System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting and normalizing information from a digital network using generated AI; means for analyzing the collected information using natural language processes and evaluating reliability; means for ranking the analyzed and evaluated information based on criteria; means for displaying the information in a ranking format on user terminals; and means for collecting user evaluation information and distributing revenue to rankers.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] In today's world, there is a huge amount of information available on the Internet, but it is difficult to quickly extract the information needed and reliable from it. Furthermore, aggregation sites often contain advertisements and unreliable information, making it difficult to access accurate information. Furthermore, there is a lack of mechanisms for appropriately distributing revenue based on user ratings. There is a need to provide a system that can solve these issues and enable users to access information efficiently and with high reliability. [Means for solving the problem]

[0005] This invention uses generative AI to normalize information collected from digital networks, analyze it using natural language processing (NLP), and evaluate its reliability. This allows reliable information to be selected from the collected information, ranked, and provided to users. It also collects user evaluation data and builds a system to distribute appropriate revenue to rankers based on that data. Furthermore, it adds a means to score the reliability of collected information using a fact-checking API, and optimizes the ranking algorithm based on user evaluation data, enabling the provision of more accurate information. This system allows users to efficiently access reliable information, and rankers to receive appropriate rewards.

[0006] "Generative AI" refers to artificial intelligence that analyzes collected information and uses natural language processing to understand meaning and context.

[0007] A "digital network" is a type of network that uses information and communications technology to send and receive digital data, including the Internet.

[0008] "Normalization" is the process of cleaning, decluttering, and formatting collected raw data.

[0009] "Natural language processing (NLP)" is a type of technology that allows computers to understand and analyze human language, and includes grammatical analysis and semantic analysis.

[0010] "Credibility assessment" is the process of evaluating and scoring the veracity and reliability of information.

[0011] "Ranking" is the process of ranking information based on specific criteria.

[0012] "User's device" refers to a device operated by a user, such as a computer, smartphone, or tablet.

[0013] "Revenue sharing" is the process of distributing generated revenue to users based on their contribution.

[0014] A "fact-checking API" is an application programming interface that automatically verifies the veracity or reliability of specific information.

[0015] A "ranking algorithm" is a set of computational rules for ranking collected and analyzed information based on specific criteria.

[0016] A "ranker" is someone who creates rankings or uses the system. [Brief explanation of the drawings]

[0017] [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

[0018] 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.

[0019] First, the terms used in the following description will be explained.

[0020] 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).

[0021] 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.

[0022] 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.

[0023] 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.

[0024] 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."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 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.

[0028] 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).

[0029] 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.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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."

[0038] This system uses generative AI to collect large amounts of data, analyzes it using natural language processing (NLP), and evaluates its reliability, allowing users to efficiently gather specific information in a short period of time and obtain highly reliable information in a ranked format.

[0039] Overall system flow

[0040] 1. Information gathering stage

[0041] On the device: The user enters specific information into a search box and submits a request.

[0042] Server: Based on the search request, they gather relevant information from websites, blogs, and social media across their digital network, using crawling and API access to obtain data.

[0043] 2. Data preprocessing stage

[0044] Server: Normalizes the collected information and removes unnecessary elements such as unnecessary HTML tags and advertisements, resulting in pure text data.

[0045] 3. Information Analysis Stage

[0046] Server: Using generative AI, natural language processing is performed to analyze the collected text, extracting important keywords and concepts, and evaluating the reliability of the information using a fact-checking API.

[0047] 4. Ranking Stage

[0048] Server: Analyzes and evaluates the information to generate a ranking according to specific criteria, such as reliability, freshness, and relevance.

[0049] 5. Display Stage

[0050] Device: The generated ranking results are displayed on the user's device, allowing the user to view the ranked, reliable information.

[0051] 6. Rating and Revenue Sharing Phase

[0052] Terminal: The user rates the displayed information, and the results are sent to the server.

[0053] Server: Distributes appropriate revenue to rankers based on user ratings. Collects rating data to help improve the algorithm.

[0054] Specific use cases

[0055] For example, if a user wants to gather information about "the most noteworthy smartphone technologies in 2023," the system works as follows:

[0056] 1. Information Collection:

[0057] A user types "2023 smartphone technology" into the search box and submits a request.

[0058] The server collects relevant information from the Internet.

[0059] 2. Data Preprocessing:

[0060] The server removes unnecessary HTML tags and advertisements from the collected information and converts it into pure text data.

[0061] 3. Information analysis:

[0062] The server uses generative AI to analyze the text and extract key keywords and topics, and it also uses fact-checking APIs to assess the reliability of the information.

[0063] 4. Ranking:

[0064] The server ranks the information based on the analysis and evaluation results, with the most important information being ranked at the top based on criteria such as reliability, recency, and relevance.

[0065] 5.Display:

[0066] The ranking information is displayed on the user's device, allowing the user to grasp important information at a glance.

[0067] 6. Ratings and Revenue Share:

[0068] The user rates the displayed information, and the results of this rating are sent to the server.

[0069] The server distributes revenue to rankers based on the evaluation data and uses it to optimize the algorithm.

