that contain inconsistencies as misinformation and manages them in a database. The server also manages the sentiment information of the posts.

[0290] Example: A post that says the server will be "provided for free" is flagged as misinformation because it contradicts official information, and is managed along with "negative" sentiment.

[0291] Generate correction information

[0292] For posts that the server determines to be misinformation, the system automatically generates a correction based on official information. Using an emotion engine, the correction is adjusted to a tone that corresponds to the emotion, such as positive, negative, or neutral. The correction can also include a citation of the official information source.

[0293] Example: The server generates a correction message that reads, "Company X's new product will not be provided for free. It costs $100. Official information is here." If the user has expressed negative sentiment, the correction message will be written in a more polite tone.

[0294] Distribution of correction information

[0295] The server posts the generated correction information to social media, and notifies users who posted or retweeted the incorrect information as needed.

[0296] Example: The server posts a correction tweet through its official account saying, "Company X's new product will not be provided for free. It costs $100. Click here for details." It also notifies users who retweeted the misinformation of the correction.

[0297] Operational Management and Reporting

[0298] The server periodically generates reports on the collected data, sentiment data, and misinformation detection and correction results, and provides them to companies and relevant public institutions. The reports include a list of detected misinformation, sentiment data, the spread of correction information, and its impact.

[0299] Example: A server generates a monthly report and sends it to a company's public relations department, which includes a list of misinformation about company X, a sentiment analysis of posts identified as misinformation, and the spread of corrections.

[0300] Prompt Sentence Examples

[0301] After a user posts news on Twitter that "Company Y has released a new product," the information begins to spread. Later, it is discovered that the content of the post is inaccurate, and the system is used to generate a correction and repost the post in a tone that matches the user's sentiment.

[0302] This system quickly detects misinformation on social media and delivers corrections in an appropriate tone based on the sentiment, thereby protecting the credibility of companies and providing accurate information.

[0303] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0304] Step 1: Gather information

[0305] Specific behavior:

[0306] Input: The server uses a social networking site's API (e.g., Twitter API) to collect public posts containing specific keywords, such as "Company X."

[0307] Data processing / calculation: The server sends an API request to retrieve data including the post content, user information, post date and time, number of retweets, and number of likes.

[0308] Output: Collected public posts and metadata stored in a database.

[0309] Example: A server collects tweets about "Company X" through an API and stores them as JSON format data.

[0310] Step 2: Analyzing and extracting post content

[0311] Specific behavior:

[0312] Input: Public submission data and metadata collected by the server.

[0313] Data processing / calculation: The server uses a Python natural language processing library (e.g., NLTK) to filter the collected posts to those containing specific keywords, and then uses an emotion engine to analyze the emotion data of the posts.

[0314] Output: Posts containing specific keywords and their sentiment data stored in a database.

[0315] Example: From a collection of 1,000 tweets, the server extracts 100 posts containing "Company X," and the sentiment engine identifies 50 as positive, 20 as negative, and 30 as neutral.

[0316] Step 3: Check against official information

[0317] Specific behavior:

[0318] Input: Post data and sentiment data extracted by the server, as well as official company information.

[0319] Data processing / calculation: The server uses a web scraping library (e.g., Beautiful Soup) to retrieve the latest official information from the company's official website and store it in a database. The extracted post content is then compared with this official information.

[0320] Output: Save the match results to a database.

[0321] Example: A server retrieves information from an official website that says "The price of a new product is $100," compares it with a post that says "Company X's new product is being offered for free," and discovers a contradiction.

[0322] Step 4: Identifying and flagging misinformation

[0323] Specific behavior:

[0324] Input: The server checks the results against the official information.

[0325] Data processing / calculation: Based on the matching results, the server flags any inconsistent posts as misinformation and manages them in a database along with the emotion data.

[0326] Output: An updated database containing misinformation flags and sentiment data.

[0327] Example: A post that says "It's provided for free" contradicts official information, so the server flags it as misinformation and manages it along with "negative" sentiment.

[0328] Step 5: Generate corrections

[0329] Specific behavior:

[0330] Input: Posts flagged by the server as misinformation, sentiment data, and official information.

[0331] Data processing / calculation: The server uses a natural language generation model (e.g., GPT-3) to automatically generate corrections to misinformation. The tone of the corrections is adjusted according to the emotion determined by the emotion engine.

[0332] Output: Corrective sentences in tone according to the emotion.

[0333] Example: A server generates a correction such as "Company X's new product is not free. It costs $100. Official information here." and uses a more polite tone for users who express negative sentiment.

[0334] Step 6: Distributing corrections

[0335] Specific behavior:

[0336] Input: Server-generated corrections and notification list (users who posted and retweeted the misinformation).

[0337] Data processing / calculation: The server uses the SNS API to post the correction from the official account and sends individual notifications to users on the notification list.

[0338] Output: The correction posted on social media and the users who received the notification.

[0339] Example: Your server tweets a correction: "Company X's new product is not free. It costs $100. Learn more." and sends notifications to relevant users.

[0340] Step 7: Operational Management and Reporting

[0341] Specific behavior:

[0342] Input: The server collects data from the database, sentiment data, misinformation history, and correction results.

[0343] Data processing / calculation: The server periodically aggregates the data, compiles it into a report, and outputs it as a PDF or CSV file.

[0344] Output: Generated reports (e.g. monthly reports).

[0345] Example: A server generates a monthly report containing a list of misinformation about Company X, the results of a sentiment analysis of posts identified as misinformation, and the spread of corrections, and sends it to the company's public relations department.

[0346] (Application example 2)

[0347] 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."

[0348] In today's information society, social networking services (SNS) have become a major means of spreading information, but this often leads to the spread of false information about companies and products. This false information is likely to have a negative impact on the company's brand image and credibility. Furthermore, corrections that do not take into account the emotions contained in user posts may leave users with a negative impression. A system that solves these problems is needed.

[0349] 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.

[0350] In this invention, the server includes means for collecting public posts, means for analyzing the collected public posts to extract posts containing specific keywords and acquiring emotion data, means for comparing the content of the extracted posts with official company information, means for identifying and flagging false information based on the comparison results, means for generating correction information in a tone appropriate to the emotion of the post determined to be false, means for posting the generated correction information to the SNS, and means for managing the history of collected data, emotion data, and correction information. This makes it possible to quickly detect false information about companies and products on the SNS and distribute correction information in an appropriate tone appropriate to the emotion.

[0351] "Public post" refers to a post that is publicly available for all users to view on a social networking service.

[0352] "Means of collection" refers to the function of using the API of a social networking service to obtain posts containing specific keywords.

[0353] "Means for analyzing and extracting" refers to a function for analyzing collected public posts using a text analysis algorithm and extracting posts that contain specific keywords.

[0354] "Means for obtaining emotion data" refers to a function that uses an emotion engine to identify emotions such as positive, negative, and neutral from the text of public posts and obtain that data.

[0355] "Means of matching" refers to the function for comparing the content of extracted posts with official company information and verifying whether they match.

[0356] "Means to identify and flag misinformation" refers to a function that identifies a post as misinformation when the content of the post contradicts official information and flags it in the database for management.

[0357] "Means for generating correction information" refers to a function that uses a generative AI model to automatically generate correction statements in an appropriate tone based on the emotion of posts that are determined to contain false information.

[0358] "Means of posting to social media" refers to the function for notifying the generated correction to users who posted or retweeted the false information on social media.

[0359] "Means for managing the history of data, emotional data, and correction information" refers to the function for systematically storing and managing collected data, emotional data, and the history of corrections to misinformation, and outputting it in report format as needed.

[0360] An "emotion engine" refers to algorithms or software that use natural language processing technology to analyze and identify emotions from text data.

[0361] A "generative AI model" refers to an artificial intelligence model that can learn from large amounts of text data and generate new text.

[0362] The system embodying the present invention includes a series of process flows for collecting, analyzing, and correcting public posts. A specific embodiment of the system is described below.

[0363] System Configuration

[0364] This system consists of the following main components:

[0365] server:

[0366] The server uses the social media platform's API to crawl public posts, analyzes them using a text analysis algorithm, and extracts posts containing specific keywords. It then uses an emotion engine to obtain the post's emotional data. Once analyzed, the post is compared with the company's official information to identify misinformation. A generative AI model then generates corrections in a tone appropriate to the emotion and posts them on social media. The data, emotional data, and correction history are stored in a database.

[0367] The specific hardware and software used are as follows:

[0368] SNS API: Twitter API, Facebook Graph API, etc.

[0369] Natural language processing libraries: NLTK, SpaCy

[0370] Sentiment analysis engine: VADER, TextBlob

[0371] Web scraping tool: Beautiful Soup

[0372] RSS Reader:RSS Feed Reader

[0373] Generative AI models: GPT-3, BERT

[0374] Data management tools: SQL databases, NoSQL databases

[0375] Examples include:

[0376] The server collects public tweets mentioning "specific brand X" via the Twitter API, analyzes the posts using the NLTK library, and extracts sentiment data using the VADER engine.

[0377] A post containing the misinformation "Brand X is offered for free" is compared with information on the company's official website and recognized as misinformation. A generative AI model (e.g., GPT-3) is then used to generate a correction, "Brand X's product is not free. It costs $100," and this correction is posted on social media.

[0378] Device:

[0379] The terminal refers to the device of the user who uses SNS and posts publicly. If the user posts incorrect information, they will receive a notification of the correction.

[0380] Examples include:

[0381] When a user tweets "Brand X is being offered for free" on a social networking site, the server collects and analyzes the tweet, and then generates a correction that is sent to the user's account.

[0382] User:

[0383] A user refers to an individual who posts and shares information on a social networking site.

[0384] For example, if a user spreads false information that "Brand X is being offered for free," the user will receive a correction that reads, "Brand X products are not free. They cost $100."

[0385] Prompt Sentence Examples

[0386] Here are some example prompts to input to a generative AI model:

[0387] markdown

[0388] To generate sentiment-based corrections for users, we provide the following information:

[0389] User Information

[0390] Username: User A

[0391] Emotion: Negative

[0392] Misinformation

[0393] Post content: New product from brand X is being offered for free

[0394] It's official: Brand X's new product is $100

[0395] Use the information above to generate a correction in the right tone. For example, "Brand X's new product is not free. It costs $100. Learn more."

[0396] In this way, it is possible to quickly detect misinformation about companies and products on social media and deliver corrections in an appropriate tone based on the sentiment.

[0397] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0398] Step 1:

[0399] The server uses the SNS API to collect public posts containing specific keywords (company names or product names). The input is the keyword and the SNS API token, and the public posts and metadata (post content, user information, posting date and time, number of retweets, number of likes) are obtained by API calls. The output is the collected post data.

[0400] Step 2:

[0401] The server analyzes the collected public posts using a text analysis algorithm (NLTK or SpaCy) to extract posts containing specific keywords. The input is the collected post data, and the text analysis algorithm extracts posts containing keywords. The output is the extracted post data.

[0402] Step 3:

[0403] The server analyzes the extracted posts using an emotion engine (VADER or TextBlob) to obtain emotion data. The input is the extracted post data, and the emotion engine obtains emotion data such as positive, negative, or neutral. The output is the post data with the emotion data added.

[0404] Step 4:

[0405] The server compares the post content with the emotional data added to it with the company's official information. The official information is obtained from web scraping or RSS readers. The input is the post data with the emotional data added and the official information data, and the matching algorithm detects matches or inconsistencies. The output is the matching result.

[0406] Step 5:

[0407] The server flags posts that contain inconsistencies based on the matching results. The input is the matching results, and if there are any inconsistencies, a misinformation flag is added. The output is the post data with the misinformation flag added.

[0408] Step 6:

[0409] Based on posts that the server determines to be misinformation, it uses a generative AI model (such as GPT-3 or BERT) to generate corrections in a tone appropriate to the emotion. The input is post data flagged as misinformation and official information data, and the generative AI model generates corrections in an appropriate tone. The output is the corrections.

[0410] Step 7:

[0411] The server posts the generated correction to SNS and notifies the users who posted the misinformation and those who retweeted it. The input is the correction and the target user information, and the correction is posted via the SNS API. The output is a notification that the post has been completed.

[0412] Step 8:

[0413] The server periodically generates a report containing the collected data, sentiment data, and correction history information, and provides it to the company and the relevant advertising agency. The input is all the data retrieved from the database, and the report is generated by the data management tool. The output is a report document.

[0414] 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.

[0415] 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.

[0416] 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.

[0417] [Second embodiment]

[0418] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0419] 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.

[0420] 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).

[0421] 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.

[0422] 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.

[0423] 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).

[0424] 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.

[0425] 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.

[0426] 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.

[0427] 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.

[0428] 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.

[0429] 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."

[0430] The present invention provides a system for preventing the spread of false information about company and product names on social media, and provides a process for automatically collecting, analyzing, and collating public posts and correcting the false information. Specific embodiments of the system are described below.

[0431] System Configuration

[0432] The system consists of the following main components:

[0433] 1. Server: A central location responsible for collecting, analyzing, collating data, identifying misinformation, and generating and distributing corrections.

[0434] 2. Device: A device on which a user uses SNS and posts publicly. If incorrect information is detected, you may receive a correction notice.

[0435] 3. Users: People who post and share information on social media.

[0436] Program processing

[0437] 1. Collection of information

[0438] The server periodically uses the social networking service's API to crawl public posts containing specific keywords, such as company names or product names.

[0439] Example: The server uses the keyword "Company X" to collect relevant tweets from Twitter's API. At the same time, metadata such as the post content, user information, posting date and time, number of retweets, and number of likes are also obtained.

[0440] 2. Analysis and extraction of post content

[0441] The server then runs the collected public posts through a text analysis algorithm, using natural language processing (NLP) techniques, to extract posts containing specific keywords.

[0442] Example: Analyze 1,000 public posts collected by the server to find posts containing the specified keyword "Company X" and extract 100 posts.

[0443] 3. Comparison with official information

[0444] The server then compares the content of the extracted public posts with official company information, often obtained through web scraping or RSS feeds, to determine whether the post matches existing official information.

[0445] Example: A server compares a post that says "Company X's new product is being offered for free" with information on the official website that says "The new product costs $100" and recognizes a discrepancy.

[0446] 4. Identifying and flagging misinformation

[0447] The server will then flag any posts that are inconsistent based on the results of the match, and posts identified as misinformation will be moderated within the system.

[0448] Example: A post containing the misinformation that servers are "provided for free" is flagged and added to a misinformation list.

[0449] 5. Generating Corrections

[0450] The server automatically generates corrections based on official information to correct misinformation, including the correct information and the source.

[0451] Example: The server creates a correction that reads, "Company X's new product costs $100. More information here."

[0452] 6. Distribution of Corrections

[0453] The server posts the generated correction information to the social networking site, and if necessary, notifies the user who posted the incorrect information and the user who retweeted it.

[0454] Example: In response to a false claim that "Company X's new product will be provided for free," the server tweets a correction from its official account and also sends a notification to users who have retweeted the false information.

[0455] Operational Management and Reporting

[0456] The server periodically generates reports on the collected data and the results of detecting and correcting misinformation, and provides them to companies and relevant public institutions. The reports include a list of detected misinformation, the spread of the correction information, and its impact.

[0457] Example: A server generates a monthly report and sends it to a company's marketing department, detailing the list of misinformation detected and the extent to which corrections to those misinformation have been disseminated.

[0458] In this way, this system efficiently detects misinformation on social media and quickly distributes corrections, thereby protecting the credibility of companies and providing accurate information.

[0459] The processing flow will be explained below.

[0460] Step 1:

[0461] The server accesses the social networking service's API and crawls public posts containing specific keywords (company names or product names). It sets the API endpoint and request parameters and periodically retrieves data.

[0462] Example: A server periodically crawls tweets containing the keyword "Company X" and retrieves metadata including the post content, user information, posting date and time, number of retweets, and number of likes.

[0463] Step 2:

[0464] The server crawls and retrieves the posted data, which is then passed through a text analysis algorithm. Natural language processing (NLP) techniques are used to analyze and extract posts containing specific keywords.

[0465] Example: The server analyzes 1,000 tweets collected and finds posts containing "Company X" and extracts 100 of them.

[0466] Step 3:

[0467] The server uses web scraping or RSS feeds to obtain official company information, and stores the information in a database.

[0468] Example: A server web scrapes new product information and product pricing information from the official website of "Company X" and stores it in a database.

[0469] Step 4:

[0470] The server compares the extracted social media posts with official information, using natural language processing technology to determine whether the posts match the official information.

[0471] Example: A server analyzes a social media post that says, "Company X's new product is being offered for free," and compares it with information on the official website that says, "The price of the new product is $100."

[0472] Step 5:

[0473] If the server does not match, it will flag the post as misinformation and keep it in a database.

[0474] Example: A post claiming that a server is "provided for free" is detected as contradicting official information, and the information is flagged as misinformation and added to a moderation list.

[0475] Step 6:

[0476] The server automatically generates a correction based on official information to correct misinformation. The correction includes the source of the official information and accurate information.

[0477] Example: The server automatically generates a correction: "Company X's new product costs $100. Official information is here."

[0478] Step 7:

[0479] The server generates a correction and posts it to the social networking site. If necessary, notifications are also sent to the user who posted the incorrect information and to users who retweeted it.

[0480] Example: In response to a false claim that "Company X's new product will be provided for free," the server tweets a correction from its official account and also sends a notification to users who have retweeted the false information.