[0070] This system allows users to efficiently gather the information they need, provides access to reliable information, and allows ranking creators to receive appropriate rewards.

[0071] The processing flow will be explained below.

[0072] Step 1:

[0073] Request reception

[0074] On the device: A user enters specific information into a search box and submits a search request, which includes search keywords, such as "smartphone technology in 2023."

[0075] Step 2:

[0076] Submitting a search query

[0077] Terminal: Sends the entered search query to the server. Here, the user's input is passed to the server as the search query.

[0078] Step 3:

[0079] Start gathering information

[0080] Server: Based on the received search query, the server launches a web crawler to collect relevant information from websites, blogs, social media, etc. on the Internet. It may also use specific APIs (such as news APIs) to obtain data.

[0081] Step 4:

[0082] Data normalization

[0083] Server: Normalizes the collected raw data. Specifically, it removes unnecessary elements such as unnecessary HTML tags, advertisements, and scripts from the information and extracts only the text information.

[0084] Step 5:

[0085] Text Cleaning

[0086] Server: Further cleans the normalized text, specifically removing line breaks and special characters, and arranging the text data in a consistent format, which improves the accuracy of text analysis.

[0087] Step 6:

[0088] Natural Language Processing

[0089] Server: Analyzes collected text information using generative AI, extracts important keywords and topics, and gains a deep understanding of the text content through grammatical and semantic analysis.

[0090] Step 7:

[0091] Reliability evaluation

[0092] Server: Using fact-checking APIs and proprietary algorithms, the reliability of collected information is evaluated. A reliability score is calculated and information with a low score is removed.

[0093] Step 8:

[0094] Ranking algorithm applied

[0095] Server: Applies a ranking algorithm to rank the information based on criteria such as reliability, recency, and relevance. Sorting is performed based on the score of each piece of information.

[0096] Step 9:

[0097] Saving ranking results

[0098] Server: Stores the generated ranking results in a database for later display and analysis.

[0099] Step 10:

[0100] Sending ranking results

[0101] Server: Sends the ranking results stored on the user's device.

[0102] Step 11:

[0103] User Interface Display

[0104] Terminal: The ranking results are displayed in a user interface designed to allow users to easily access reliable information.

[0105] Step 12:

[0106] Accepting user reviews

[0107] Device: The user rates the displayed information. Specifically, the rating is completed by clicking a rating button such as "useful" or "trustworthy."

[0108] Step 13:

[0109] Submitting evaluation data

[0110] Terminal: Sends user evaluation data to the server, which reflects the user's evaluation in the system.

[0111] Step 14:

[0112] Revenue Share Calculation

[0113] Server: Based on user rating data, distributes appropriate revenue to rankers. Advertising revenue and sponsorship income are calculated according to the ratings, and the corresponding points and revenue are awarded to rankers.

[0114] Step 15:

[0115] Collecting feedback

[0116] Server: Collects and analyzes user evaluation data and behavioral data, which helps identify areas for system improvement.

[0117] Step 16:

[0118] Algorithm Updates

[0119] Server: Optimizes the ranking algorithm based on the collected feedback data. Uses adaptive learning to improve the accuracy of ranking results from next time onwards.

[0120] This system allows users to efficiently gather the information they need and access reliable information, while also allowing rankers to receive appropriate rewards.

[0121] Example 1

[0122] 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."

[0123] Conventional information collection and analysis systems often lack the means to efficiently collect large amounts of information and evaluate it based on its reliability. As a result, it is difficult for users to quickly access the reliable information they need. Furthermore, there are issues with efficiency and accuracy in the process of verifying the accuracy of the collected information and generating rankings based on it. Furthermore, there is a lack of means to optimize the system itself using user evaluation data.

[0124] 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.

[0125] In this invention, the server includes: a means for a user to enter keywords into a search box on a terminal and send a request; a means for collecting information from a digital network based on the search request; a means for normalizing the collected information and removing unnecessary data to convert it into pure text data; a means for analyzing the collected text data using a generative AI model and extracting key keywords and concepts; a means for evaluating the reliability of information using a fact-checking API; a means for ranking information based on the analyzed and evaluated information; a means for displaying the ranking results on the user's terminal; and a means for collecting user evaluation data and distributing revenue to rankers. This enables efficient collection and analysis of large amounts of information, and evaluation and ranking based on reliability. Furthermore, utilizing user evaluation data enables system optimization and enables the provision of more accurate information.

[0126] "User" refers to an entity that uses the system to search for and evaluate information.

[0127] "Device" refers to a device used by a User, such as a computer, smartphone, or tablet.

[0128] A "search box" is an interface element that allows a user to enter search keywords.

[0129] A "request" refers to a search or operation request sent by a user from a terminal to a server.

[0130] A "server" is a collection of hardware and software that receives requests from users and collects and processes information.

[0131] "Digital network" refers to the entire system for electronic information exchange, including the Internet.

[0132] "Means of collecting information" refers to methods of obtaining the necessary data from the Internet using web crawlers or API access.

[0133] "Normalization" refers to the process of removing unnecessary elements from collected data and converting it into a unified format.

[0134] "Unnecessary data" refers to elements of collected information, such as HTML tags and advertisements, that may reduce the accuracy of search results.

[0135] "Pure text data" refers to information after it has been normalized and converted into a parsable text format.

[0136] "Generative AI model" refers to an artificial intelligence model for analyzing collected text data.

[0137] "Means for analyzing text data" refers to a method of using a generative AI model to analyze collected text data and extract key keywords and concepts.

[0138] "Major keywords and concepts" refer to information elements that are particularly important within the text data and serve as the basis for ranking.

[0139] A "fact-checking API" refers to an interface for external services that allow users to verify the reliability of information.

[0140] "Means for evaluating the reliability of information" refers to methods for scoring the veracity and reliability of collected information using fact-checking APIs, etc.

[0141] "Ranking" refers to the process of ranking information according to certain criteria (such as reliability, recency, relevance, etc.) based on evaluated information.

[0142] "Ranking results" refers to a collection of ranked information and a list of ranked information displayed to users.

[0143] "User rating data" refers to the rating a user gives to the displayed information, including feedback such as star ratings and comments.

[0144] "Ranker" refers to the person who creates the ranking or receives compensation based on the ranking.

[0145] "Means of revenue distribution" refers to a method of distributing appropriate rewards to rankers based on user evaluation data.

[0146] MODE FOR CARRYING OUT THE INVENTION

[0147] The system of the present invention is designed to enable users to efficiently collect reliable information. The system automatically performs a series of processes, from processing search requests to collecting, normalizing, analyzing, ranking, and displaying information to users. Furthermore, it collects user evaluation data and uses it for system optimization and revenue distribution.

[0148] Information collection and preprocessing

[0149] When a user enters a specific keyword (e.g., "Smartphone technology in 2023") into the device's search box and presses the search button, the server receives the request. Based on the request, the server uses a web crawler (e.g., Scrapy) to retrieve information from digital networks. It also collects the latest information from social media and search engines through API access (e.g., Twitter API, Google® Custom Search API).

[0150] The collected information is normalized on the server using libraries such as Python's BeautifulSoup, which removes unnecessary HTML tags and advertisements and converts it into pure text data.

[0151] Information analysis

[0152] The normalized text data is analyzed by a generative AI model (for example, GPT-4 (registered trademark) by OpenAI (registered trademark)). The generative AI model is used to extract key keywords and concepts from the collected text data. At the same time, the reliability of the information is evaluated using a fact-checking API (for example, Factmata API). The fact-checking API is used to score the veracity and reliability of the information.

[0153] Ranking and Display

[0154] Once the information has been analyzed and evaluated, it is ranked based on criteria such as reliability, recency, and relevance. In this ranking process, an overall score is assigned to each piece of information, and the information is sorted in descending order of its score. The ranking results are stored in a database on the server and displayed on the user's device. Users can use this ranking to quickly access the information they need.

[0155] To give a concrete example, let's say a user searches for information on "smartphone technologies to watch in 2023." Here's how the system works:

[0156] 1. A user types "2023 smartphone technology" into the search box and submits a request.

[0157] 2. The server collects relevant information from the Internet.

[0158] 3. The server removes unnecessary HTML tags and advertisements from the collected information and converts it into pure text data.

[0159] 4. The server analyzes the text using a generative AI model (e.g., GPT-4) to extract key keywords and topics, and also uses a fact-checking API to assess the reliability of the information.

[0160] 5. The server ranks the information based on the analysis and evaluation results. The most important information is ranked at the top based on criteria such as reliability, recency, and relevance.

[0161] 6. Ranking information is displayed on the user's device, allowing the user to grasp important information at a glance.

[0162] An example of a prompt sentence might be:

[0163] "Please explain the specific information gathering procedures for the system that allows users to gather information about 'notable smartphone technologies in 2023.'"

[0164] This system allows users to efficiently gather the information they need and quickly access reliable information. Furthermore, by using user evaluation data, the system can be optimized to provide even higher quality information.

[0165] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0166] Step 1: Receiving a search request

[0167] Device: A user enters a specific keyword (e.g., "smartphone technology 2023") into the device's search box and presses the search button. This action generates a search request.

[0168] Input: The search keyword entered by the user.

[0169] Output: The search request sent to the server.

[0170] Step 2: Gather information

[0171] Server: Based on the received search request, launch a web crawler (e.g., Scrapy) to collect relevant information from the Internet, and then use API access (e.g., Twitter API, Google Custom Search API) to obtain relevant data.

[0172] Input: Keywords based on your search request.

[0173] Output: Collected data from web pages and APIs.

[0174] Step 3: Normalize the data

[0175] Server: To normalize the collected information, we use Python's BeautifulSoup library to remove unnecessary HTML tags and advertisements and convert it into pure text data.

[0176] Input: Raw data collected from web pages and APIs.

[0177] Output: Cleaned and normalized text data.

[0178] Step 4: Analyzing the text data

[0179] Server: Analyzes the normalized text data using a generative AI model (e.g., OpenAI's GPT-4), extracting key keywords and concepts.