[0481] Step 8:

[0482] The server generates a report of the collected data, the history of correction information, and the correction results, and provides it to companies and relevant public institutions.

[0483] Example: A server generates a monthly report and sends it to a company's marketing department, detailing the list of misinformation detected and the extent to which corrections to those misinformation have been disseminated.

[0484] In this way, this system efficiently detects misinformation on social media and quickly distributes corrections, thereby protecting the credibility of companies and providing accurate information.

[0485] Example 1

[0486] 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."

[0487] Currently, inaccurate and false information can spread quickly on online social networking services (SNS), causing problems that undermine the credibility of companies and products. Furthermore, delays in responding to false information can lead to further confusion, which could damage brand image and cause consumer confusion.

[0488] 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.

[0489] In this invention, the server includes means for collecting public posts, means for applying a text analysis algorithm to the collected public posts to extract posts containing specific keywords, means for comparing the content of the extracted posts with official information, means for identifying and flagging misinformation based on the comparison results, means for generating corrections containing correct information for posts determined to be misinformation, means for posting the generated corrections to a social networking service, and means for sending notifications to users who posted and retweeted the misinformation as needed, thereby enabling inaccurate and misinformation to be detected quickly and efficiently and corresponding corrections to be provided.

[0490] "Public Post" means a user's post that is publicly available on a social networking service.

[0491] "Means of collection" refers to algorithms and programs that automatically obtain public posts using APIs of social networking services, etc.

[0492] A "text analysis algorithm" refers to a program that uses natural language processing technology to extract specific keywords and meanings from text data.

[0493] "Means of extraction" refers to the process of using a text analysis algorithm to select data that meets certain conditions.

[0494] "Verification measures" refer to programs that compare the collected public posts with official information to check for matches or inconsistencies.

[0495] "Misinformation" refers to information that does not match official information or contains factually incorrect information.

[0496] "Flagging" refers to the process of assigning an identifying mark to information that meets certain conditions.

[0497] "Correction information" refers to content that corrects misinformation and includes correct information and sources of official information.

[0498] "Means for generating" refers to a program for automatically creating correct correction information for erroneous information.

[0499] "Means of posting" refers to the algorithms or programs used to disseminate the generated correction information on social networking services.

[0500] "Means for sending notifications" refers to a program for sending corrections to users who posted false information or spread it.

[0501] "Social networking service" refers to a service that allows users to share information with each other over the Internet.

[0502] "Report format" refers to a document format that visually presents the history of collected data and correction information in an easy-to-understand manner.

[0503] The present invention provides a system for preventing the spread of false information about company and product names on social networking services (SNS), and provides a process for automatically collecting, analyzing, and collating public posts and correcting the false information. A specific embodiment of this system will be described.

[0504] System Configuration

[0505] The system consists of the following main components:

[0506] 1. Server: A central location responsible for collecting, analyzing, collating data, identifying misinformation, and generating and distributing corrections.

[0507] 2. Device: A device on which a user uses SNS and posts publicly. If incorrect information is detected, you may receive a correction notice.

[0508] 3. Users: People who post and share information on social media.

[0509] Hardware and Software Use

[0510] The server should be a computer with powerful computing capabilities. Recommended hardware includes a fast CPU, large memory capacity, and an SSD. Software used includes Python, NLTK, spaCy, BeautifulSoup, and the Twitter API.

[0511] For example, a server periodically uses the Twitter API to collect tweets containing the keyword "Company X." It then uses a Python script to perform text analysis using NLTK and spaCy to extract posts containing specific keywords. The server then uses BeautifulSoup to scrape the latest information from the company's official website and compares the extracted posts with official information.

[0512] Generation and distribution of corrections

[0513] If a contradiction is detected, the server automatically generates a correction based on the official information. The correction includes the correct information and its source. The server then reposts the correction to the social networking site, and notifies the user who posted the misinformation and those who retweeted it. This quickly prevents the spread of misinformation.

[0514] Operational Management and Reporting

[0515] The server periodically generates reports on the collected data, misinformation detection, and correction results, and provides them to companies and relevant public institutions. The reports include a list of detected misinformation, the spread of correction information, and its impact. For example, the server generates a monthly report, converts it to PDF format, and emails it to the company's marketing department.

[0516] Prompt Sentence Examples

[0517] Please describe a system that detects misinformation about Company X on social media and automatically generates and distributes corrections. Please provide details, including examples, of how this system works.

[0518] In this way, this system efficiently detects misinformation on social media and quickly distributes corrections, thereby protecting the credibility of companies and providing accurate information.

[0519] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0520] Step 1:

[0521] The server collects public posts. The server periodically crawls public posts containing specific keywords, such as "Company X," using the API of the social networking site. Specifically, the server uses Python to access the Twitter API and collect tweets containing the keyword "Company X." The data obtained from the API of the social networking site includes metadata such as the post content, user information, posting date and time, number of retweets, and number of likes. This provides the public posts to be collected as input data.

[0522] Step 2:

[0523] The server runs the collected public posts through a text analysis algorithm to extract posts containing specific keywords. Specifically, the server uses natural language processing libraries such as NLTK and spaCy to analyze collected tweets for posts containing the keyword "Company X" and extract the relevant posts. At this stage, the collected post data is input and posts that match the keywords are extracted.

[0524] Step 3:

[0525] The server compares the extracted public post content with official information. Specifically, the server uses a web scraping tool such as BeautifulSoup to retrieve the latest product information and news from the company's official website. It then compares the extracted post content with the retrieved official information to check for matches and inconsistencies. The input data is the extracted post and official information, and the comparison results are output based on this.

[0526] Step 4:

[0527] The server identifies and flags misinformation based on the matching results. Specifically, the server flags posts containing the misinformation "provided for free" as misinformation and stores it in a database. The input data is the matching results, and based on the results, misinformation is identified and flagged data is output.

[0528] Step 5:

[0529] The server generates correction information based on official information. Specifically, the server automatically generates a correction statement that includes the correct information and its source. For example, it generates a correction statement such as, "Company X's new product costs $100. Click here for more information." The input data is official information and incorrect information, and the output is correction information.

[0530] Step 6:

[0531] The server posts the generated correction information to SNS, and if necessary, sends notifications to the user who posted the misinformation and to users who retweeted it.Specifically, the server tweets the correction information on its official account and also sends individual notifications to users who retweeted the misinformation.The input data is the generated correction information, and the output is posted to SNS, and notifications are sent.

[0532] Step 7:

[0533] The server generates and provides the collected data, correction information history, and correction results in report format.Specifically, the server creates a monthly report, converts it to PDF format, and emails it to the company's marketing department.The input data are the collected data and correction results, and the report is generated and provided as output.

[0534] (Application example 1)

[0535] 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."

[0536] The spread of misinformation on social media is a serious problem that damages the trust of companies and individuals. In particular, with the rapid increase in the number of social media users, misinformation spreads quickly, creating a need for systems that provide fast and accurate corrections. There is also a growing need for automated solutions that can save users the trouble of monitoring and correcting relevant misinformation themselves.

[0537] 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.

[0538] In this invention, the server includes means for collecting public posts, means for analyzing the collected public posts to extract posts containing specific keywords, means for comparing the content of the extracted posts with official information, means for identifying misinformation based on the comparison results and flagging the misinformation, means for generating corrections to posts determined to be misinformation, means for posting the generated corrections to an online environment, means for setting specific keywords and notifying users when misinformation is detected, means for automatically generating corrections using a generative AI model, and means for setting prompt text for the generative AI model. This enables quick detection and correction of misinformation spreading on SNS and quick notification to users.

[0539] A "means for collecting public posts" is a device or program that has the functionality to automatically collect posts that are publicly available on social media or online platforms.

[0540] "Means for analyzing collected public posts and extracting posts containing specific keywords" refers to a device or program that has the function of analyzing collected post data and identifying and extracting posts containing specified keywords.

[0541] "Means for comparing the content of extracted posts with official information" refers to a device or program that has the function of comparing and matching the content of extracted posts with official information provided by companies or official institutions.

[0542] "Means for identifying false information based on the matching results and flagging false information" refers to a device or program that has the function of detecting false information from the matching results and flagging posts that are determined to be false information.

[0543] "Means for generating correction information for posts determined to contain false information" refers to a device or program that has the function of automatically generating a correction statement to provide accurate information for posts identified as false information.

[0544] "Means for posting the generated correction information online" refers to a device or program that has the function of automatically posting the generated correction information to social media or other online platforms.

[0545] "Means for setting specific keywords and notifying users when misinformation is detected" refers to a device or program that has the function of allowing users to set keywords of interest, collecting and analyzing information based on those keywords, and notifying users when misinformation is detected.

[0546] "Means for automatically generating correction information using a generative AI model" refers to a device or program that has the function of automatically generating correction information for erroneous information using an artificial intelligence model.

[0547] "Means for setting a prompt sentence to a generative AI model" refers to a device or program that has the function of setting a prompt sentence to be input to a generative AI model and performing information processing based on that prompt.

[0548] The present invention is a system that effectively detects misinformation spread on social media and quickly distributes corrections. The system includes the following main components:

[0549] System Configuration

[0550] 1. Server:

[0551] The server has the ability to collect public posts using the API of a social networking site. For example, it uses the Twython library to collect posts from Twitter that contain specific keywords.

[0552] The server uses the SpaCy library to process the collected data in natural language, analyzing and extracting posts containing specific keywords.

[0553] The server scrapes official information using BeautifulSoup and the requests library to match the extracted public posts with official information obtained from the websites of companies or official institutions.

[0554] Based on the results of the comparison, the server flags any posts that contradict official information as misinformation.

[0555] 2. Use of generative AI models:

[0556] For posts that are determined to contain false information, the server automatically generates corrections using a generative AI model. The generative AI model uses the latest natural language processing technology and generates corrections by setting prompt sentences.

[0557] For example, a prompt such as "We have detected a tweet that says, 'Company X's new product will be offered for free.' Please correct this information." can be input into the generative AI model to obtain appropriate correction information.

[0558] 3. User Notice:

[0559] If certain keywords are set, the server will notify users when misinformation is detected, allowing users to respond quickly.

[0560] 4. Distribution of Corrections:

[0561] The server has the function of posting the generated correction information online, for example, by posting the correction information from an official account on social media to spread it.

[0562] Specific examples

[0563] As a specific example, consider a case where misinformation related to Company X is spread on social media. In this case, the server collects related posts from Twitter using the keyword "Company X." By analyzing the posts, a tweet containing the misinformation "provided for free" is detected. The server compares it with official information, flags it as misinformation, and then uses a generative AI model to automatically generate a correction such as "Company X's new product costs $100. Click here for more information." The server then posts the generated correction on social media and sends a notification to the affected users.

[0564] This system enables the rapid detection and correction of misinformation, protecting the credibility of businesses and individuals. Furthermore, by utilizing generative AI models and prompts, corrections can be automatically generated, enabling efficient correction of information. Specific examples of prompts are as follows:

[0565] Example prompt sentence:

[0566] "We have detected a tweet that states 'Company X's new product will be offered for free.' Please correct this information."

[0567] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0568] Step 1:

[0569] The server uses the SNS API to collect public posts containing specific keywords. Specifically, it uses the Twython library to set keywords such as "Company X" and collects tweets that match the keywords. The input is the specific keyword, and the output is the collected post data. This collected data includes the post content, poster information, posting date and time, number of retweets, number of likes, etc.

[0570] Step 2:

[0571] The server analyzes the collected public posts using natural language processing (NLP) technology and extracts posts containing specific keywords. Specifically, it performs text analysis using the SpaCy library to extract posts containing the keyword "Company X." The input is the collected post data, and the output is the analyzed and extracted post data. The analysis includes data processing such as part-of-speech tagging and named entity extraction.

[0572] Step 3:

[0573] The server compares the post content extracted through analysis with official information. Specifically, it uses BeautifulSoup and the requests library to scrape official information from the company's official website and compares it with the extracted post content. The input is the extracted post data and official information, and the output is the comparison result. This comparison involves data calculations to determine whether the post content matches the official information.

[0574] Step 4:

[0575] Based on the matching results, the server identifies misinformation and flags the post. For example, if a post saying "Company X's new product will be offered for free" contradicts official information, it will be flagged as misinformation. The input is the matching results, and the output is the flagged post data. This flagging is a data processing process to identify misinformation.

[0576] Step 5:

[0577] The server uses a generative AI model to automatically generate corrections for posts determined to contain false information. Specifically, the prompt "We have detected a tweet stating that 'Company X's new product will be offered for free.' Please correct this information." is input into the generative AI model, which then outputs appropriate corrections. The input is the flagged post data and the prompt, and the output is the corrections. The AI ​​model performs data calculations based on this prompt, generating the corrections.

[0578] Step 6:

[0579] The server posts the generated correction information to the online environment. Specifically, it tweets the correction information from the official SNS account and, if necessary, sends a notification to users who retweeted the misinformation. The input is the generated correction information, and the output is the posting of the correction information on the SNS and a user notification. This posting and notification are specific actions that transmit data.

[0580] 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.

[0581] The present invention provides a system that prevents the spread of false information about company names and product names on social networking sites and also takes into consideration the feelings of users. Specific embodiments of the invention are described below.

[0582] System Configuration

[0583] The system consists of the following main components:

[0584] 1. Server:

[0585] Responsible for collecting, analyzing, collating data, identifying misinformation, and generating and distributing corrections.

[0586] Using an emotion engine, the content and context of users' posts are analyzed to identify emotions.

[0587] 2. Terminal:

[0588] A device that allows users to use social networking services and make public posts.

[0589] If incorrect information is detected, you will receive a notification of the correction.

[0590] 3. User:

[0591] People who post and share information on social media.

[0592] Program processing

[0593] 1. Collection of information

[0594] The server uses the social networking site's API to crawl public posts containing specific keywords (such as company or product names), and also collects metadata including the post content, user information, posting date and time, number of retweets, and number of likes.

[0595] Example: The server collects tweets containing the keyword "Company X" and obtains the post content, user information, posting date and time, number of retweets, and number of likes.

[0596] 2. Analysis and extraction of post content

[0597] The server runs a text analysis algorithm through the collected public posts to extract posts containing specific keywords, and in the process also uses an emotion engine to obtain sentiment data for the posts.

[0598] Example: A server collects 1,000 tweets, analyzes 100 posts containing "Company X," and the sentiment engine identifies that 50 of them have positive sentiment, 20 have negative sentiment, and 30 are neutral.

[0599] 3. Comparison with official information

[0600] The server compares the extracted social media posts with the company's official information, which is obtained using web scraping or RSS feeds and stored in a database. This comparison determines whether the posts match existing official information.

[0601] Example: A server compares a post that says "Company X's new product is being offered for free" with information on Company X's official website that says "The new product costs $100" and recognizes a contradiction.

[0602] 4. Identifying and flagging misinformation

[0603] Based on the results of the comparison, the server flags any posts that contain inconsistencies as misinformation and manages them in a database. The server also manages the sentiment information of the posts.

[0604] Example: A post that says the server will be "provided for free" is flagged as misinformation because it contradicts official information, and is managed along with "negative" sentiment.

[0605] 5. Generating Corrections

[0606] For posts that the server determines to be misinformation, the system automatically generates a correction based on official information. Using an emotion engine, the correction is adjusted to a tone that corresponds to the emotion, such as positive, negative, or neutral. The correction can also include a citation of the official information source.

[0607] Example: The server generates a correction message that reads, "Company X's new product will not be provided for free. It costs $100. Official information is here." If the user has expressed negative sentiment, the correction message will be written in a more polite tone.

[0608] 6. Distribution of Corrections

[0609] The server posts the generated correction information to social media, and notifies users who posted or retweeted the incorrect information as needed.

[0610] Example: The server posts a correction tweet through its official account saying, "Company X's new product will not be provided for free. It costs $100. Click here for details." It also notifies users who retweeted the misinformation of the correction.

[0611] Operational Management and Reporting

[0612] The server periodically generates reports on the collected data, sentiment data, and misinformation detection and correction results, and provides them to companies and relevant public institutions. The reports include a list of detected misinformation, sentiment data, the spread of correction information, and its impact.

[0613] Example: A server generates a monthly report and sends it to a company's public relations department, which includes a list of misinformation about company X, a sentiment analysis of posts identified as misinformation, and the spread of corrections.

[0614] This system quickly detects misinformation on social media and delivers corrections in an appropriate tone based on the sentiment, thereby protecting the credibility of companies and providing accurate information.

[0615] The processing flow will be explained below.

[0616] Step 1:

[0617] The server accesses the social networking service's API and crawls public posts containing specific keywords (company names or product names). It sets the API endpoint and request parameters and periodically retrieves data.

[0618] Example: The server crawls tweets containing the keyword "Company X" every hour and retrieves the post content, user information, posting date and time, number of retweets, and number of likes.

[0619] Step 2:

[0620] The server crawls and collects post data, then runs it through a text analysis algorithm to extract posts containing specific keywords. Keyword analysis is performed using natural language processing (NLP) technology.

[0621] Example: Analyze and extract 100 posts containing "Company X" from 1,000 tweets collected by the server.

[0622] Step 3:

[0623] The server then runs the extracted public posts through an emotion engine to obtain the emotion data of the post content, which identifies emotion categories such as positive, negative, and neutral.

[0624] Example: A server analyzes the sentiment of 100 posts about "Company X" and identifies 50 as positive, 30 as negative, and 20 as neutral.