[0180] Input: Normalized text data.

[0181] Output: Extracted keywords and concepts.

[0182] Step 5: Assess reliability

[0183] Server: Use a fact-checking API (e.g., Factmata API) to evaluate the reliability of the extracted information. Using this API, the truthfulness and reliability score of the information can be obtained.

[0184] Input: Keywords and concepts extracted by the generative AI model.

[0185] Output: Evaluation results including confidence scores.

[0186] Step 6: Ranking the information

[0187] Server: Based on the analyzed and evaluated information, the information is ranked based on specific criteria (reliability, freshness, relevance, etc.) and the rankings are stored in a SQL database or similar.

[0188] Input: Confidence score and evaluation results.

[0189] Output: A list of information in ranking format.

[0190] Step 7: View ranking information

[0191] Device: Ranking information is displayed on the user's device, allowing users to quickly access important information.

[0192] Input: A ranking list of information sent from the server.

[0193] Output: A viewable list of information displayed on the user's terminal.

[0194] Step 8: Collecting ratings and sharing revenue

[0195] Terminal: The user rates the displayed information, and the results are sent to the server.

[0196] Input: User rating data (star ratings, comments, etc.).

[0197] Output: The rating data sent to the server.

[0198] Server: Receives user evaluation data and distributes revenue to rankers based on that data. Additionally, the collected evaluation data is used to optimize the system.

[0199] Input: User rating data.

[0200] Output: Revenue distribution results to rankers and algorithm improvements using evaluation data.

[0201] In this way, the system performs specific data processing and calculations at each step, enabling users to access reliable information in a short time, and also optimizes the system itself by utilizing user evaluation data.

[0202] (Application example 1)

[0203] 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."

[0204] A major issue facing modern online shopping sites is the low reliability of product reviews. Many reviews are inaccurate or exaggerated, making it difficult for users to make appropriate decisions. This has created a need for users to efficiently find reliable reviews. Furthermore, existing ranking systems tend to rely on subjective evaluations and lack objective evaluations of reliability. This has led to major challenges in improving the reliability of reviews and user satisfaction.

[0205] 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.

[0206] In this invention, the server includes means for collecting and normalizing information from a digital network using a generative AI, means for analyzing the collected information using natural language processing and evaluating its reliability, means for ranking the analyzed and evaluated information based on criteria, means for displaying the information in a ranking format on the user's terminal, means for collecting user evaluation data and distributing revenue to rankers, and means for collecting product reviews and displaying highly reliable reviews at the top, thereby enabling users to efficiently access highly reliable product reviews.

[0207] "Generative AI" refers to generative artificial intelligence, a technology that generates new information and answers based on large amounts of data.

[0208] "Normalization" refers to the process of removing unnecessary information and formats from collected data.

[0209] "Natural language processing" refers to the technology that enables computers to understand, analyze, and process human language.

[0210] "Assessing trustworthiness" is the process of determining how accurate or relevant the information collected is.

[0211] "Ranking" refers to ranking data according to specific criteria.

[0212] "User's terminal" refers to a device used by a user, such as a smartphone or computer.

[0213] "User Ratings Data" means feedback and rating information provided by Users.

[0214] A "ranker" refers to the person or system that creates rankings based on collected information and evaluation data.

[0215] "Means for distributing revenue" refers to a mechanism for appropriately distributing revenue generated based on the generated information and rankings.

[0216] "Product reviews" refer to the impressions and evaluation data left by users after using a product.

[0217] A "reliable review" is a review that contains objective and accurate information and is deemed to be useful to other users.

[0218] This invention is a system that uses generative AI technology to collect reliable product reviews and provide them to users in a ranking format. This system consists of the following steps: information collection, data preprocessing, information analysis, reliability evaluation, ranking generation, display, user evaluation, and revenue distribution.

[0219] The server first uses generative AI to collect product review information from digital networks and normalizes it by removing unnecessary elements such as HTML tags and advertisements. Libraries such as BeautifulSoup and requests are used in this process. The collected data is then analyzed using natural language processing (NLP) to extract important keywords and topics. Software such as SpaCy and transformers are used in this stage.

[0220] The server evaluates the reliability of the reviews based on the analysis results and assigns them a score using a fact-checking API. Based on the obtained score, the reviews are ranked according to criteria. Data analysis tools such as scikit-learn are used to generate the rankings. Once the reliability evaluation and ranking are complete, the ranking results are sent to the user's device, where they can be viewed on their smartphone or other device.

[0221] Users rate the displayed ranking results, and the rating data is sent to the server. The server then optimizes the ranking algorithm based on the user rating data and distributes revenue to the rankers. As a concrete example, when a user searches for "latest smartphones," reliable reviews are displayed at the top. For example, specific reviews such as "This smartphone has a long battery life" are displayed in a prominent position.

[0222] An example of a prompt is, "Please collect reviews of the latest smartphones, rate their reliability, and rank them." By using such a prompt, users can efficiently obtain the information they need.

[0223] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0224] Step 1:

[0225] Information gathering

[0226] The server collects relevant review information from the digital network based on the keywords entered by the user in the search box. Specifically, it retrieves the information from the relevant product's web page or review site using the requests library. The retrieved information is saved on the server in HTML format.