[0625] Step 4:

[0626] The server uses web scraping and RSS feeds to obtain official company information, which is then stored in a database.

[0627] Example: A server scrapes new product information and pricing information from the official website of "Company X" and stores it in a database.

[0628] Step 5:

[0629] The server compares the extracted social media posts with official information, using natural language processing technology to determine whether the posts match the official information.

[0630] Example: A server compares a social media post that says "Company X's new product is being offered for free" with information on Company X's official website that says "The new product costs $100" and recognizes a contradiction.

[0631] Step 6:

[0632] Based on the results of the comparison, the server flags any inconsistent posts as misinformation and manages them in a database, along with the sentiment information of the posts.

[0633] Example: A post that says the server will be "provided for free" contradicts official information, so we flag it as misinformation and manage it together with "negative" sentiment information.

[0634] Step 7:

[0635] For posts that the server determines to be misinformation, the system automatically generates a correction based on official information. Using an emotion engine, the correction is adjusted to a tone that corresponds to the emotion of the post, and includes a citation of the official source.

[0636] Example: The server generates a correction that reads, "Company X's new product is not free. It costs $100. Official information is here." In this case, the correction is written in a more polite tone for posts that express negative sentiment.

[0637] Step 8:

[0638] The server posts the generated correction information to the social networking site, and if necessary, sends notifications to users who posted or retweeted the incorrect information.

[0639] Example: The server tweets a correction through its official account: "Company X's new product will not be free. It will cost $100." It also notifies users who retweeted the false information.

[0640] Step 9:

[0641] The server generates reports on the collected data, sentiment data, misinformation detection, and correction results, and provides them to companies and relevant public institutions.

[0642] Example: A server generates a monthly report and sends it to a company's public relations department, which includes a list of misinformation about company X, a sentiment analysis of posts identified as misinformation, and the spread of corrections.

[0643] In this way, this system quickly detects misinformation on social media and delivers corrections in an appropriate tone based on the sentiment, thereby protecting the credibility of companies and providing accurate information.

[0644] Example 2

[0645] 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."

[0646] The rapid and widespread spread of misinformation about company and product names on the internet, particularly on social media platforms, has become a problem. This misinformation not only has a negative impact on a company's credibility and brand image, but also conveys incorrect information to consumers. It is also difficult to correct the misinformation, and it is also difficult to contain the impact of the misinformation that has spread. Furthermore, there is a lack of technology to generate appropriate corrections that take user sentiment into account. A system is needed to solve this issue and provide accurate information while maintaining a company's credibility.

[0647] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting public posts, means for analyzing the collected public posts and extracting posts containing specific keywords, means for acquiring emotion data of the extracted posts, means for comparing the content of the extracted posts with official information, means for identifying false information based on the comparison result and flagging the false information, means for generating correction information in a tone corresponding to the emotion for posts determined to be false information, means for posting the generated correction information to an SNS, and means for notifying the generated correction information to users who posted the false information and users who retweeted the false information. This makes it possible to quickly detect false information and provide appropriate correction information while taking user emotions into consideration.

[0648] "Public posts" refer to information posted by users with public settings on social media, online forums, etc.

[0649] The "means of collection" refers to a part of the system that has the function of collecting public posts using social media APIs and web scraping technology.

[0650] The "means for analyzing" is a part of a system that has the function of analyzing collected public posts using text analysis algorithms and natural language processing techniques and extracting posts that contain specific keywords.

[0651] The "means for extracting" is a part of the system that has the function of selecting information that meets specific conditions from the analyzed data.

[0652] "Emotional data" is data that is classified based on an analysis of the content of posts to determine emotions such as positive, negative, or neutral.

[0653] "Official information" is reliable information officially released by a company or organization, such as from a website, press release, RSS feed, etc.

[0654] A "mechanism for matching" is a part of the system that has the ability to compare the collected content of public posts with official information and detect matches or inconsistencies.

[0655] "Misinformation" refers to public posts whose content contradicts official information.

[0656] "Flagging measures" are part of a system that has the ability to manage posts identified as misinformation by attaching specific marks or tags to them.

[0657] "Correction information" is content generated to provide correct information based on official information in response to posts identified as misinformation.

[0658] The "means for generating" is a part of the system that has the function of automatically generating correction information based on official information.

[0659] "Means of notification" refers to the part of the system that has the function of communicating the generated correction information to the user who posted the misinformation or retweeted it.

[0660] The present invention provides a system that prevents the spread of false information about company names and product names on social networking sites and also takes into consideration the feelings of users. A specific embodiment of the present invention will now be described.

[0661] System Configuration

[0662] The system consists of the following main components:

[0663] 1. Server:

[0664] Responsible for collecting, analyzing, collating data, identifying misinformation, and generating and distributing corrections.

[0665] Using an emotion engine, the content and context of users' posts are analyzed to identify emotions.

[0666] Specifically, we use Python natural language processing libraries (e.g., NLTK), sentiment analysis APIs, and web scraping libraries (e.g., Beautiful Soup).

[0667] 2. Terminal:

[0668] A device that users use to use SNS and post publicly. This includes smartphones and PCs.

[0669] If incorrect information is detected, you will receive a notification of the correction.

[0670] 3. User:

[0671] These are people who post and share information on social media.

[0672] Program processing

[0673] Collection of information

[0674] The server uses the social networking site's API to crawl public posts containing specific keywords (e.g., "Company X"), collecting metadata including the post content, user information, posting date and time, number of retweets, and number of likes.

[0675] Example: The server collects tweets containing the keyword "Company X" and obtains the post content, user information, posting date and time, number of retweets, and number of likes.

[0676] Analyzing and extracting post content

[0677] The server runs a text analysis algorithm through the collected public posts to extract posts containing specific keywords, and in the process also uses an emotion engine to obtain sentiment data for the posts.

[0678] Example: A server collects 1,000 tweets, analyzes 100 posts containing "Company X," and the sentiment engine identifies that 50 of them have positive sentiment, 20 have negative sentiment, and 30 are neutral.

[0679] Comparison with official information

[0680] The server compares the extracted social media posts with the company's official information, which is obtained using web scraping or RSS feeds and stored in a database. This comparison determines whether the posts match existing official information.

[0681] Example: A server compares a post that says "Company X's new product is being offered for free" with information on the company's official website that says "The new product costs $100" and recognizes a contradiction.

[0682] Identifying and flagging misinformation

[0683] Based on the results of the comparison, the server flags any posts that contain inconsistencies as misinformation and manages them in a database. The server also manages the sentiment information of the posts.

[0684] Example: A post that says the server will be "provided for free" is flagged as misinformation because it contradicts official information, and is managed along with "negative" sentiment.

[0685] Generate correction information

[0686] For posts that the server determines to be misinformation, the system automatically generates a correction based on official information. Using an emotion engine, the correction is adjusted to a tone that corresponds to the emotion, such as positive, negative, or neutral. The correction can also include a citation of the official information source.

[0687] Example: The server generates a correction message that reads, "Company X's new product will not be provided for free. It costs $100. Official information is here." If the user has expressed negative sentiment, the correction message will be written in a more polite tone.

[0688] Distribution of correction information

[0689] The server posts the generated correction information to social media, and notifies users who posted or retweeted the incorrect information as needed.

[0690] Example: The server posts a correction tweet through its official account saying, "Company X's new product will not be provided for free. It costs $100. Click here for details." It also notifies users who retweeted the misinformation of the correction.

[0691] Operational Management and Reporting

[0692] The server periodically generates reports on the collected data, sentiment data, and misinformation detection and correction results, and provides them to companies and relevant public institutions. The reports include a list of detected misinformation, sentiment data, the spread of correction information, and its impact.

[0693] Example: A server generates a monthly report and sends it to a company's public relations department, which includes a list of misinformation about company X, a sentiment analysis of posts identified as misinformation, and the spread of corrections.

[0694] Prompt Sentence Examples

[0695] After a user posts news on Twitter that "Company Y has released a new product," the information begins to spread. Later, it is discovered that the content of the post is inaccurate, and the system is used to generate a correction and repost the post in a tone that matches the user's sentiment.

[0696] This system quickly detects misinformation on social media and delivers corrections in an appropriate tone based on the sentiment, thereby protecting the credibility of companies and providing accurate information.

[0697] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0698] Step 1: Gather information

[0699] Specific behavior:

[0700] Input: The server uses a social networking site's API (e.g., Twitter API) to collect public posts containing specific keywords, such as "Company X."

[0701] Data processing / calculation: The server sends an API request to retrieve data including the post content, user information, post date and time, number of retweets, and number of likes.

[0702] Output: Collected public posts and metadata stored in a database.

[0703] Example: A server collects tweets about "Company X" through an API and stores them as JSON format data.

[0704] Step 2: Analyzing and extracting post content

[0705] Specific behavior:

[0706] Input: Public submission data and metadata collected by the server.

[0707] Data processing / calculation: The server uses a Python natural language processing library (e.g., NLTK) to filter the collected posts to those containing specific keywords, and then uses an emotion engine to analyze the emotion data of the posts.

[0708] Output: Posts containing specific keywords and their sentiment data stored in a database.

[0709] Example: From a collection of 1,000 tweets, the server extracts 100 posts containing "Company X," and the sentiment engine identifies 50 as positive, 20 as negative, and 30 as neutral.

[0710] Step 3: Check against official information

[0711] Specific behavior:

[0712] Input: Post data and sentiment data extracted by the server, as well as official company information.

[0713] Data processing / calculation: The server uses a web scraping library (e.g., Beautiful Soup) to retrieve the latest official information from the company's official website and store it in a database. The extracted post content is then compared with this official information.

[0714] Output: Save the match results to a database.

[0715] Example: A server retrieves information from an official website that says "The price of a new product is $100," compares it with a post that says "Company X's new product is being offered for free," and discovers a contradiction.

[0716] Step 4: Identifying and flagging misinformation

[0717] Specific behavior:

[0718] Input: The server checks the results against the official information.

[0719] Data processing / calculation: Based on the matching results, the server flags any inconsistent posts as misinformation and manages them in a database along with the emotion data.

[0720] Output: An updated database containing misinformation flags and sentiment data.

[0721] Example: A post that says "It's provided for free" contradicts official information, so the server flags it as misinformation and manages it along with "negative" sentiment.

[0722] Step 5: Generate corrections

[0723] Specific behavior:

[0724] Input: Posts flagged by the server as misinformation, sentiment data, and official information.

[0725] Data processing / calculation: The server uses a natural language generation model (e.g., GPT-3) to automatically generate corrections to misinformation. The tone of the corrections is adjusted according to the emotion determined by the emotion engine.

[0726] Output: Corrective sentences in tone according to the emotion.

[0727] Example: A server generates a correction such as "Company X's new product is not free. It costs $100. Official information here." and uses a more polite tone for users who express negative sentiment.

[0728] Step 6: Distributing corrections

[0729] Specific behavior:

[0730] Input: Server-generated corrections and notification list (users who posted and retweeted the misinformation).

[0731] Data processing / calculation: The server uses the SNS API to post the correction from the official account and sends individual notifications to users on the notification list.

[0732] Output: The correction posted on social media and the users who received the notification.

[0733] Example: Your server tweets a correction: "Company X's new product is not free. It costs $100. Learn more." and sends notifications to relevant users.

[0734] Step 7: Operational Management and Reporting

[0735] Specific behavior:

[0736] Input: The server collects data from the database, sentiment data, misinformation history, and correction results.

[0737] Data processing / calculation: The server periodically aggregates the data, compiles it into a report, and outputs it as a PDF or CSV file.

[0738] Output: Generated reports (e.g. monthly reports).

[0739] Example: A server generates a monthly report containing a list of misinformation about Company X, the results of a sentiment analysis of posts identified as misinformation, and the spread of corrections, and sends it to the company's public relations department.

[0740] (Application example 2)

[0741] 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."

[0742] In today's information society, social networking services (SNS) have become a major means of spreading information, but this often leads to the spread of false information about companies and products. This false information is likely to have a negative impact on the company's brand image and credibility. Furthermore, corrections that do not take into account the emotions contained in user posts may leave users with a negative impression. A system that solves these problems is needed.

[0743] 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.

[0744] In this invention, the server includes means for collecting public posts, means for analyzing the collected public posts to extract posts containing specific keywords and acquiring emotion data, means for comparing the content of the extracted posts with official company information, means for identifying and flagging false information based on the comparison results, means for generating correction information in a tone appropriate to the emotion of the post determined to be false, means for posting the generated correction information to the SNS, and means for managing the history of collected data, emotion data, and correction information. This makes it possible to quickly detect false information about companies and products on the SNS and distribute correction information in an appropriate tone appropriate to the emotion.

[0745] "Public post" refers to a post that is publicly available for all users to view on a social networking service.

[0746] "Means of collection" refers to the function of using the API of a social networking service to obtain posts containing specific keywords.

[0747] "Means for analyzing and extracting" refers to a function for analyzing collected public posts using a text analysis algorithm and extracting posts that contain specific keywords.

[0748] "Means for obtaining emotion data" refers to a function that uses an emotion engine to identify emotions such as positive, negative, and neutral from the text of public posts and obtain that data.

[0749] "Means of matching" refers to the function for comparing the content of extracted posts with official company information and verifying whether they match.

[0750] "Means to identify and flag misinformation" refers to a function that identifies a post as misinformation when the content of the post contradicts official information and flags it in the database for management.

[0751] "Means for generating correction information" refers to a function that uses a generative AI model to automatically generate correction statements in an appropriate tone based on the emotion of posts that are determined to contain false information.

[0752] "Means of posting to social media" refers to the function for notifying the generated correction to users who posted or retweeted the false information on social media.

[0753] "Means for managing the history of data, emotional data, and correction information" refers to the function for systematically storing and managing collected data, emotional data, and the history of corrections to misinformation, and outputting it in report format as needed.

[0754] An "emotion engine" refers to algorithms or software that use natural language processing technology to analyze and identify emotions from text data.

[0755] A "generative AI model" refers to an artificial intelligence model that can learn from large amounts of text data and generate new text.

[0756] The system embodying the present invention includes a series of process flows for collecting, analyzing, and correcting public posts. A specific embodiment of the system is described below.

[0757] System Configuration

[0758] This system consists of the following main components:

[0759] server:

[0760] The server uses the social media platform's API to crawl public posts, analyzes them using a text analysis algorithm, and extracts posts containing specific keywords. It then uses an emotion engine to obtain the post's emotional data. Once analyzed, the post is compared with the company's official information to identify misinformation. A generative AI model then generates corrections in a tone appropriate to the emotion and posts them on social media. The data, emotional data, and correction history are stored in a database.

[0761] The specific hardware and software used are as follows:

[0762] SNS API: Twitter API, Facebook Graph API, etc.

[0763] Natural language processing libraries: NLTK, SpaCy

[0764] Sentiment analysis engine: VADER, TextBlob

[0765] Web scraping tool: Beautiful Soup

[0766] RSS Reader:RSS Feed Reader

[0767] Generative AI models: GPT-3, BERT

[0768] Data management tools: SQL databases, NoSQL databases

[0769] Examples include:

[0770] The server collects public tweets mentioning "specific brand X" via the Twitter API, analyzes the posts using the NLTK library, and extracts sentiment data using the VADER engine.

[0771] A post containing the misinformation "Brand X is offered for free" is compared with information on the company's official website and recognized as misinformation. A generative AI model (e.g., GPT-3) is then used to generate a correction, "Brand X's product is not free. It costs $100," and this correction is posted on social media.

[0772] Device:

[0773] The terminal refers to the device of the user who uses SNS and posts publicly. If the user posts incorrect information, they will receive a notification of the correction.

[0774] Examples include:

[0775] When a user tweets "Brand X is being offered for free" on a social networking site, the server collects and analyzes the tweet, and then generates a correction that is sent to the user's account.

[0776] User:

[0777] A user refers to an individual who posts and shares information on a social networking site.

[0778] For example, if a user spreads false information that "Brand X is being offered for free," the user will receive a correction that reads, "Brand X products are not free. They cost $100."

[0779] Prompt Sentence Examples

[0780] Here are some example prompts to input to a generative AI model:

[0781] markdown

[0782] To generate sentiment-based corrections for users, we provide the following information:

[0783] User Information

[0784] Username: User A

[0785] Emotion: Negative

[0786] Misinformation

[0787] Post content: New product from brand X is being offered for free

[0788] It's official: Brand X's new product is $100

[0789] Use the information above to generate a correction in the right tone. For example, "Brand X's new product is not free. It costs $100. Learn more."

[0790] In this way, it is possible to quickly detect misinformation about companies and products on social media and deliver corrections in an appropriate tone based on the sentiment.

[0791] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0792] Step 1:

[0793] The server uses the SNS API to collect public posts containing specific keywords (company names or product names). The input is the keyword and the SNS API token, and the public posts and metadata (post content, user information, posting date and time, number of retweets, number of likes) are obtained by API calls. The output is the collected post data.

[0794] Step 2:

[0795] The server analyzes the collected public posts using a text analysis algorithm (NLTK or SpaCy) to extract posts containing specific keywords. The input is the collected post data, and the text analysis algorithm extracts posts containing keywords. The output is the extracted post data.

[0796] Step 3:

[0797] The server analyzes the extracted posts using an emotion engine (VADER or TextBlob) to obtain emotion data. The input is the extracted post data, and the emotion engine obtains emotion data such as positive, negative, or neutral. The output is the post data with the emotion data added.