[0227] Input: User's search keyword

[0228] Output: Collected review information in HTML format

[0229] Step 2:

[0230] Data Preprocessing

[0231] The server normalizes the collected HTML review information by using the BeautifulSoup library to remove unnecessary elements such as HTML tags and advertisements, converting it into pure text data, and removing line spacing, whitespace, and special characters.

[0232] Input: Collected review information in HTML format

[0233] Output: Normalized text data

[0234] Step 3:

[0235] Natural Language Processing (NLP) Analysis

[0236] The server then performs natural language processing on the normalized text data using SpaCy or transformers, extracting important keywords and topics from the reviews and calculating features (such as TF-IDF) for each review.

[0237] Input: normalized text data

[0238] Output: Extracted keywords, topics, and features

[0239] Step 4:

[0240] Reliability evaluation

[0241] The server uses a fact-checking API to score the credibility of each review. Specifically, it sends the analyzed data to the fact-checking API and receives a credibility score, which quantifies the credibility of each review.

[0242] Input: Extracted keywords, topics, and features

[0243] Output: Confidence score

[0244] Step 5:

[0245] Ranking Generation

[0246] The server ranks reviews based on their credibility score. The ranking algorithm uses factors such as credibility score, recency, and relevance. The ranking is generated using libraries such as scikit-learn.

[0247] Inputs: Confidence score, recency, relevance

[0248] Output: Ranked reviews

[0249] Step 6:

[0250] Displaying ranking results

[0251] The device displays the ranking results sent from the server to the user, who can then view the reviews ranked in order of reliability on their smartphone or computer screen.

[0252] Input: Ranked reviews

[0253] Output: Ranking results displayed on the user's device

[0254] Step 7:

[0255] User ratings and data collection

[0256] Users rate the displayed ranking results, and the rating data is sent from the device to the server, including feedback on the reliability and usefulness of the reviews.

[0257] Input: User Rating

[0258] Output: Rating data sent to the server

[0259] Step 8:

[0260] Revenue sharing and algorithm optimization

[0261] The server optimizes the ranking algorithm based on the evaluation data collected from users, and also executes the process of distributing profits to rankers. Appropriate data analysis is performed to improve the accuracy of the ranking algorithm and distribute profits.

[0262] Input: User rating data

[0263] Output: Optimized ranking algorithm, revenue sharing results

[0264] 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.

[0265] This invention provides a system that combines generative AI and an emotion engine to efficiently collect, analyze, assess credibility, rank, evaluate users, and distribute revenue. Furthermore, by analyzing user emotions, it optimizes the ranking algorithm and dynamically adjusts the content displayed.

[0266] Overall system flow

[0267] 1. Information gathering stage

[0268] On the device: A user types specific information (e.g., "smartphone technology in 2023") into a search box and submits a request.

[0269] Server: Based on the search request, they gather relevant information from websites, blogs, and social media across their digital network, using crawling and API access to obtain data.

[0270] 2. Data preprocessing stage

[0271] Server: Normalizes the collected information and removes unnecessary elements such as unnecessary HTML tags and advertisements, resulting in pure text data.

[0272] 3. Information Analysis Stage

[0273] Server: Using generative AI, natural language processing is performed to analyze the collected text, extracting important keywords and topics, and evaluating the reliability of the information using fact-checking APIs.

[0274] 4. Ranking Stage

[0275] Server: Analyzes and evaluates the information to generate a ranking according to specific criteria, such as reliability, freshness, and relevance.

[0276] 5. Display Stage

[0277] Device: The generated ranking results are displayed on the user's device, allowing the user to view the ranked, reliable information.

[0278] 6. Rating and Revenue Sharing Phase

[0279] Terminal: The user rates the displayed information. The rating data is sent to the server.

[0280] Server: Distributes appropriate revenue to rankers based on user rating data. Collects rating data to help improve the algorithm.

[0281] 7. Sentiment Analysis Stage

[0282] Device: An emotion engine that recognizes the user's emotions runs and collects the user's emotion data.

[0283] Example: An emotion engine analyzes in real time whether a user is very interested, angry, sad, etc.

[0284] 8. Sentiment data analysis and algorithm optimization

[0285] Server: The ranking algorithm is optimized based on the collected emotional data, such as giving priority to information where the user expressed positive emotions.

[0286] Example: If the emotion engine analyzes the user's emotion as "interesting," it will rank similarly interesting information higher.

[0287] 9. Dynamic Content Adjustment

[0288] On the device: Dynamically adjust the content displayed based on the analysis results of the emotion engine, prioritizing the display of topics and related information that interest the user.

[0289] Example: If a user is determined to be interested in smartphone technology, the app will automatically show them new related articles and videos.

[0290] Specific use cases

[0291] For example, if a user wants to gather information about "the most noteworthy smartphone technologies in 2023," the system works as follows:

[0292] 1. Information Collection:

[0293] A user types "2023 smartphone technology" into the search box and submits a request.

[0294] The server collects relevant information from the Internet.

[0295] 2. Data Preprocessing:

[0296] The server removes unnecessary HTML tags and advertisements from the collected information and converts it into pure text data.

[0297] 3. Information analysis:

[0298] The server uses generative AI to analyze the text and extract key keywords and topics, and it also uses fact-checking APIs to assess the reliability of the information.

[0299] 4. Emotion analysis:

[0300] While users browse information, the emotion engine analyzes their emotions in real time and changes relevant information if the user is interested or dissatisfied.

[0301] 5. Ranking:

[0302] The server ranks the information based on the analysis and evaluation results, incorporating factors such as reliability, recency, and relevance, as well as sentiment data, to rank the most important information at the top.

[0303] 6. Dynamic content adjustment:

[0304] The content displayed is dynamically adjusted based on the user's emotional data, prioritizing new and relevant information that will interest the user.

[0305] 7. Ratings and Revenue Share:

[0306] Users rate the displayed information, and the results are sent to the server, which distributes revenue to rankers based on the rating data and uses it to optimize the algorithm.

[0307] This system allows users to efficiently gather the information they need and access reliable information. By taking user sentiment into consideration, it also enables more personalized information to be provided, and rankers can receive appropriate rewards.

[0308] The processing flow will be explained below.

[0309] Step 1:

[0310] Request reception

[0311] On the device: A user enters specific information into a search box and submits a search request, which includes search keywords, such as "smartphone technology in 2023."

[0312] Example: A user searches for "smartphone technology 2023."

[0313] Step 2:

[0314] Submitting a search query

[0315] Terminal: Sends the entered search query to the server.

[0316] Step 3:

[0317] Start gathering information

[0318] Server: Based on the received search query, the server launches a web crawler to collect relevant information from websites, blogs, social media, etc. on the Internet. It may also use specific APIs (such as news APIs) to obtain data.

[0319] Example: The server crawls news sites and social media sites using the keyword "smartphone technology 2023" to collect related information.

[0320] Step 4:

[0321] Data normalization

[0322] Server: The collected raw data is normalized. Specifically, unnecessary elements such as unnecessary HTML tags, advertisements, and scripts are removed from the information, and only the text information is extracted.

[0323] Example: Removing advertisements and unnecessary HTML tags from aggregated news articles and blog posts.

[0324] Step 5:

[0325] Text Cleaning

[0326] Server: Further cleans the normalized text, specifically removing line breaks and special characters, and arranging the text data in a consistent format, which improves the accuracy of text analysis.

[0327] For example: cleaning text by removing unnecessary spaces and special characters.

[0328] Step 6:

[0329] Natural Language Processing

[0330] Server: Analyzes collected text information using generative AI, extracts important keywords and topics, and gains a deep understanding of the text content through grammatical and semantic analysis.

[0331] Example: Extracting keywords related to smartphone technology (5G, camera resolution, battery life, etc.).

[0332] Step 7:

[0333] Reliability evaluation

[0334] Server: Using fact-checking APIs and proprietary algorithms, the server evaluates the reliability of collected information, calculates a reliability score, and filters out information with a low score.

[0335] Example: Calculate a credibility score for each source and filter out information with a low score.

[0336] Step 8:

[0337] Ranking algorithm applied

[0338] Server: Applies a ranking algorithm to rank the information based on criteria such as reliability, recency, and relevance. Sorting is performed based on the score of each piece of information.

[0339] Example: Combining reliability, recency, and relevance scores and ranking the information with the highest score at the top.

[0340] Step 9:

[0341] Saving ranking results

[0342] Server: Stores the generated ranking results in a database for later display and analysis.

[0343] Example: Storing ranked smartphone technology information in a database.

[0344] Step 10:

[0345] Sending ranking results

[0346] Server: Sends the ranking results stored on the user's device.

[0347] Example: Sending information about ranked smartphone technologies to the user's device.

[0348] Step 11:

[0349] User Interface Display

[0350] Terminal: The ranking results are displayed on the user interface, allowing users to access reliable information through the displayed rankings.

[0351] Example: A smartphone screen displays a list of reliable smartphone technology information.

[0352] Step 12:

[0353] Accepting user reviews

[0354] Device: The user rates the displayed information. Specifically, the rating is completed by pressing a rating button such as "useful" or "trustworthy."

[0355] Example: A user rates this review as helpful.

[0356] Step 13:

[0357] Submitting evaluation data

[0358] Terminal: Sends user evaluation data to the server, which reflects the user's evaluation in the system.

[0359] Example: User rating data is sent to the server.

[0360] Step 14:

[0361] Revenue Share Calculation

[0362] Server: Based on user rating data, distributes appropriate revenue to rankers. Advertising revenue and sponsorship income are calculated according to the ratings, and the corresponding points and revenue are awarded to rankers.

[0363] Example: Revenues are distributed to rankers based on user ratings.

[0364] Step 15:

[0365] Collecting feedback

[0366] Server: Collects and analyzes user evaluation data and behavioral data, which helps identify areas for system improvement.

[0367] Example: Collecting evaluation data to help improve the system.