[0798] Step 4:

[0799] The server compares the post content with the emotional data added to it with the company's official information. The official information is obtained from web scraping or RSS readers. The input is the post data with the emotional data added and the official information data, and the matching algorithm detects matches or inconsistencies. The output is the matching result.

[0800] Step 5:

[0801] The server flags posts that contain inconsistencies based on the matching results. The input is the matching results, and if there are any inconsistencies, a misinformation flag is added. The output is the post data with the misinformation flag added.

[0802] Step 6:

[0803] Based on posts that the server determines to be misinformation, it uses a generative AI model (such as GPT-3 or BERT) to generate corrections in a tone appropriate to the emotion. The input is post data flagged as misinformation and official information data, and the generative AI model generates corrections in an appropriate tone. The output is the corrections.

[0804] Step 7:

[0805] The server posts the generated correction to SNS and notifies the users who posted the misinformation and those who retweeted it. The input is the correction and the target user information, and the correction is posted via the SNS API. The output is a notification that the post has been completed.

[0806] Step 8:

[0807] The server periodically generates a report containing the collected data, sentiment data, and correction history information, and provides it to the company and the relevant advertising agency. The input is all the data retrieved from the database, and the report is generated by the data management tool. The output is a report document.

[0808] 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.

[0809] 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.

[0810] 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.

[0811] [Third embodiment]

[0812] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0813] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0814] 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).

[0815] 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.

[0816] 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.

[0817] 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).

[0818] 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.

[0819] 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.

[0820] 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.

[0821] 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.

[0822] 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.

[0823] 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."

[0824] The present invention provides a system for preventing the spread of false information about company and product names on social media, and provides a process for automatically collecting, analyzing, and collating public posts and correcting the false information. Specific embodiments of the system are described below.

[0825] System Configuration

[0826] The system consists of the following main components:

[0827] 1. Server: A central location responsible for collecting, analyzing, collating data, identifying misinformation, and generating and distributing corrections.

[0828] 2. Device: A device on which a user uses SNS and posts publicly. If incorrect information is detected, you may receive a correction notice.

[0829] 3. Users: People who post and share information on social media.

[0830] Program processing

[0831] 1. Collection of information

[0832] The server periodically uses the social networking service's API to crawl public posts containing specific keywords, such as company names or product names.

[0833] Example: The server uses the keyword "Company X" to collect relevant tweets from Twitter's API. At the same time, metadata such as the post content, user information, posting date and time, number of retweets, and number of likes are also obtained.

[0834] 2. Analysis and extraction of post content

[0835] The server then runs the collected public posts through a text analysis algorithm, using natural language processing (NLP) techniques, to extract posts containing specific keywords.

[0836] Example: Analyze 1,000 public posts collected by the server to find posts containing the specified keyword "Company X" and extract 100 posts.

[0837] 3. Comparison with official information

[0838] The server then compares the content of the extracted public posts with official company information, often obtained through web scraping or RSS feeds, to determine whether the post matches existing official information.

[0839] Example: A server compares a post that says "Company X's new product is being offered for free" with information on the official website that says "The new product costs $100" and recognizes a discrepancy.

[0840] 4. Identifying and flagging misinformation

[0841] The server will then flag any posts that are inconsistent based on the results of the match, and posts identified as misinformation will be moderated within the system.

[0842] Example: A post containing the misinformation that servers are "provided for free" is flagged and added to a misinformation list.

[0843] 5. Generating Corrections

[0844] The server automatically generates corrections based on official information to correct misinformation, including the correct information and the source.

[0845] Example: The server creates a correction that reads, "Company X's new product costs $100. More information here."

[0846] 6. Distribution of Corrections

[0847] The server posts the generated correction information to the social networking site, and if necessary, notifies the user who posted the incorrect information and the user who retweeted it.

[0848] Example: In response to a false claim that "Company X's new product will be provided for free," the server tweets a correction from its official account and also sends a notification to users who have retweeted the false information.

[0849] Operational Management and Reporting

[0850] The server periodically generates reports on the collected data and the results of detecting and correcting misinformation, and provides them to companies and relevant public institutions. The reports include a list of detected misinformation, the spread of the correction information, and its impact.

[0851] Example: A server generates a monthly report and sends it to a company's marketing department, detailing the list of misinformation detected and the extent to which corrections to those misinformation have been disseminated.

[0852] In this way, this system efficiently detects misinformation on social media and quickly distributes corrections, thereby protecting the credibility of companies and providing accurate information.

[0853] The processing flow will be explained below.

[0854] Step 1:

[0855] The server accesses the social networking service's API and crawls public posts containing specific keywords (company names or product names). It sets the API endpoint and request parameters and periodically retrieves data.

[0856] Example: A server periodically crawls tweets containing the keyword "Company X" and retrieves metadata including the post content, user information, posting date and time, number of retweets, and number of likes.

[0857] Step 2:

[0858] The server crawls and retrieves the posted data, which is then passed through a text analysis algorithm. Natural language processing (NLP) techniques are used to analyze and extract posts containing specific keywords.

[0859] Example: The server analyzes 1,000 tweets collected and finds posts containing "Company X" and extracts 100 of them.

[0860] Step 3:

[0861] The server uses web scraping or RSS feeds to obtain official company information, and stores the information in a database.

[0862] Example: A server web scrapes new product information and product pricing information from the official website of "Company X" and stores it in a database.

[0863] Step 4:

[0864] The server compares the extracted social media posts with official information, using natural language processing technology to determine whether the posts match the official information.

[0865] Example: A server analyzes a social media post that says, "Company X's new product is being offered for free," and compares it with information on the official website that says, "The price of the new product is $100."

[0866] Step 5:

[0867] If the server does not match, it will flag the post as misinformation and keep it in a database.

[0868] Example: A post claiming that a server is "provided for free" is detected as contradicting official information, and the information is flagged as misinformation and added to a moderation list.

[0869] Step 6:

[0870] The server automatically generates a correction based on official information to correct misinformation. The correction includes the source of the official information and accurate information.

[0871] Example: The server automatically generates a correction: "Company X's new product costs $100. Official information is here."

[0872] Step 7:

[0873] The server generates a correction and posts it to the social networking site. If necessary, notifications are also sent to the user who posted the incorrect information and to users who retweeted it.

[0874] Example: In response to a false claim that "Company X's new product will be provided for free," the server tweets a correction from its official account and also sends a notification to users who have retweeted the false information.

[0875] Step 8:

[0876] The server generates a report of the collected data, the history of correction information, and the correction results, and provides it to companies and relevant public institutions.

[0877] Example: A server generates a monthly report and sends it to a company's marketing department, detailing the list of misinformation detected and the extent to which corrections to those misinformation have been disseminated.

[0878] In this way, this system efficiently detects misinformation on social media and quickly distributes corrections, thereby protecting the credibility of companies and providing accurate information.

[0879] Example 1

[0880] 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."

[0881] Currently, inaccurate and false information can spread quickly on online social networking services (SNS), causing problems that undermine the credibility of companies and products. Furthermore, delays in responding to false information can lead to further confusion, which could damage brand image and cause consumer confusion.

[0882] 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.

[0883] In this invention, the server includes means for collecting public posts, means for applying a text analysis algorithm to the collected public posts to extract posts containing specific keywords, means for comparing the content of the extracted posts with official information, means for identifying and flagging misinformation based on the comparison results, means for generating corrections containing correct information for posts determined to be misinformation, means for posting the generated corrections to a social networking service, and means for sending notifications to users who posted and retweeted the misinformation as needed, thereby enabling inaccurate and misinformation to be detected quickly and efficiently and corresponding corrections to be provided.

[0884] "Public Post" means a user's post that is publicly available on a social networking service.

[0885] "Means of collection" refers to algorithms and programs that automatically obtain public posts using APIs of social networking services, etc.

[0886] A "text analysis algorithm" refers to a program that uses natural language processing technology to extract specific keywords and meanings from text data.

[0887] "Means of extraction" refers to the process of using a text analysis algorithm to select data that meets certain conditions.

[0888] "Verification measures" refer to programs that compare the collected public posts with official information to check for matches or inconsistencies.

[0889] "Misinformation" refers to information that does not match official information or contains factually incorrect information.

[0890] "Flagging" refers to the process of assigning an identifying mark to information that meets certain conditions.

[0891] "Correction information" refers to content that corrects misinformation and includes correct information and sources of official information.

[0892] "Means for generating" refers to a program for automatically creating correct correction information for erroneous information.

[0893] "Means of posting" refers to the algorithms or programs used to disseminate the generated correction information on social networking services.

[0894] "Means for sending notifications" refers to a program for sending corrections to users who posted false information or spread it.

[0895] "Social networking service" refers to a service that allows users to share information with each other over the Internet.

[0896] "Report format" refers to a document format that visually presents the history of collected data and correction information in an easy-to-understand manner.

[0897] The present invention provides a system for preventing the spread of false information about company and product names on social networking services (SNS), and provides a process for automatically collecting, analyzing, and collating public posts and correcting the false information. A specific embodiment of this system will be described.

[0898] System Configuration

[0899] The system consists of the following main components:

[0900] 1. Server: A central location responsible for collecting, analyzing, collating data, identifying misinformation, and generating and distributing corrections.

[0901] 2. Device: A device on which a user uses SNS and posts publicly. If incorrect information is detected, you may receive a correction notice.

[0902] 3. Users: People who post and share information on social media.

[0903] Hardware and Software Use

[0904] The server should be a computer with powerful computing capabilities. Recommended hardware includes a fast CPU, large memory capacity, and an SSD. Software used includes Python, NLTK, spaCy, BeautifulSoup, and the Twitter API.

[0905] For example, a server periodically uses the Twitter API to collect tweets containing the keyword "Company X." It then uses a Python script to perform text analysis using NLTK and spaCy to extract posts containing specific keywords. The server then uses BeautifulSoup to scrape the latest information from the company's official website and compares the extracted posts with official information.

[0906] Generation and distribution of corrections

[0907] If a contradiction is detected, the server automatically generates a correction based on the official information. The correction includes the correct information and its source. The server then reposts the correction to the social networking site, and notifies the user who posted the misinformation and those who retweeted it. This quickly prevents the spread of misinformation.

[0908] Operational Management and Reporting

[0909] The server periodically generates reports on the collected data, misinformation detection, and correction results, and provides them to companies and relevant public institutions. The reports include a list of detected misinformation, the spread of correction information, and its impact. For example, the server generates a monthly report, converts it to PDF format, and emails it to the company's marketing department.

[0910] Prompt Sentence Examples

[0911] Please describe a system that detects misinformation about Company X on social media and automatically generates and distributes corrections. Please provide details, including examples, of how this system works.

[0912] In this way, this system efficiently detects misinformation on social media and quickly distributes corrections, thereby protecting the credibility of companies and providing accurate information.

[0913] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0914] Step 1:

[0915] The server collects public posts. The server periodically crawls public posts containing specific keywords, such as "Company X," using the API of the social networking site. Specifically, the server uses Python to access the Twitter API and collect tweets containing the keyword "Company X." The data obtained from the API of the social networking site includes metadata such as the post content, user information, posting date and time, number of retweets, and number of likes. This provides the public posts to be collected as input data.

[0916] Step 2:

[0917] The server runs the collected public posts through a text analysis algorithm to extract posts containing specific keywords. Specifically, the server uses natural language processing libraries such as NLTK and spaCy to analyze collected tweets for posts containing the keyword "Company X" and extract the relevant posts. At this stage, the collected post data is input and posts that match the keywords are extracted.

[0918] Step 3:

[0919] The server compares the extracted public post content with official information. Specifically, the server uses a web scraping tool such as BeautifulSoup to retrieve the latest product information and news from the company's official website. It then compares the extracted post content with the retrieved official information to check for matches and inconsistencies. The input data is the extracted post and official information, and the comparison results are output based on this.

[0920] Step 4:

[0921] The server identifies and flags misinformation based on the matching results. Specifically, the server flags posts containing the misinformation "provided for free" as misinformation and stores it in a database. The input data is the matching results, and based on the results, misinformation is identified and flagged data is output.

[0922] Step 5:

[0923] The server generates correction information based on official information. Specifically, the server automatically generates a correction statement that includes the correct information and its source. For example, it generates a correction statement such as, "Company X's new product costs $100. Click here for more information." The input data is official information and incorrect information, and the output is correction information.

[0924] Step 6:

[0925] The server posts the generated correction information to SNS, and if necessary, sends notifications to the user who posted the misinformation and to users who retweeted it.Specifically, the server tweets the correction information on its official account and also sends individual notifications to users who retweeted the misinformation.The input data is the generated correction information, and the output is posted to SNS, and notifications are sent.

[0926] Step 7:

[0927] The server generates and provides the collected data, correction information history, and correction results in report format.Specifically, the server creates a monthly report, converts it to PDF format, and emails it to the company's marketing department.The input data are the collected data and correction results, and the report is generated and provided as output.

[0928] (Application example 1)

[0929] 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."

[0930] The spread of misinformation on social media is a serious problem that damages the trust of companies and individuals. In particular, with the rapid increase in the number of social media users, misinformation spreads quickly, creating a need for systems that provide fast and accurate corrections. There is also a growing need for automated solutions that can save users the trouble of monitoring and correcting relevant misinformation themselves.

[0931] 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.

[0932] In this invention, the server includes means for collecting public posts, means for analyzing the collected public posts to extract posts containing specific keywords, means for comparing the content of the extracted posts with official information, means for identifying misinformation based on the comparison results and flagging the misinformation, means for generating corrections to posts determined to be misinformation, means for posting the generated corrections to an online environment, means for setting specific keywords and notifying users when misinformation is detected, means for automatically generating corrections using a generative AI model, and means for setting prompt text for the generative AI model. This enables quick detection and correction of misinformation spreading on SNS and quick notification to users.

[0933] A "means for collecting public posts" is a device or program that has the functionality to automatically collect posts that are publicly available on social media or online platforms.

[0934] "Means for analyzing collected public posts and extracting posts containing specific keywords" refers to a device or program that has the function of analyzing collected post data and identifying and extracting posts containing specified keywords.

[0935] "Means for comparing the content of extracted posts with official information" refers to a device or program that has the function of comparing and matching the content of extracted posts with official information provided by companies or official institutions.

[0936] "Means for identifying false information based on the matching results and flagging false information" refers to a device or program that has the function of detecting false information from the matching results and flagging posts that are determined to be false information.

[0937] "Means for generating correction information for posts determined to contain false information" refers to a device or program that has the function of automatically generating a correction statement to provide accurate information for posts identified as false information.

[0938] "Means for posting the generated correction information online" refers to a device or program that has the function of automatically posting the generated correction information to social media or other online platforms.

[0939] "Means for setting specific keywords and notifying users when misinformation is detected" refers to a device or program that has the function of allowing users to set keywords of interest, collecting and analyzing information based on those keywords, and notifying users when misinformation is detected.

[0940] "Means for automatically generating correction information using a generative AI model" refers to a device or program that has the function of automatically generating correction information for erroneous information using an artificial intelligence model.

[0941] "Means for setting a prompt sentence to a generative AI model" refers to a device or program that has the function of setting a prompt sentence to be input to a generative AI model and performing information processing based on that prompt.

[0942] The present invention is a system that effectively detects misinformation spread on social media and quickly distributes corrections. The system includes the following main components:

[0943] System Configuration

[0944] 1. Server:

[0945] The server has the ability to collect public posts using the API of a social networking site. For example, it uses the Twython library to collect posts from Twitter that contain specific keywords.

[0946] The server uses the SpaCy library to process the collected data in natural language, analyzing and extracting posts containing specific keywords.

[0947] The server scrapes official information using BeautifulSoup and the requests library to match the extracted public posts with official information obtained from the websites of companies or official institutions.

[0948] Based on the results of the comparison, the server flags any posts that contradict official information as misinformation.

[0949] 2. Use of generative AI models:

[0950] For posts that are determined to contain false information, the server automatically generates corrections using a generative AI model. The generative AI model uses the latest natural language processing technology and generates corrections by setting prompt sentences.

[0951] For example, a prompt such as "We have detected a tweet that says, 'Company X's new product will be offered for free.' Please correct this information." can be input into the generative AI model to obtain appropriate correction information.

[0952] 3. User Notice:

[0953] If certain keywords are set, the server will notify users when misinformation is detected, allowing users to respond quickly.

[0954] 4. Distribution of Corrections:

[0955] The server has the function of posting the generated correction information online, for example, by posting the correction information from an official account on social media to spread it.

[0956] Specific examples

[0957] As a specific example, consider a case where misinformation related to Company X is spread on social media. In this case, the server collects related posts from Twitter using the keyword "Company X." By analyzing the posts, a tweet containing the misinformation "provided for free" is detected. The server compares it with official information, flags it as misinformation, and then uses a generative AI model to automatically generate a correction such as "Company X's new product costs $100. Click here for more information." The server then posts the generated correction on social media and sends a notification to the affected users.

[0958] This system enables the rapid detection and correction of misinformation, protecting the credibility of businesses and individuals. Furthermore, by utilizing generative AI models and prompts, corrections can be automatically generated, enabling efficient correction of information. Specific examples of prompts are as follows:

[0959] Example prompt sentence:

[0960] "We have detected a tweet that states 'Company X's new product will be offered for free.' Please correct this information."

[0961] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0962] Step 1:

[0963] The server uses the SNS API to collect public posts containing specific keywords. Specifically, it uses the Twython library to set keywords such as "Company X" and collects tweets that match the keywords. The input is the specific keyword, and the output is the collected post data. This collected data includes the post content, poster information, posting date and time, number of retweets, number of likes, etc.

[0964] Step 2:

[0965] The server analyzes the collected public posts using natural language processing (NLP) technology and extracts posts containing specific keywords. Specifically, it performs text analysis using the SpaCy library to extract posts containing the keyword "Company X." The input is the collected post data, and the output is the analyzed and extracted post data. The analysis includes data processing such as part-of-speech tagging and named entity extraction.

[0966] Step 3:

[0967] The server compares the post content extracted through analysis with official information. Specifically, it uses BeautifulSoup and the requests library to scrape official information from the company's official website and compares it with the extracted post content. The input is the extracted post data and official information, and the output is the comparison result. This comparison involves data calculations to determine whether the post content matches the official information.

[0968] Step 4:

[0969] Based on the matching results, the server identifies misinformation and flags the post. For example, if a post saying "Company X's new product will be offered for free" contradicts official information, it will be flagged as misinformation. The input is the matching results, and the output is the flagged post data. This flagging is a data processing process to identify misinformation.

[0970] Step 5:

[0971] The server uses a generative AI model to automatically generate corrections for posts determined to contain false information. Specifically, the prompt "We have detected a tweet stating that 'Company X's new product will be offered for free.' Please correct this information." is input into the generative AI model, which then outputs appropriate corrections. The input is the flagged post data and the prompt, and the output is the corrections. The AI ​​model performs data calculations based on this prompt, generating the corrections.

[0972] Step 6:

[0973] The server posts the generated correction information to the online environment. Specifically, it tweets the correction information from the official SNS account and, if necessary, sends a notification to users who retweeted the misinformation. The input is the generated correction information, and the output is the posting of the correction information on the SNS and a user notification. This posting and notification are specific actions that transmit data.

[0974] 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.

[0975] The present invention provides a system that prevents the spread of false information about company names and product names on social networking sites and also takes into consideration the feelings of users. Specific embodiments of the invention are described below.

[0976] System Configuration

[0977] The system consists of the following main components:

[0978] 1. Server:

[0979] Responsible for collecting, analyzing, collating data, identifying misinformation, and generating and distributing corrections.

[0980] Using an emotion engine, the content and context of users' posts are analyzed to identify emotions.

[0981] 2. Terminal:

[0982] A device that allows users to use social networking services and make public posts.

[0983] If incorrect information is detected, you will receive a notification of the correction.

[0984] 3. User:

[0985] People who post and share information on social media.

[0986] Program processing

[0987] 1. Collection of information

[0988] The server uses the social networking site's API to crawl public posts containing specific keywords (such as company or product names), and also collects metadata including the post content, user information, posting date and time, number of retweets, and number of likes.

[0989] Example: The server collects tweets containing the keyword "Company X" and obtains the post content, user information, posting date and time, number of retweets, and number of likes.

[0990] 2. Analysis and extraction of post content

[0991] The server runs a text analysis algorithm through the collected public posts to extract posts containing specific keywords, and in the process also uses an emotion engine to obtain sentiment data for the posts.

[0992] Example: A server collects 1,000 tweets, analyzes 100 posts containing "Company X," and the sentiment engine identifies that 50 of them have positive sentiment, 20 have negative sentiment, and 30 are neutral.

[0993] 3. Comparison with official information

[0994] The server compares the extracted social media posts with the company's official information, which is obtained using web scraping or RSS feeds and stored in a database. This comparison determines whether the posts match existing official information.

[0995] Example: A server compares a post that says "Company X's new product is being offered for free" with information on Company X's official website that says "The new product costs $100" and recognizes a contradiction.

[0996] 4. Identifying and flagging misinformation

[0997] Based on the results of the comparison, the server flags any posts that contain inconsistencies as misinformation and manages them in a database. The server also manages the sentiment information of the posts.

[0998] Example: A post that says the server will be "provided for free" is flagged as misinformation because it contradicts official information, and is managed along with "negative" sentiment.

[0999] 5. Generating Corrections

[1000] For posts that the server determines to be misinformation, the system automatically generates a correction based on official information. Using an emotion engine, the correction is adjusted to a tone that corresponds to the emotion, such as positive, negative, or neutral. The correction can also include a citation of the official information source.

[1001] Example: The server generates a correction message that reads, "Company X's new product will not be provided for free. It costs $100. Official information is here." If the user has expressed negative sentiment, the correction message will be written in a more polite tone.

[1002] 6. Distribution of Corrections

[1003] The server posts the generated correction information to social media, and notifies users who posted or retweeted the incorrect information as needed.

[1004] Example: The server posts a correction tweet through its official account saying, "Company X's new product will not be provided for free. It costs $100. Click here for details." It also notifies users who retweeted the misinformation of the correction.

[1005] Operational Management and Reporting

[1006] The server periodically generates reports on the collected data, sentiment data, and misinformation detection and correction results, and provides them to companies and relevant public institutions. The reports include a list of detected misinformation, sentiment data, the spread of correction information, and its impact.

[1007] Example: A server generates a monthly report and sends it to a company's public relations department, which includes a list of misinformation about company X, a sentiment analysis of posts identified as misinformation, and the spread of corrections.

[1008] This system quickly detects misinformation on social media and delivers corrections in an appropriate tone based on the sentiment, thereby protecting the credibility of companies and providing accurate information.

[1009] The processing flow will be explained below.

[1010] Step 1:

[1011] The server accesses the social networking service's API and crawls public posts containing specific keywords (company names or product names). It sets the API endpoint and request parameters and periodically retrieves data.

[1012] Example: The server crawls tweets containing the keyword "Company X" every hour and retrieves the post content, user information, posting date and time, number of retweets, and number of likes.

[1013] Step 2:

[1014] The server crawls and collects post data, then runs it through a text analysis algorithm to extract posts containing specific keywords. Keyword analysis is performed using natural language processing (NLP) technology.

[1015] Example: Analyze and extract 100 posts containing "Company X" from 1,000 tweets collected by the server.

[1016] Step 3:

[1017] The server then runs the extracted public posts through an emotion engine to obtain the emotion data of the post content, which identifies emotion categories such as positive, negative, and neutral.

[1018] Example: A server analyzes the sentiment of 100 posts about "Company X" and identifies 50 as positive, 30 as negative, and 20 as neutral.

[1019] Step 4:

[1020] The server uses web scraping and RSS feeds to obtain official company information, which is then stored in a database.

[1021] Example: A server scrapes new product information and pricing information from the official website of "Company X" and stores it in a database.

[1022] Step 5:

[1023] The server compares the extracted social media posts with official information, using natural language processing technology to determine whether the posts match the official information.

[1024] Example: A server compares a social media post that says "Company X's new product is being offered for free" with information on Company X's official website that says "The new product costs $100" and recognizes a contradiction.

[1025] Step 6:

[1026] Based on the results of the comparison, the server flags any inconsistent posts as misinformation and manages them in a database, along with the sentiment information of the posts.

[1027] Example: A post that says the server will be "provided for free" contradicts official information, so we flag it as misinformation and manage it together with "negative" sentiment information.

[1028] Step 7:

[1029] For posts that the server determines to be misinformation, the system automatically generates a correction based on official information. Using an emotion engine, the correction is adjusted to a tone that corresponds to the emotion of the post, and includes a citation of the official source.

[1030] Example: The server generates a correction that reads, "Company X's new product is not free. It costs $100. Official information is here." In this case, the correction is written in a more polite tone for posts that express negative sentiment.

[1031] Step 8:

[1032] The server posts the generated correction information to the social networking site, and if necessary, sends notifications to users who posted or retweeted the incorrect information.

[1033] Example: The server tweets a correction through its official account: "Company X's new product will not be free. It will cost $100." It also notifies users who retweeted the false information.

[1034] Step 9:

[1035] The server generates reports on the collected data, sentiment data, misinformation detection, and correction results, and provides them to companies and relevant public institutions.

[1036] Example: A server generates a monthly report and sends it to a company's public relations department, which includes a list of misinformation about company X, a sentiment analysis of posts identified as misinformation, and the spread of corrections.

[1037] In this way, this system quickly detects misinformation on social media and delivers corrections in an appropriate tone based on the sentiment, thereby protecting the credibility of companies and providing accurate information.

[1038] Example 2

[1039] 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."

[1040] The rapid and widespread spread of misinformation about company and product names on the internet, particularly on social media platforms, has become a problem. This misinformation not only has a negative impact on a company's credibility and brand image, but also conveys incorrect information to consumers. It is also difficult to correct the misinformation, and it is also difficult to contain the impact of the misinformation that has spread. Furthermore, there is a lack of technology to generate appropriate corrections that take user sentiment into account. A system is needed to solve this issue and provide accurate information while maintaining a company's credibility.

[1041] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting public posts, means for analyzing the collected public posts and extracting posts containing specific keywords, means for acquiring emotion data of the extracted posts, means for comparing the content of the extracted posts with official information, means for identifying false information based on the comparison result and flagging the false information, means for generating correction information in a tone corresponding to the emotion for posts determined to be false information, means for posting the generated correction information to an SNS, and means for notifying the generated correction information to users who posted the false information and users who retweeted the false information. This makes it possible to quickly detect false information and provide appropriate correction information while taking user emotions into consideration.

[1042] "Public posts" refer to information posted by users with public settings on social media, online forums, etc.

[1043] The "means of collection" refers to a part of the system that has the function of collecting public posts using social media APIs and web scraping technology.

[1044] The "means for analyzing" is a part of a system that has the function of analyzing collected public posts using text analysis algorithms and natural language processing techniques and extracting posts that contain specific keywords.

[1045] The "means for extracting" is a part of the system that has the function of selecting information that meets specific conditions from the analyzed data.

[1046] "Emotional data" is data that is classified based on an analysis of the content of posts to determine emotions such as positive, negative, or neutral.

[1047] "Official information" is reliable information officially released by a company or organization, such as from a website, press release, RSS feed, etc.

[1048] A "mechanism for matching" is a part of the system that has the ability to compare the collected content of public posts with official information and detect matches or inconsistencies.

[1049] "Misinformation" refers to public posts whose content contradicts official information.

[1050] "Flagging measures" are part of a system that has the ability to manage posts identified as misinformation by attaching specific marks or tags to them.

[1051] "Correction information" is content generated to provide correct information based on official information in response to posts identified as misinformation.

[1052] The "means for generating" is a part of the system that has the function of automatically generating correction information based on official information.

[1053] "Means of notification" refers to the part of the system that has the function of communicating the generated correction information to the user who posted the misinformation or retweeted it.

[1054] The present invention provides a system that prevents the spread of false information about company names and product names on social networking sites and also takes into consideration the feelings of users. A specific embodiment of the present invention will now be described.

[1055] System Configuration

[1056] The system consists of the following main components:

[1057] 1. Server:

[1058] Responsible for collecting, analyzing, collating data, identifying misinformation, and generating and distributing corrections.

[1059] Using an emotion engine, the content and context of users' posts are analyzed to identify emotions.

[1060] Specifically, we use Python natural language processing libraries (e.g., NLTK), sentiment analysis APIs, and web scraping libraries (e.g., Beautiful Soup).

[1061] 2. Terminal:

[1062] A device that users use to use SNS and post publicly. This includes smartphones and PCs.

[1063] If incorrect information is detected, you will receive a notification of the correction.

[1064] 3. User:

[1065] These are people who post and share information on social media.

[1066] Program processing

[1067] Collection of information

[1068] The server uses the social networking site's API to crawl public posts containing specific keywords (e.g., "Company X"), collecting metadata including the post content, user information, posting date and time, number of retweets, and number of likes.

[1069] Example: The server collects tweets containing the keyword "Company X" and obtains the post content, user information, posting date and time, number of retweets, and number of likes.

[1070] Analyzing and extracting post content

[1071] The server runs a text analysis algorithm through the collected public posts to extract posts containing specific keywords, and in the process also uses an emotion engine to obtain sentiment data for the posts.

[1072] Example: A server collects 1,000 tweets, analyzes 100 posts containing "Company X," and the sentiment engine identifies that 50 of them have positive sentiment, 20 have negative sentiment, and 30 are neutral.

[1073] Comparison with official information

[1074] The server compares the extracted social media posts with the company's official information, which is obtained using web scraping or RSS feeds and stored in a database. This comparison determines whether the posts match existing official information.

[1075] Example: A server compares a post that says "Company X's new product is being offered for free" with information on the company's official website that says "The new product costs $100" and recognizes a contradiction.

[1076] Identifying and flagging misinformation

[1077] Based on the results of the comparison, the server flags any posts that contain inconsistencies as misinformation and manages them in a database. The server also manages the sentiment information of the posts.

[1078] Example: A post that says the server will be "provided for free" is flagged as misinformation because it contradicts official information, and is managed along with "negative" sentiment.

[1079] Generate correction information

[1080] For posts that the server determines to be misinformation, the system automatically generates a correction based on official information. Using an emotion engine, the correction is adjusted to a tone that corresponds to the emotion, such as positive, negative, or neutral. The correction can also include a citation of the official information source.

[1081] Example: The server generates a correction message that reads, "Company X's new product will not be provided for free. It costs $100. Official information is here." If the user has expressed negative sentiment, the correction message will be written in a more polite tone.

[1082] Distribution of correction information

[1083] The server posts the generated correction information to social media, and notifies users who posted or retweeted the incorrect information as needed.

[1084] Example: The server posts a correction tweet through its official account saying, "Company X's new product will not be provided for free. It costs $100. Click here for details." It also notifies users who retweeted the misinformation of the correction.

[1085] Operational Management and Reporting

[1086] The server periodically generates reports on the collected data, sentiment data, and misinformation detection and correction results, and provides them to companies and relevant public institutions. The reports include a list of detected misinformation, sentiment data, the spread of correction information, and its impact.

[1087] Example: A server generates a monthly report and sends it to a company's public relations department, which includes a list of misinformation about company X, a sentiment analysis of posts identified as misinformation, and the spread of corrections.

[1088] Prompt Sentence Examples

[1089] After a user posts news on Twitter that "Company Y has released a new product," the information begins to spread. Later, it is discovered that the content of the post is inaccurate, and the system is used to generate a correction and repost the post in a tone that matches the user's sentiment.

[1090] This system quickly detects misinformation on social media and delivers corrections in an appropriate tone based on the sentiment, thereby protecting the credibility of companies and providing accurate information.

[1091] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1092] Step 1: Gather information

[1093] Specific behavior:

[1094] Input: The server uses a social networking site's API (e.g., Twitter API) to collect public posts containing specific keywords, such as "Company X."

[1095] Data processing / calculation: The server sends an API request to retrieve data including the post content, user information, post date and time, number of retweets, and number of likes.

[1096] Output: Collected public posts and metadata stored in a database.

[1097] Example: A server collects tweets about "Company X" through an API and stores them as JSON format data.

[1098] Step 2: Analyzing and extracting post content

[1099] Specific behavior:

[1100] Input: Public submission data and metadata collected by the server.

[1101] Data processing / calculation: The server uses a Python natural language processing library (e.g., NLTK) to filter the collected posts to those containing specific keywords, and then uses an emotion engine to analyze the emotion data of the posts.

[1102] Output: Posts containing specific keywords and their sentiment data stored in a database.

[1103] Example: From a collection of 1,000 tweets, the server extracts 100 posts containing "Company X," and the sentiment engine identifies 50 as positive, 20 as negative, and 30 as neutral.

[1104] Step 3: Check against official information

[1105] Specific behavior:

[1106] Input: Post data and sentiment data extracted by the server, as well as official company information.

[1107] Data processing / calculation: The server uses a web scraping library (e.g., Beautiful Soup) to retrieve the latest official information from the company's official website and store it in a database. The extracted post content is then compared with this official information.

[1108] Output: Save the match results to a database.

[1109] Example: A server retrieves information from an official website that says "The price of a new product is $100," compares it with a post that says "Company X's new product is being offered for free," and discovers a contradiction.

[1110] Step 4: Identifying and flagging misinformation

[1111] Specific behavior:

[1112] Input: The server checks the results against the official information.

[1113] Data processing / calculation: Based on the matching results, the server flags any inconsistent posts as misinformation and manages them in a database along with the emotion data.

[1114] Output: An updated database containing misinformation flags and sentiment data.

[1115] Example: A post that says "It's provided for free" contradicts official information, so the server flags it as misinformation and manages it along with "negative" sentiment.

[1116] Step 5: Generate corrections

[1117] Specific behavior:

[1118] Input: Posts flagged by the server as misinformation, sentiment data, and official information.

[1119] Data processing / calculation: The server uses a natural language generation model (e.g., GPT-3) to automatically generate corrections to misinformation. The tone of the corrections is adjusted according to the emotion determined by the emotion engine.

[1120] Output: Corrective sentences in tone according to the emotion.

[1121] Example: A server generates a correction such as "Company X's new product is not free. It costs $100. Official information here." and uses a more polite tone for users who express negative sentiment.

[1122] Step 6: Distributing corrections

[1123] Specific behavior:

[1124] Input: Server-generated corrections and notification list (users who posted and retweeted the misinformation).

[1125] Data processing / calculation: The server uses the SNS API to post the correction from the official account and sends individual notifications to users on the notification list.