[0368] Step 16:

[0369] Algorithm Updates

[0370] Server: Optimizes the ranking algorithm based on the collected feedback data. Uses adaptive learning to improve the accuracy of ranking results from next time onwards.

[0371] Example: Learn the characteristics of highly rated articles and reflect them in future rankings.

[0372] Step 17:

[0373] Emotion analysis

[0374] On-device: Utilizing an emotion engine to recognize the user's emotional state in real time, including interest, excitement, anger, sadness, etc.

[0375] Example: An emotion engine analyzes user emotions in real time while the user is browsing information.

[0376] Step 18:

[0377] Sending emotional data

[0378] Terminal: Sends the user's emotional data to the server.

[0379] Example: The analyzed emotion data is sent to the server.

[0380] Step 19:

[0381] Sentiment data analysis and algorithm optimization

[0382] Server: The ranking algorithm is optimized based on the collected emotional data, such as giving priority to information where the user expressed positive emotions.

[0383] For example, if the emotion engine analyzes the user's emotion as "interesting," it will rank similarly interesting information higher.

[0384] Step 20:

[0385] Dynamic Content Adjustment

[0386] On the device: Dynamically adjust the content displayed based on the analysis results of the emotion engine, prioritizing the display of topics and related information that interest the user.

[0387] Example: If a user expresses an interest in smartphone technology, automatically show them new related articles and videos.

[0388] This system allows users to efficiently gather the information they need and access reliable information. By taking user sentiment into consideration, it also enables more personalized information to be provided, and rankers can receive appropriate rewards.

[0389] Example 2

[0390] 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."

[0391] Conventional information gathering and organization systems have a process for ranking collected information based on its reliability, recency, and relevance, but they lack optimization that takes into account user emotions and evaluation data, limiting their ability to improve user experience. In particular, they are unable to reflect users' emotional reactions to content evaluations in real time, making it difficult to provide personalized information. Furthermore, there are also insufficient means for automatically evaluating and scoring the reliability of collected information, creating a risk of misinformation being included.

[0392] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0393] In this invention, the server includes means for collecting and normalizing information from a digital network using generative AI, means for analyzing the collected information using natural language processing and evaluating its reliability, means for ranking the analyzed and evaluated information based on criteria, means for displaying the information in a ranking format on the user's terminal, means for collecting user evaluation data and distributing revenue to rankers, means for analyzing user emotions using an emotion engine, means for optimizing the ranking algorithm based on the analyzed emotion data, and means for dynamically adjusting the content to be displayed based on the emotion data. This enables the provision of personalized information based on the user's emotion and evaluation data, resulting in more reliable information and an improved user experience.

[0394] "Generative AI" is a type of artificial intelligence technology used for natural language processing and data analysis, and has the ability to generate, convert, and analyze information from a wide range of data.

[0395] "Digital network" refers to the Internet and other digital communications networks that enable the collection, sharing, and transfer of information.

[0396] "Normalization" refers to the process of removing unnecessary data (e.g., HTML tags and advertisements) from collected information and converting it into pure text data.

[0397] "Natural language processing" refers to the technology of analyzing collected text data, extracting keywords and topics, and understanding the meaning and relationships of the information.

[0398] "Credibility assessment" is the process of assessing whether collected information is accurate, and includes assigning a credibility score using fact-checking APIs, etc.

[0399] "Ranking" refers to the process of analyzing and evaluating information and ranking it based on certain criteria (reliability, recency, relevance).

[0400] "User rating data" refers to data including ratings (e.g., star ratings, comments, etc.) made by users on information.

[0401] "Revenue sharing" refers to the process of distributing appropriate rewards to information providers based on users' rating data.

[0402] An "emotion engine" refers to technology for analyzing emotions from a user's facial expressions, voice, etc., and is used to collect and analyze emotional data in real time.

[0403] "Dynamic adjustment" refers to the process of changing and adjusting displayed content in real time based on the results of user sentiment analysis.

[0404] A "fact-checking API" is an application program interface used to assess the reliability of collected information and is a tool for fact-checking information.

[0405] This invention provides a system that combines generative AI and an emotion engine to efficiently collect, analyze, assess reliability, rank, and distribute revenues to users. Additionally, analyzing user emotions can optimize the ranking algorithm and dynamically adjust the content displayed. Below, we will explain each element of the system and how they are combined.

[0406] First, the device accepts the user's search request. The device has a search box, and the user enters the keyword of the information they want to find (e.g., "Smartphone technology in 2023") into this search box. The device then sends the search request to the server using the HTTP protocol.

[0407] Based on the received search request, the server collects relevant information from websites, blogs, social media, etc. on the digital network. This information collection is done using crawling tools such as Apache (registered trademark) Nutch and services such as the Twitter API.

[0408] The server then normalizes the collected information by using an HTML parser such as BeautifulSoup to remove unnecessary HTML tags and advertisements from the collected web pages and convert them into pure text data.

[0409] A generative AI model (e.g., GPT-4) runs on a server and analyzes the collected and normalized text data. During this analysis, natural language processing techniques are used to extract key keywords and topics, and fact-checking APIs (e.g., Snopes API) are used to assess the reliability of the information.