[1126] Output: The correction posted on social media and the users who received the notification.

[1127] Example: Your server tweets a correction: "Company X's new product is not free. It costs $100. Learn more." and sends notifications to relevant users.

[1128] Step 7: Operational Management and Reporting

[1129] Specific behavior:

[1130] Input: The server collects data from the database, sentiment data, misinformation history, and correction results.

[1131] Data processing / calculation: The server periodically aggregates the data, compiles it into a report, and outputs it as a PDF or CSV file.

[1132] Output: Generated reports (e.g. monthly reports).

[1133] Example: A server generates a monthly report containing a list of misinformation about Company X, the results of a sentiment analysis of posts identified as misinformation, and the spread of corrections, and sends it to the company's public relations department.

[1134] (Application example 2)

[1135] 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."

[1136] In today's information society, social networking services (SNS) have become a major means of spreading information, but this often leads to the spread of false information about companies and products. This false information is likely to have a negative impact on the company's brand image and credibility. Furthermore, corrections that do not take into account the emotions contained in user posts may leave users with a negative impression. A system that solves these problems is needed.

[1137] 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.

[1138] In this invention, the server includes means for collecting public posts, means for analyzing the collected public posts to extract posts containing specific keywords and acquiring emotion data, means for comparing the content of the extracted posts with official company information, means for identifying and flagging false information based on the comparison results, means for generating correction information in a tone appropriate to the emotion of the post determined to be false, means for posting the generated correction information to the SNS, and means for managing the history of collected data, emotion data, and correction information. This makes it possible to quickly detect false information about companies and products on the SNS and distribute correction information in an appropriate tone appropriate to the emotion.

[1139] "Public post" refers to a post that is publicly available for all users to view on a social networking service.

[1140] "Means of collection" refers to the function of using the API of a social networking service to obtain posts containing specific keywords.

[1141] "Means for analyzing and extracting" refers to a function for analyzing collected public posts using a text analysis algorithm and extracting posts that contain specific keywords.

[1142] "Means for obtaining emotion data" refers to a function that uses an emotion engine to identify emotions such as positive, negative, and neutral from the text of public posts and obtain that data.

[1143] "Means of matching" refers to the function for comparing the content of extracted posts with official company information and verifying whether they match.

[1144] "Means to identify and flag misinformation" refers to a function that identifies a post as misinformation when the content of the post contradicts official information and flags it in the database for management.

[1145] "Means for generating correction information" refers to a function that uses a generative AI model to automatically generate correction statements in an appropriate tone based on the emotion of posts that are determined to contain false information.

[1146] "Means of posting to social media" refers to the function for notifying the generated correction to users who posted or retweeted the false information on social media.

[1147] "Means for managing the history of data, emotional data, and correction information" refers to the function for systematically storing and managing collected data, emotional data, and the history of corrections to misinformation, and outputting it in report format as needed.

[1148] An "emotion engine" refers to algorithms or software that use natural language processing technology to analyze and identify emotions from text data.

[1149] A "generative AI model" refers to an artificial intelligence model that can learn from large amounts of text data and generate new text.

[1150] The system embodying the present invention includes a series of process flows for collecting, analyzing, and correcting public posts. A specific embodiment of the system is described below.

[1151] System Configuration

[1152] This system consists of the following main components:

[1153] server:

[1154] The server uses the social media platform's API to crawl public posts, analyzes them using a text analysis algorithm, and extracts posts containing specific keywords. It then uses an emotion engine to obtain the post's emotional data. Once analyzed, the post is compared with the company's official information to identify misinformation. A generative AI model then generates corrections in a tone appropriate to the emotion and posts them on social media. The data, emotional data, and correction history are stored in a database.

[1155] The specific hardware and software used are as follows:

[1156] SNS API: Twitter API, Facebook Graph API, etc.

[1157] Natural language processing libraries: NLTK, SpaCy

[1158] Sentiment analysis engine: VADER, TextBlob

[1159] Web scraping tool: Beautiful Soup

[1160] RSS Reader:RSS Feed Reader

[1161] Generative AI models: GPT-3, BERT

[1162] Data management tools: SQL databases, NoSQL databases

[1163] Examples include:

[1164] The server collects public tweets mentioning "specific brand X" via the Twitter API, analyzes the posts using the NLTK library, and extracts sentiment data using the VADER engine.

[1165] A post containing the misinformation "Brand X is offered for free" is compared with information on the company's official website and recognized as misinformation. A generative AI model (e.g., GPT-3) is then used to generate a correction, "Brand X's product is not free. It costs $100," and this correction is posted on social media.

[1166] Device:

[1167] The terminal refers to the device of the user who uses SNS and posts publicly. If the user posts incorrect information, they will receive a notification of the correction.

[1168] Examples include:

[1169] When a user tweets "Brand X is being offered for free" on a social networking site, the server collects and analyzes the tweet, and then generates a correction that is sent to the user's account.

[1170] User:

[1171] A user refers to an individual who posts and shares information on a social networking site.

[1172] For example, if a user spreads false information that "Brand X is being offered for free," the user will receive a correction that reads, "Brand X products are not free. They cost $100."

[1173] Prompt Sentence Examples

[1174] Here are some example prompts to input to a generative AI model:

[1175] markdown

[1176] To generate sentiment-based corrections for users, we provide the following information:

[1177] User Information

[1178] Username: User A

[1179] Emotion: Negative

[1180] Misinformation

[1181] Post content: New product from brand X is being offered for free

[1182] It's official: Brand X's new product is $100

[1183] Use the information above to generate a correction in the right tone. For example, "Brand X's new product is not free. It costs $100. Learn more."

[1184] In this way, it is possible to quickly detect misinformation about companies and products on social media and deliver corrections in an appropriate tone based on the sentiment.

[1185] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1186] Step 1:

[1187] The server uses the SNS API to collect public posts containing specific keywords (company names or product names). The input is the keyword and the SNS API token, and the public posts and metadata (post content, user information, posting date and time, number of retweets, number of likes) are obtained by API calls. The output is the collected post data.

[1188] Step 2:

[1189] The server analyzes the collected public posts using a text analysis algorithm (NLTK or SpaCy) to extract posts containing specific keywords. The input is the collected post data, and the text analysis algorithm extracts posts containing keywords. The output is the extracted post data.

[1190] Step 3:

[1191] The server analyzes the extracted posts using an emotion engine (VADER or TextBlob) to obtain emotion data. The input is the extracted post data, and the emotion engine obtains emotion data such as positive, negative, or neutral. The output is the post data with the emotion data added.

[1192] Step 4:

[1193] The server compares the post content with the emotional data added to it with the company's official information. The official information is obtained from web scraping or RSS readers. The input is the post data with the emotional data added and the official information data, and the matching algorithm detects matches or inconsistencies. The output is the matching result.

[1194] Step 5:

[1195] The server flags posts that contain inconsistencies based on the matching results. The input is the matching results, and if there are any inconsistencies, a misinformation flag is added. The output is the post data with the misinformation flag added.

[1196] Step 6:

[1197] Based on posts that the server determines to be misinformation, it uses a generative AI model (such as GPT-3 or BERT) to generate corrections in a tone appropriate to the emotion. The input is post data flagged as misinformation and official information data, and the generative AI model generates corrections in an appropriate tone. The output is the corrections.

[1198] Step 7:

[1199] The server posts the generated correction to SNS and notifies the users who posted the misinformation and those who retweeted it. The input is the correction and the target user information, and the correction is posted via the SNS API. The output is a notification that the post has been completed.

[1200] Step 8:

[1201] The server periodically generates a report containing the collected data, sentiment data, and correction history information, and provides it to the company and the relevant advertising agency. The input is all the data retrieved from the database, and the report is generated by the data management tool. The output is a report document.

[1202] 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.

[1203] 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.

[1204] 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.

[1205] [Fourth embodiment]

[1206] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1207] 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.

[1208] 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).

[1209] 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.

[1210] 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.

[1211] 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).

[1212] 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.

[1213] 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.

[1214] 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.

[1215] 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.

[1216] 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.

[1217] 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.

[1218] 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."

[1219] The present invention provides a system for preventing the spread of false information about company and product names on social media, and provides a process for automatically collecting, analyzing, and collating public posts and correcting the false information. Specific embodiments of the system are described below.

[1220] System Configuration

[1221] The system consists of the following main components:

[1222] 1. Server: A central location responsible for collecting, analyzing, collating data, identifying misinformation, and generating and distributing corrections.

[1223] 2. Device: A device on which a user uses SNS and posts publicly. If incorrect information is detected, you may receive a correction notice.

[1224] 3. Users: People who post and share information on social media.

[1225] Program processing

[1226] 1. Collection of information

[1227] The server periodically uses the social networking service's API to crawl public posts containing specific keywords, such as company names or product names.

[1228] Example: The server uses the keyword "Company X" to collect relevant tweets from Twitter's API. At the same time, metadata such as the post content, user information, posting date and time, number of retweets, and number of likes are also obtained.

[1229] 2. Analysis and extraction of post content

[1230] The server then runs the collected public posts through a text analysis algorithm, using natural language processing (NLP) techniques, to extract posts containing specific keywords.

[1231] Example: Analyze 1,000 public posts collected by the server to find posts containing the specified keyword "Company X" and extract 100 posts.

[1232] 3. Comparison with official information

[1233] The server then compares the content of the extracted public posts with official company information, often obtained through web scraping or RSS feeds, to determine whether the post matches existing official information.

[1234] Example: A server compares a post that says "Company X's new product is being offered for free" with information on the official website that says "The new product costs $100" and recognizes a discrepancy.

[1235] 4. Identifying and flagging misinformation

[1236] The server will then flag any posts that are inconsistent based on the results of the match, and posts identified as misinformation will be moderated within the system.

[1237] Example: A post containing the misinformation that servers are "provided for free" is flagged and added to a misinformation list.

[1238] 5. Generating Corrections

[1239] The server automatically generates corrections based on official information to correct misinformation, including the correct information and the source.

[1240] Example: The server creates a correction that reads, "Company X's new product costs $100. More information here."

[1241] 6. Distribution of Corrections

[1242] The server posts the generated correction information to the social networking site, and if necessary, notifies the user who posted the incorrect information and the user who retweeted it.

[1243] Example: In response to a false claim that "Company X's new product will be provided for free," the server tweets a correction from its official account and also sends a notification to users who have retweeted the false information.

[1244] Operational Management and Reporting

[1245] The server periodically generates reports on the collected data and the results of detecting and correcting misinformation, and provides them to companies and relevant public institutions. The reports include a list of detected misinformation, the spread of the correction information, and its impact.

[1246] Example: A server generates a monthly report and sends it to a company's marketing department, detailing the list of misinformation detected and the extent to which corrections to those misinformation have been disseminated.

[1247] In this way, this system efficiently detects misinformation on social media and quickly distributes corrections, thereby protecting the credibility of companies and providing accurate information.

[1248] The processing flow will be explained below.

[1249] Step 1:

[1250] The server accesses the social networking service's API and crawls public posts containing specific keywords (company names or product names). It sets the API endpoint and request parameters and periodically retrieves data.

[1251] Example: A server periodically crawls tweets containing the keyword "Company X" and retrieves metadata including the post content, user information, posting date and time, number of retweets, and number of likes.

[1252] Step 2:

[1253] The server crawls and retrieves the posted data, which is then passed through a text analysis algorithm. Natural language processing (NLP) techniques are used to analyze and extract posts containing specific keywords.

[1254] Example: The server analyzes 1,000 tweets collected and finds posts containing "Company X" and extracts 100 of them.

[1255] Step 3:

[1256] The server uses web scraping or RSS feeds to obtain official company information, and stores the information in a database.

[1257] Example: A server web scrapes new product information and product pricing information from the official website of "Company X" and stores it in a database.

[1258] Step 4:

[1259] The server compares the extracted social media posts with official information, using natural language processing technology to determine whether the posts match the official information.

[1260] Example: A server analyzes a social media post that says, "Company X's new product is being offered for free," and compares it with information on the official website that says, "The price of the new product is $100."

[1261] Step 5:

[1262] If the server does not match, it will flag the post as misinformation and keep it in a database.

[1263] Example: A post claiming that a server is "provided for free" is detected as contradicting official information, and the information is flagged as misinformation and added to a moderation list.

[1264] Step 6:

[1265] The server automatically generates a correction based on official information to correct misinformation. The correction includes the source of the official information and accurate information.

[1266] Example: The server automatically generates a correction: "Company X's new product costs $100. Official information is here."

[1267] Step 7:

[1268] The server generates a correction and posts it to the social networking site. If necessary, notifications are also sent to the user who posted the incorrect information and to users who retweeted it.

[1269] Example: In response to a false claim that "Company X's new product will be provided for free," the server tweets a correction from its official account and also sends a notification to users who have retweeted the false information.

[1270] Step 8:

[1271] The server generates a report of the collected data, the history of correction information, and the correction results, and provides it to companies and relevant public institutions.

[1272] Example: A server generates a monthly report and sends it to a company's marketing department, detailing the list of misinformation detected and the extent to which corrections to those misinformation have been disseminated.

[1273] In this way, this system efficiently detects misinformation on social media and quickly distributes corrections, thereby protecting the credibility of companies and providing accurate information.

[1274] Example 1

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

[1276] Currently, inaccurate and false information can spread quickly on online social networking services (SNS), causing problems that undermine the credibility of companies and products. Furthermore, delays in responding to false information can lead to further confusion, which could damage brand image and cause consumer confusion.

[1277] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1278] In this invention, the server includes means for collecting public posts, means for applying a text analysis algorithm to the collected public posts to extract posts containing specific keywords, means for comparing the content of the extracted posts with official information, means for identifying and flagging misinformation based on the comparison results, means for generating corrections containing correct information for posts determined to be misinformation, means for posting the generated corrections to a social networking service, and means for sending notifications to users who posted and retweeted the misinformation as needed, thereby enabling inaccurate and misinformation to be detected quickly and efficiently and corresponding corrections to be provided.

[1279] "Public Post" means a user's post that is publicly available on a social networking service.

[1280] "Means of collection" refers to algorithms and programs that automatically obtain public posts using APIs of social networking services, etc.

[1281] A "text analysis algorithm" refers to a program that uses natural language processing technology to extract specific keywords and meanings from text data.

[1282] "Means of extraction" refers to the process of using a text analysis algorithm to select data that meets certain conditions.

[1283] "Verification measures" refer to programs that compare the collected public posts with official information to check for matches or inconsistencies.

[1284] "Misinformation" refers to information that does not match official information or contains factually incorrect information.

[1285] "Flagging" refers to the process of assigning an identifying mark to information that meets certain conditions.

[1286] "Correction information" refers to content that corrects misinformation and includes correct information and sources of official information.

[1287] "Means for generating" refers to a program for automatically creating correct correction information for erroneous information.

[1288] "Means of posting" refers to the algorithms or programs used to disseminate the generated correction information on social networking services.

[1289] "Means for sending notifications" refers to a program for sending corrections to users who posted false information or spread it.

[1290] "Social networking service" refers to a service that allows users to share information with each other over the Internet.

[1291] "Report format" refers to a document format that visually presents the history of collected data and correction information in an easy-to-understand manner.

[1292] The present invention provides a system for preventing the spread of false information about company and product names on social networking services (SNS), and provides a process for automatically collecting, analyzing, and collating public posts and correcting the false information. A specific embodiment of this system will be described.

[1293] System Configuration

[1294] The system consists of the following main components:

[1295] 1. Server: A central location responsible for collecting, analyzing, collating data, identifying misinformation, and generating and distributing corrections.

[1296] 2. Device: A device on which a user uses SNS and posts publicly. If incorrect information is detected, you may receive a correction notice.

[1297] 3. Users: People who post and share information on social media.

[1298] Hardware and Software Use

[1299] The server should be a computer with powerful computing capabilities. Recommended hardware includes a fast CPU, large memory capacity, and an SSD. Software used includes Python, NLTK, spaCy, BeautifulSoup, and the Twitter API.

[1300] For example, a server periodically uses the Twitter API to collect tweets containing the keyword "Company X." It then uses a Python script to perform text analysis using NLTK and spaCy to extract posts containing specific keywords. The server then uses BeautifulSoup to scrape the latest information from the company's official website and compares the extracted posts with official information.

[1301] Generation and distribution of corrections

[1302] If a contradiction is detected, the server automatically generates a correction based on the official information. The correction includes the correct information and its source. The server then reposts the correction to the social networking site, and notifies the user who posted the misinformation and those who retweeted it. This quickly prevents the spread of misinformation.

[1303] Operational Management and Reporting

[1304] The server periodically generates reports on the collected data, misinformation detection, and correction results, and provides them to companies and relevant public institutions. The reports include a list of detected misinformation, the spread of correction information, and its impact. For example, the server generates a monthly report, converts it to PDF format, and emails it to the company's marketing department.

[1305] Prompt Sentence Examples

[1306] Please describe a system that detects misinformation about Company X on social media and automatically generates and distributes corrections. Please provide details, including examples, of how this system works.

[1307] In this way, this system efficiently detects misinformation on social media and quickly distributes corrections, thereby protecting the credibility of companies and providing accurate information.

[1308] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1309] Step 1:

[1310] The server collects public posts. The server periodically crawls public posts containing specific keywords, such as "Company X," using the API of the social networking site. Specifically, the server uses Python to access the Twitter API and collect tweets containing the keyword "Company X." The data obtained from the API of the social networking site includes metadata such as the post content, user information, posting date and time, number of retweets, and number of likes. This provides the public posts to be collected as input data.

[1311] Step 2:

[1312] The server runs the collected public posts through a text analysis algorithm to extract posts containing specific keywords. Specifically, the server uses natural language processing libraries such as NLTK and spaCy to analyze collected tweets for posts containing the keyword "Company X" and extract the relevant posts. At this stage, the collected post data is input and posts that match the keywords are extracted.

[1313] Step 3:

[1314] The server compares the extracted public post content with official information. Specifically, the server uses a web scraping tool such as BeautifulSoup to retrieve the latest product information and news from the company's official website. It then compares the extracted post content with the retrieved official information to check for matches and inconsistencies. The input data is the extracted post and official information, and the comparison results are output based on this.

[1315] Step 4:

[1316] The server identifies and flags misinformation based on the matching results. Specifically, the server flags posts containing the misinformation "provided for free" as misinformation and stores it in a database. The input data is the matching results, and based on the results, misinformation is identified and flagged data is output.

[1317] Step 5:

[1318] The server generates correction information based on official information. Specifically, the server automatically generates a correction statement that includes the correct information and its source. For example, it generates a correction statement such as, "Company X's new product costs $100. Click here for more information." The input data is official information and incorrect information, and the output is correction information.

[1319] Step 6:

[1320] The server posts the generated correction information to SNS, and if necessary, sends notifications to the user who posted the misinformation and to users who retweeted it.Specifically, the server tweets the correction information on its official account and also sends individual notifications to users who retweeted the misinformation.The input data is the generated correction information, and the output is posted to SNS, and notifications are sent.

[1321] Step 7:

[1322] The server generates and provides the collected data, correction information history, and correction results in report format.Specifically, the server creates a monthly report, converts it to PDF format, and emails it to the company's marketing department.The input data are the collected data and correction results, and the report is generated and provided as output.

[1323] (Application example 1)

[1324] 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."

[1325] The spread of misinformation on social media is a serious problem that damages the trust of companies and individuals. In particular, with the rapid increase in the number of social media users, misinformation spreads quickly, creating a need for systems that provide fast and accurate corrections. There is also a growing need for automated solutions that can save users the trouble of monitoring and correcting relevant misinformation themselves.

[1326] 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.

[1327] In this invention, the server includes means for collecting public posts, means for analyzing the collected public posts to extract posts containing specific keywords, means for comparing the content of the extracted posts with official information, means for identifying misinformation based on the comparison results and flagging the misinformation, means for generating corrections to posts determined to be misinformation, means for posting the generated corrections to an online environment, means for setting specific keywords and notifying users when misinformation is detected, means for automatically generating corrections using a generative AI model, and means for setting prompt text for the generative AI model. This enables quick detection and correction of misinformation spreading on SNS and quick notification to users.

[1328] A "means for collecting public posts" is a device or program that has the functionality to automatically collect posts that are publicly available on social media or online platforms.

[1329] "Means for analyzing collected public posts and extracting posts containing specific keywords" refers to a device or program that has the function of analyzing collected post data and identifying and extracting posts containing specified keywords.

[1330] "Means for comparing the content of extracted posts with official information" refers to a device or program that has the function of comparing and matching the content of extracted posts with official information provided by companies or official institutions.

[1331] "Means for identifying false information based on the matching results and flagging false information" refers to a device or program that has the function of detecting false information from the matching results and flagging posts that are determined to be false information.

[1332] "Means for generating correction information for posts determined to contain false information" refers to a device or program that has the function of automatically generating a correction statement to provide accurate information for posts identified as false information.

[1333] "Means for posting the generated correction information online" refers to a device or program that has the function of automatically posting the generated correction information to social media or other online platforms.

[1334] "Means for setting specific keywords and notifying users when misinformation is detected" refers to a device or program that has the function of allowing users to set keywords of interest, collecting and analyzing information based on those keywords, and notifying users when misinformation is detected.

[1335] "Means for automatically generating correction information using a generative AI model" refers to a device or program that has the function of automatically generating correction information for erroneous information using an artificial intelligence model.

[1336] "Means for setting a prompt sentence to a generative AI model" refers to a device or program that has the function of setting a prompt sentence to be input to a generative AI model and performing information processing based on that prompt.

[1337] The present invention is a system that effectively detects misinformation spread on social media and quickly distributes corrections. The system includes the following main components:

[1338] System Configuration

[1339] 1. Server:

[1340] The server has the ability to collect public posts using the API of a social networking site. For example, it uses the Twython library to collect posts from Twitter that contain specific keywords.

[1341] The server uses the SpaCy library to process the collected data in natural language, analyzing and extracting posts containing specific keywords.

[1342] The server scrapes official information using BeautifulSoup and the requests library to match the extracted public posts with official information obtained from the websites of companies or official institutions.

[1343] Based on the results of the comparison, the server flags any posts that contradict official information as misinformation.

[1344] 2. Use of generative AI models:

[1345] For posts that are determined to contain false information, the server automatically generates corrections using a generative AI model. The generative AI model uses the latest natural language processing technology and generates corrections by setting prompt sentences.

[1346] For example, a prompt such as "We have detected a tweet that says, 'Company X's new product will be offered for free.' Please correct this information." can be input into the generative AI model to obtain appropriate correction information.

[1347] 3. User Notice:

[1348] If certain keywords are set, the server will notify users when misinformation is detected, allowing users to respond quickly.

[1349] 4. Distribution of Corrections:

[1350] The server has the function of posting the generated correction information online, for example, by posting the correction information from an official account on social media to spread it.

[1351] Specific examples

[1352] As a specific example, consider a case where misinformation related to Company X is spread on social media. In this case, the server collects related posts from Twitter using the keyword "Company X." By analyzing the posts, a tweet containing the misinformation "provided for free" is detected. The server compares it with official information, flags it as misinformation, and then uses a generative AI model to automatically generate a correction such as "Company X's new product costs $100. Click here for more information." The server then posts the generated correction on social media and sends a notification to the affected users.

[1353] This system enables the rapid detection and correction of misinformation, protecting the credibility of businesses and individuals. Furthermore, by utilizing generative AI models and prompts, corrections can be automatically generated, enabling efficient correction of information. Specific examples of prompts are as follows:

[1354] Example prompt sentence:

[1355] "We have detected a tweet that states 'Company X's new product will be offered for free.' Please correct this information."

[1356] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1357] Step 1:

[1358] The server uses the SNS API to collect public posts containing specific keywords. Specifically, it uses the Twython library to set keywords such as "Company X" and collects tweets that match the keywords. The input is the specific keyword, and the output is the collected post data. This collected data includes the post content, poster information, posting date and time, number of retweets, number of likes, etc.

[1359] Step 2:

[1360] The server analyzes the collected public posts using natural language processing (NLP) technology and extracts posts containing specific keywords. Specifically, it performs text analysis using the SpaCy library to extract posts containing the keyword "Company X." The input is the collected post data, and the output is the analyzed and extracted post data. The analysis includes data processing such as part-of-speech tagging and named entity extraction.

[1361] Step 3:

[1362] The server compares the post content extracted through analysis with official information. Specifically, it uses BeautifulSoup and the requests library to scrape official information from the company's official website and compares it with the extracted post content. The input is the extracted post data and official information, and the output is the comparison result. This comparison involves data calculations to determine whether the post content matches the official information.

[1363] Step 4:

[1364] Based on the matching results, the server identifies misinformation and flags the post. For example, if a post saying "Company X's new product will be offered for free" contradicts official information, it will be flagged as misinformation. The input is the matching results, and the output is the flagged post data. This flagging is a data processing process to identify misinformation.

[1365] Step 5:

[1366] The server uses a generative AI model to automatically generate corrections for posts determined to contain false information. Specifically, the prompt "We have detected a tweet stating that 'Company X's new product will be offered for free.' Please correct this information." is input into the generative AI model, which then outputs appropriate corrections. The input is the flagged post data and the prompt, and the output is the corrections. The AI ​​model performs data calculations based on this prompt, generating the corrections.

[1367] Step 6:

[1368] The server posts the generated correction information to the online environment. Specifically, it tweets the correction information from the official SNS account and, if necessary, sends a notification to users who retweeted the misinformation. The input is the generated correction information, and the output is the posting of the correction information on the SNS and a user notification. This posting and notification are specific actions that transmit data.

[1369] 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.

[1370] The present invention provides a system that prevents the spread of false information about company names and product names on social networking sites and also takes into consideration the feelings of users. Specific embodiments of the invention are described below.

[1371] System Configuration

[1372] The system consists of the following main components:

[1373] 1. Server:

[1374] Responsible for collecting, analyzing, collating data, identifying misinformation, and generating and distributing corrections.

[1375] Using an emotion engine, the content and context of users' posts are analyzed to identify emotions.

[1376] 2. Terminal:

[1377] A device that allows users to use social networking services and make public posts.

[1378] If incorrect information is detected, you will receive a notification of the correction.

[1379] 3. User:

[1380] People who post and share information on social media.

[1381] Program processing

[1382] 1. Collection of information

[1383] The server uses the social networking site's API to crawl public posts containing specific keywords (such as company or product names), and also collects metadata including the post content, user information, posting date and time, number of retweets, and number of likes.

[1384] Example: The server collects tweets containing the keyword "Company X" and obtains the post content, user information, posting date and time, number of retweets, and number of likes.

[1385] 2. Analysis and extraction of post content

[1386] The server runs a text analysis algorithm through the collected public posts to extract posts containing specific keywords, and in the process also uses an emotion engine to obtain sentiment data for the posts.

[1387] Example: A server collects 1,000 tweets, analyzes 100 posts containing "Company X," and the sentiment engine identifies that 50 of them have positive sentiment, 20 have negative sentiment, and 30 are neutral.

[1388] 3. Comparison with official information

[1389] The server compares the extracted social media posts with the company's official information, which is obtained using web scraping or RSS feeds and stored in a database. This comparison determines whether the posts match existing official information.

[1390] Example: A server compares a post that says "Company X's new product is being offered for free" with information on Company X's official website that says "The new product costs $100" and recognizes a contradiction.

[1391] 4. Identifying and flagging misinformation

[1392] Based on the results of the comparison, the server flags any posts that contain inconsistencies as misinformation and manages them in a database. The server also manages the sentiment information of the posts.

[1393] Example: A post that says the server will be "provided for free" is flagged as misinformation because it contradicts official information, and is managed along with "negative" sentiment.

[1394] 5. Generating Corrections

[1395] For posts that the server determines to be misinformation, the system automatically generates a correction based on official information. Using an emotion engine, the correction is adjusted to a tone that corresponds to the emotion, such as positive, negative, or neutral. The correction can also include a citation of the official information source.

[1396] Example: The server generates a correction message that reads, "Company X's new product will not be provided for free. It costs $100. Official information is here." If the user has expressed negative sentiment, the correction message will be written in a more polite tone.

[1397] 6. Distribution of Corrections

[1398] The server posts the generated correction information to social media, and notifies users who posted or retweeted the incorrect information as needed.

[1399] Example: The server posts a correction tweet through its official account saying, "Company X's new product will not be provided for free. It costs $100. Click here for details." It also notifies users who retweeted the misinformation of the correction.

[1400] Operational Management and Reporting

[1401] The server periodically generates reports on the collected data, sentiment data, and misinformation detection and correction results, and provides them to companies and relevant public institutions. The reports include a list of detected misinformation, sentiment data, the spread of correction information, and its impact.

[1402] Example: A server generates a monthly report and sends it to a company's public relations department, which includes a list of misinformation about company X, a sentiment analysis of posts identified as misinformation, and the spread of corrections.

[1403] This system quickly detects misinformation on social media and delivers corrections in an appropriate tone based on the sentiment, thereby protecting the credibility of companies and providing accurate information.

[1404] The processing flow will be explained below.

[1405] Step 1:

[1406] The server accesses the social networking service's API and crawls public posts containing specific keywords (company names or product names). It sets the API endpoint and request parameters and periodically retrieves data.

[1407] Example: The server crawls tweets containing the keyword "Company X" every hour and retrieves the post content, user information, posting date and time, number of retweets, and number of likes.

[1408] Step 2:

[1409] The server crawls and collects post data, then runs it through a text analysis algorithm to extract posts containing specific keywords. Keyword analysis is performed using natural language processing (NLP) technology.

[1410] Example: Analyze and extract 100 posts containing "Company X" from 1,000 tweets collected by the server.

[1411] Step 3:

[1412] The server then runs the extracted public posts through an emotion engine to obtain the emotion data of the post content, which identifies emotion categories such as positive, negative, and neutral.

[1413] Example: A server analyzes the sentiment of 100 posts about "Company X" and identifies 50 as positive, 30 as negative, and 20 as neutral.

[1414] Step 4:

[1415] The server uses web scraping and RSS feeds to obtain official company information, which is then stored in a database.

[1416] Example: A server scrapes new product information and pricing information from the official website of "Company X" and stores it in a database.

[1417] Step 5:

[1418] The server compares the extracted social media posts with official information, using natural language processing technology to determine whether the posts match the official information.

[1419] Example: A server compares a social media post that says "Company X's new product is being offered for free" with information on Company X's official website that says "The new product costs $100" and recognizes a contradiction.

[1420] Step 6:

[1421] Based on the results of the comparison, the server flags any inconsistent posts as misinformation and manages them in a database, along with the sentiment information of the posts.

[1422] Example: A post that says the server will be "provided for free" contradicts official information, so we flag it as misinformation and manage it together with "negative" sentiment information.

[1423] Step 7:

[1424] For posts that the server determines to be misinformation, the system automatically generates a correction based on official information. Using an emotion engine, the correction is adjusted to a tone that corresponds to the emotion of the post, and includes a citation of the official source.

[1425] Example: The server generates a correction that reads, "Company X's new product is not free. It costs $100. Official information is here." In this case, the correction is written in a more polite tone for posts that express negative sentiment.

[1426] Step 8:

[1427] The server posts the generated correction information to the social networking site, and if necessary, sends notifications to users who posted or retweeted the incorrect information.

[1428] Example: The server tweets a correction through its official account: "Company X's new product will not be free. It will cost $100." It also notifies users who retweeted the false information.

[1429] Step 9:

[1430] The server generates reports on the collected data, sentiment data, misinformation detection, and correction results, and provides them to companies and relevant public institutions.

[1431] Example: A server generates a monthly report and sends it to a company's public relations department, which includes a list of misinformation about company X, a sentiment analysis of posts identified as misinformation, and the spread of corrections.

[1432] In this way, this system quickly detects misinformation on social media and delivers corrections in an appropriate tone based on the sentiment, thereby protecting the credibility of companies and providing accurate information.

[1433] Example 2

[1434] 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."

[1435] The rapid and widespread spread of misinformation about company and product names on the internet, particularly on social media platforms, has become a problem. This misinformation not only has a negative impact on a company's credibility and brand image, but also conveys incorrect information to consumers. It is also difficult to correct the misinformation, and it is also difficult to contain the impact of the misinformation that has spread. Furthermore, there is a lack of technology to generate appropriate corrections that take user sentiment into account. A system is needed to solve this issue and provide accurate information while maintaining a company's credibility.

[1436] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting public posts, means for analyzing the collected public posts and extracting posts containing specific keywords, means for acquiring emotion data of the extracted posts, means for comparing the content of the extracted posts with official information, means for identifying false information based on the comparison result and flagging the false information, means for generating correction information in a tone corresponding to the emotion for posts determined to be false information, means for posting the generated correction information to an SNS, and means for notifying the generated correction information to users who posted the false information and users who retweeted the false information. This makes it possible to quickly detect false information and provide appropriate correction information while taking user emotions into consideration.

[1437] "Public posts" refer to information posted by users with public settings on social media, online forums, etc.

[1438] The "means of collection" refers to a part of the system that has the function of collecting public posts using social media APIs and web scraping technology.

[1439] The "means for analyzing" is a part of a system that has the function of analyzing collected public posts using text analysis algorithms and natural language processing techniques and extracting posts that contain specific keywords.

[1440] The "means for extracting" is a part of the system that has the function of selecting information that meets specific conditions from the analyzed data.

[1441] "Emotional data" is data that is classified based on an analysis of the content of posts to determine emotions such as positive, negative, or neutral.

[1442] "Official information" is reliable information officially released by a company or organization, such as from a website, press release, RSS feed, etc.

[1443] A "mechanism for matching" is a part of the system that has the ability to compare the collected content of public posts with official information and detect matches or inconsistencies.

[1444] "Misinformation" refers to public posts whose content contradicts official information.

[1445] "Flagging measures" are part of a system that has the ability to manage posts identified as misinformation by attaching specific marks or tags to them.

[1446] "Correction information" is content generated to provide correct information based on official information in response to posts identified as misinformation.

[1447] The "means for generating" is a part of the system that has the function of automatically generating correction information based on official information.

[1448] "Means of notification" refers to the part of the system that has the function of communicating the generated correction information to the user who posted the misinformation or retweeted it.

[1449] The present invention provides a system that prevents the spread of false information about company names and product names on social networking sites and also takes into consideration the feelings of users. A specific embodiment of the present invention will now be described.

[1450] System Configuration

[1451] The system consists of the following main components:

[1452] 1. Server:

[1453] Responsible for collecting, analyzing, collating data, identifying misinformation, and generating and distributing corrections.

[1454] Using an emotion engine, the content and context of users' posts are analyzed to identify emotions.

[1455] Specifically, we use Python natural language processing libraries (e.g., NLTK), sentiment analysis APIs, and web scraping libraries (e.g., Beautiful Soup).

[1456] 2. Terminal:

[1457] A device that users use to use SNS and post publicly. This includes smartphones and PCs.

[1458] If incorrect information is detected, you will receive a notification of the correction.

[1459] 3. User:

[1460] These are people who post and share information on social media.

[1461] Program processing

[1462] Collection of information

[1463] The server uses the social networking site's API to crawl public posts containing specific keywords (e.g., "Company X"), collecting metadata including the post content, user information, posting date and time, number of retweets, and number of likes.

[1464] Example: The server collects tweets containing the keyword "Company X" and obtains the post content, user information, posting date and time, number of retweets, and number of likes.

[1465] Analyzing and extracting post content

[1466] The server runs a text analysis algorithm through the collected public posts to extract posts containing specific keywords, and in the process also uses an emotion engine to obtain sentiment data for the posts.

[1467] Example: A server collects 1,000 tweets, analyzes 100 posts containing "Company X," and the sentiment engine identifies that 50 of them have positive sentiment, 20 have negative sentiment, and 30 are neutral.

[1468] Comparison with official information

[1469] The server compares the extracted social media posts with the company's official information, which is obtained using web scraping or RSS feeds and stored in a database. This comparison determines whether the posts match existing official information.

[1470] Example: A server compares a post that says "Company X's new product is being offered for free" with information on the company's official website that says "The new product costs $100" and recognizes a contradiction.

[1471] Identifying and flagging misinformation

[1472] Based on the results of the comparison, the server flags any posts that contain inconsistencies as misinformation and manages them in a database. The server also manages the sentiment information of the posts.

[1473] Example: A post that says the server will be "provided for free" is flagged as misinformation because it contradicts official information, and is managed along with "negative" sentiment.

[1474] Generate correction information

[1475] For posts that the server determines to be misinformation, the system automatically generates a correction based on official information. Using an emotion engine, the correction is adjusted to a tone that corresponds to the emotion, such as positive, negative, or neutral. The correction can also include a citation of the official information source.

[1476] Example: The server generates a correction message that reads, "Company X's new product will not be provided for free. It costs $100. Official information is here." If the user has expressed negative sentiment, the correction message will be written in a more polite tone.

[1477] Distribution of correction information

[1478] The server posts the generated correction information to social media, and notifies users who posted or retweeted the incorrect information as needed.

[1479] Example: The server posts a correction tweet through its official account saying, "Company X's new product will not be provided for free. It costs $100. Click here for details." It also notifies users who retweeted the misinformation of the correction.

[1480] Operational Management and Reporting

[1481] The server periodically generates reports on the collected data, sentiment data, and misinformation detection and correction results, and provides them to companies and relevant public institutions. The reports include a list of detected misinformation, sentiment data, the spread of correction information, and its impact.

[1482] Example: A server generates a monthly report and sends it to a company's public relations department, which includes a list of misinformation about company X, a sentiment analysis of posts identified as misinformation, and the spread of corrections.

[1483] Prompt Sentence Examples

[1484] After a user posts news on Twitter that "Company Y has released a new product," the information begins to spread. Later, it is discovered that the content of the post is inaccurate, and the system is used to generate a correction and repost the post in a tone that matches the user's sentiment.

[1485] This system quickly detects misinformation on social media and delivers corrections in an appropriate tone based on the sentiment, thereby protecting the credibility of companies and providing accurate information.

[1486] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1487] Step 1: Gather information

[1488] Specific behavior:

[1489] Input: The server uses a social networking site's API (e.g., Twitter API) to collect public posts containing specific keywords, such as "Company X."

[1490] Data processing / calculation: The server sends an API request to retrieve data including the post content, user information, post date and time, number of retweets, and number of likes.

[1491] Output: Collected public posts and metadata stored in a database.

[1492] Example: A server collects tweets about "Company X" through an API and stores them as JSON format data.

[1493] Step 2: Analyzing and extracting post content

[1494] Specific behavior:

[1495] Input: Public submission data and metadata collected by the server.

[1496] Data processing / calculation: The server uses a Python natural language processing library (e.g., NLTK) to filter the collected posts to those containing specific keywords, and then uses an emotion engine to analyze the emotion data of the posts.

[1497] Output: Posts containing specific keywords and their sentiment data stored in a database.

[1498] Example: From a collection of 1,000 tweets, the server extracts 100 posts containing "Company X," and the sentiment engine identifies 50 as positive, 20 as negative, and 30 as neutral.

[1499] Step 3: Check against official information

[1500] Specific behavior:

[1501] Input: Post data and sentiment data extracted by the server, as well as official company information.

[1502] Data processing / calculation: The server uses a web scraping library (e.g., Beautiful Soup) to retrieve the latest official information from the company's official website and store it in a database. The extracted post content is then compared with this official information.

[1503] Output: Save the match results to a database.

[1504] Example: A server retrieves information from an official website that says "The price of a new product is $100," compares it with a post that says "Company X's new product is being offered for free," and discovers a contradiction.

[1505] Step 4: Identifying and flagging misinformation

[1506] Specific behavior:

[1507] Input: The server checks the results against the official information.

[1508] Data processing / calculation: Based on the matching results, the server flags any inconsistent posts as misinformation and manages them in a database along with the emotion data.

[1509] Output: An updated database containing misinformation flags and sentiment data.

[1510] Example: A post that says "It's provided for free" contradicts official information, so the server flags it as misinformation and manages it along with "negative" sentiment.

[1511] Step 5: Generate corrections

[1512] Specific behavior:

[1513] Input: Posts flagged by the server as misinformation, sentiment data, and official information.

[1514] Data processing / calculation: The server uses a natural language generation model (e.g., GPT-3) to automatically generate corrections to misinformation. The tone of the corrections is adjusted according to the emotion determined by the emotion engine.

[1515] Output: Corrective sentences in tone according to the emotion.

[1516] Example: A server generates a correction such as "Company X's new product is not free. It costs $100. Official information here." and uses a more polite tone for users who express negative sentiment.

[1517] Step 6: Distributing corrections

[1518] Specific behavior:

[1519] Input: Server-generated corrections and notification list (users who posted and retweeted the misinformation).

[1520] Data processing / calculation: The server uses the SNS API to post the correction from the official account and sends individual notifications to users on the notification list.

[1521] Output: The correction posted on social media and the users who received the notification.

[1522] Example: Your server tweets a correction: "Company X's new product is not free. It costs $100. Learn more." and sends notifications to relevant users.

[1523] Step 7: Operational Management and Reporting

[1524] Specific behavior:

[1525] Input: The server collects data from the database, sentiment data, misinformation history, and correction results.

[1526] Data processing / calculation: The server periodically aggregates the data, compiles it into a report, and outputs it as a PDF or CSV file.

[1527] Output: Generated reports (e.g. monthly reports).

[1528] Example: A server generates a monthly report containing a list of misinformation about Company X, the results of a sentiment analysis of posts identified as misinformation, and the spread of corrections, and sends it to the company's public relations department.

[1529] (Application example 2)

[1530] 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."

[1531] In today's information society, social networking services (SNS) have become a major means of spreading information, but this often leads to the spread of false information about companies and products. This false information is likely to have a negative impact on the company's brand image and credibility. Furthermore, corrections that do not take into account the emotions contained in user posts may leave users with a negative impression. A system that solves these problems is needed.

[1532] 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.

[1533] In this invention, the server includes means for collecting public posts, means for analyzing the collected public posts to extract posts containing specific keywords and acquiring emotion data, means for comparing the content of the extracted posts with official company information, means for identifying and flagging false information based on the comparison results, means for generating correction information in a tone appropriate to the emotion of the post determined to be false, means for posting the generated correction information to the SNS, and means for managing the history of collected data, emotion data, and correction information. This makes it possible to quickly detect false information about companies and products on the SNS and distribute correction information in an appropriate tone appropriate to the emotion.

[1534] "Public post" refers to a post that is publicly available for all users to view on a social networking service.

[1535] "Means of collection" refers to the function of using the API of a social networking service to obtain posts containing specific keywords.

[1536] "Means for analyzing and extracting" refers to a function for analyzing collected public posts using a text analysis algorithm and extracting posts that contain specific keywords.

[1537] "Means for obtaining emotion data" refers to a function that uses an emotion engine to identify emotions such as positive, negative, and neutral from the text of public posts and obtain that data.

[1538] "Means of matching" refers to the function for comparing the content of extracted posts with official company information and verifying whether they match.

[1539] "Means to identify and flag misinformation" refers to a function that identifies a post as misinformation when the content of the post contradicts official information and flags it in the database for management.

[1540] "Means for generating correction information" refers to a function that uses a generative AI model to automatically generate correction statements in an appropriate tone based on the emotion of posts that are determined to contain false information.

[1541] "Means of posting to social media" refers to the function for notifying the generated correction to users who posted or retweeted the false information on social media.

[1542] "Means for managing the history of data, emotional data, and correction information" refers to the function for systematically storing and managing collected data, emotional data, and the history of corrections to misinformation, and outputting it in report format as needed.

[1543] An "emotion engine" refers to algorithms or software that use natural language processing technology to analyze and identify emotions from text data.

[1544] A "generative AI model" refers to an artificial intelligence model that can learn from large amounts of text data and generate new text.

[1545] The system embodying the present invention includes a series of process flows for collecting, analyzing, and correcting public posts. A specific embodiment of the system is described below.

[1546] System Configuration

[1547] This system consists of the following main components:

[1548] server:

[1549] The server uses the social media platform's API to crawl public posts, analyzes them using a text analysis algorithm, and extracts posts containing specific keywords. It then uses an emotion engine to obtain the post's emotional data. Once analyzed, the post is compared with the company's official information to identify misinformation. A generative AI model then generates corrections in a tone appropriate to the emotion and posts them on social media. The data, emotional data, and correction history are stored in a database.

[1550] The specific hardware and software used are as follows:

[1551] SNS API: Twitter API, Facebook Graph API, etc.

[1552] Natural language processing libraries: NLTK, SpaCy

[1553] Sentiment analysis engine: VADER, TextBlob

[1554] Web scraping tool: Beautiful Soup

[1555] RSS Reader:RSS Feed Reader

[1556] Generative AI models: GPT-3, BERT

[1557] Data management tools: SQL databases, NoSQL databases

[1558] Examples include:

[1559] The server collects public tweets mentioning "specific brand X" via the Twitter API, analyzes the posts using the NLTK library, and extracts sentiment data using the VADER engine.

[1560] A post containing the misinformation "Brand X is offered for free" is compared with information on the company's official website and recognized as misinformation. A generative AI model (e.g., GPT-3) is then used to generate a correction, "Brand X's product is not free. It costs $100," and this correction is posted on social media.

[1561] Device:

[1562] The terminal refers to the device of the user who uses SNS and posts publicly. If the user posts incorrect information, they will receive a notification of the correction.

[1563] Examples include:

[1564] When a user tweets "Brand X is being offered for free" on a social networking site, the server collects and analyzes the tweet, and then generates a correction that is sent to the user's account.

[1565] User:

[1566] A user refers to an individual who posts and shares information on a social networking site.

[1567] For example, if a user spreads false information that "Brand X is being offered for free," the user will receive a correction that reads, "Brand X products are not free. They cost $100."

[1568] Prompt Sentence Examples

[1569] Here are some example prompts to input to a generative AI model:

[1570] markdown

[1571] To generate sentiment-based corrections for users, we provide the following information:

[1572] User Information

[1573] Username: User A

[1574] Emotion: Negative

[1575] Misinformation

[1576] Post content: New product from brand X is being offered for free

[1577] It's official: Brand X's new product is $100

[1578] Use the information above to generate a correction in the right tone. For example, "Brand X's new product is not free. It costs $100. Learn more."

[1579] In this way, it is possible to quickly detect misinformation about companies and products on social media and deliver corrections in an appropriate tone based on the sentiment.

[1580] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1581] Step 1:

[1582] The server uses the SNS API to collect public posts containing specific keywords (company names or product names). The input is the keyword and the SNS API token, and the public posts and metadata (post content, user information, posting date and time, number of retweets, number of likes) are obtained by API calls. The output is the collected post data.

[1583] Step 2:

[1584] The server analyzes the collected public posts using a text analysis algorithm (NLTK or SpaCy) to extract posts containing specific keywords. The input is the collected post data, and the text analysis algorithm extracts posts containing keywords. The output is the extracted post data.

[1585] Step 3:

[1586] The server analyzes the extracted posts using an emotion engine (VADER or TextBlob) to obtain emotion data. The input is the extracted post data, and the emotion engine obtains emotion data such as positive, negative, or neutral. The output is the post data with the emotion data added.

[1587] Step 4:

[1588] The server compares the post content with the emotional data added to it with the company's official information. The official information is obtained from web scraping or RSS readers. The input is the post data with the emotional data added and the official information data, and the matching algorithm detects matches or inconsistencies. The output is the matching result.

[1589] Step 5:

[1590] The server flags posts that contain inconsistencies based on the matching results. The input is the matching results, and if there are any inconsistencies, a misinformation flag is added. The output is the post data with the misinformation flag added.

[1591] Step 6:

[1592] Based on posts that the server determines to be misinformation, it uses a generative AI model (such as GPT-3 or BERT) to generate corrections in a tone appropriate to the emotion. The input is post data flagged as misinformation and official information data, and the generative AI model generates corrections in an appropriate tone. The output is the corrections.

[1593] Step 7:

[1594] The server posts the generated correction to SNS and notifies the users who posted the misinformation and those who retweeted it. The input is the correction and the target user information, and the correction is posted via the SNS API. The output is a notification that the post has been completed.

[1595] Step 8:

[1596] The server periodically generates a report containing the collected data, sentiment data, and correction history information, and provides it to the company and the relevant advertising agency. The input is all the data retrieved from the database, and the report is generated by the data management tool. The output is a report document.

[1597] 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.

[1598] 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.

[1599] 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 robot 414.

[1600] 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.

[1601] 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.

[1602] 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.

[1603] 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).

[1604] 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.

[1605] 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."

[1606] 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.

[1607] 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).

[1608] 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.

[1609] 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.

[1610] 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.

[1611] 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.

[1612] 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.

[1613] 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.

[1614] 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.

[1615] 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.

[1616] 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.

[1617] 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.

[1618] The following is further disclosed regarding the above embodiment.

[1619] (Claim 1)

[1620] a means of collecting public posts;

[1621] A means for analyzing the collected public posts and extracting posts containing specific keywords;

[1622] A means of comparing the content of the extracted posts with official company information;

[1623] a means for identifying and flagging misinformation based on the matching results;

[1624] A means for generating correction information for posts determined to be false information;

[1625] A means to post the generated correction information on social media,

[1626] A system including:

[1627] (Claim 2)

[1628] 10. The system of claim 1, further comprising a citation for official information along with the correction information.

[1629] (Claim 3)

[1630] 10. The system of claim 1, further comprising means for generating and providing a history of collected data and correction information, and correction results, in the form of a report.

[1631] "Example 1"

[1632] (Claim 1)

[1633] a means of collecting public posts;

[1634] A means for applying a text analysis algorithm to the collected public posts to extract posts containing specific keywords;

[1635] A means for comparing the content of the extracted posts with official information;

[1636] a means for identifying and flagging misinformation based on the matching results;

[1637] A means for generating correction information containing correct information for a post determined to be false information;

[1638] A means for posting the generated correction information to a social networking service;

[1639] A means to send notifications to users who posted and retweeted misinformation, as appropriate; and

[1640] A system including:

[1641] (Claim 2)

[1642] 10. The system of claim 1, wherein the correction information includes the correct information and the source of the information.

[1643] (Claim 3)

[1644] 10. The system according to claim 1, further comprising means for generating and providing a history of collected data and correction information, and correction results, in the form of a report.

[1645] "Application Example 1"

[1646] (Claim 1)

[1647] a means of collecting public posts;

[1648] A means for analyzing the collected public posts and extracting posts containing specific keywords;

[1649] A means for comparing the content of the extracted posts with official information;

[1650] a means for identifying and flagging misinformation based on the matching results;

[1651] A means for generating correction information for posts determined to be false information;

[1652] a means for posting the generated corrections to an online environment;

[1653] A method to set specific keywords and notify users when misinformation is detected, and

[1654] A means for automatically generating correction information using a generative AI model;

[1655] A means for setting a prompt sentence for the generative AI model;

[1656] A system including:

[1657] (Claim 2)

[1658] 10. The system of claim 1, further comprising a citation for official information along with the correction information.

[1659] (Claim 3)

[1660] 10. The system of claim 1, further comprising means for generating and providing a history of collected data and correction information, and correction results, in the form of a re...

Claims

1. a means of collecting public posts; A means for analyzing the collected public posts and extracting posts containing specific keywords; A means of comparing the content of the extracted posts with official company information; a means for identifying and flagging misinformation based on the matching results; A means for generating correction information for posts determined to be false information; A means to post the generated correction information on social media, A system including:

2. 10. The system of claim 1, further comprising a citation for official information along with the correction information.

3. 10. The system of claim 1, further comprising means for generating and providing a history of collected data and correction information, and correction results, in the form of a report.

Images