[0410] The server then generates a ranking based on the analyzed and evaluated information according to certain criteria (authority, recency, relevance), using a modified version of the PageRank algorithm or a recommendation system for this ranking process.

[0411] The ranking results are displayed on the user's device. The data is presented in a visually easy-to-read format using HTML and CSS. For example, the "latest articles about AI camera technology" could be displayed at the top.

[0412] Users rate the displayed information (e.g., star ratings or comments). The rating data is sent to the server, which then distributes appropriate revenue to the rankers based on the rating data. This rating data is also used to optimize the algorithm.

[0413] The user's device is equipped with an emotion engine, which uses technologies such as Emotion API to collect emotional data in real time from the user's facial expressions and voice. For example, it analyzes whether the user is expressing the emotion of "joy."

[0414] The server optimizes the ranking algorithm based on the collected sentiment data, for example, adjusting the algorithm to prioritize ranking information that users find "interesting," thereby improving the user experience.

[0415] Furthermore, the content displayed can be dynamically adjusted based on emotional data. For example, if it is determined that a user is interested in "AI camera technology," new related articles and videos will be dynamically displayed.

[0416] This system allows users to efficiently gather the information they need and access reliable information. It also enables personalized information provision that takes into account the user's emotions, and information providers can receive appropriate rewards.

[0417] Specific prompt examples:

[0418] "Please explain in detail the process flow for a user wanting to gather information about smartphone technology in 2023."

[0419] By combining generative AI models and emotion analysis technology, this system can respond quickly and appropriately to diverse user needs, significantly improving the quality and efficiency of information provision.

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

[0421] Step 1:

[0422] User: Enters "Smartphone technology in 2023" into the search box on the device and sends a search request, which generates search keywords as input data and sends them to the server using the HTTP protocol.

[0423] Input: Search keyword "Smartphone technology in 2023"

[0424] Output: Search request

[0425] Step 2:

[0426] Server: Based on the received search request, the server uses crawling tools such as Apache Nutch and Twitter APIs to collect relevant information from websites, blogs, and social media on the digital network. The collected information is stored as text data.

[0427] Input: Search request

[0428] Output: Collected information (text data)

[0429] Step 3:

[0430] Server: Normalize the collected information. Use an HTML parser such as BeautifulSoup to remove unnecessary HTML tags and advertisements and convert it into pure text data. This data processing results in clean text data.

[0431] Input: Collected information (text data)

[0432] Output: Normalized text data

[0433] Step 4:

[0434] Server: Uses a generative AI model (e.g., GPT-4) to extract key keywords and topics from the normalized text data, and uses fact-checking APIs (e.g., Snopes API) to assess the reliability of the information, resulting in analytics data and a reliability score.

[0435] Input: normalized text data

[0436] Output: Analysis data (major keywords, topics), reliability score

[0437] Step 5:

[0438] Server: Based on the analyzed information and the reliability score, it generates a ranking according to certain criteria (reliability, recency, relevance). This ranking is done using a modified version of the PageRank algorithm or a recommendation system.

[0439] Input: Analysis data (main keywords, topics), reliability score

[0440] Output: Ranking results

[0441] Step 6:

[0442] Device: The ranking results are visually formatted using HTML and CSS and displayed to the user. Users can view the ranking information on their devices.

[0443] Input: Ranking results

[0444] Output: Visualized ranking results (HTML, CSS)

[0445] Step 7:

[0446] User: Rates the displayed information (e.g., star ratings or comments). The rating data is sent to the server using the HTTP protocol.

[0447] Input: Visualized ranking results

[0448] Output: Rating data (star ratings, comments)

[0449] Step 8:

[0450] Server: Based on the collected evaluation data, the server distributes appropriate revenue to rankers. This evaluation data is also used to optimize the ranking algorithm.

[0451] Input: Evaluation data

[0452] Output: Revenue share data, optimized ranking algorithm

[0453] Step 9:

[0454] Device: Runs the emotion engine and collects emotional data from the user's facial expressions and voice. For example, it uses the Emotion API to analyze emotions such as "happiness," "interest," and "dissatisfaction" in real time.

[0455] Input: User's facial expressions and voice

[0456] Output: Emotion data

[0457] Step 10:

[0458] Server: The ranking algorithm is further optimized based on the data obtained from the sentiment analysis. For example, the algorithm is adjusted to prioritize ranking information that users find "interesting."

[0459] Input: Emotion data

[0460] Output: Optimized ranking algorithm

[0461] Step 11:

[0462] On-device: Dynamically adjusts content displayed based on sentiment analysis. For example, if a user is determined to be interested in "AI camera technology," relevant new articles and videos will be displayed.

[0463] Input: Sentiment analysis results

[0464] Output: Dynamically adjusted content placement

[0465] (Application example 2)

[0466] 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."

[0467] Conventional information gathering and ranking systems have had difficulty providing dynamic content that takes user sentiment into account, resulting in a lack of optimization for the user experience. Furthermore, while there is a demand for reliable information, there are limited methods for evaluating the accuracy and reliability of collected information in real time. Furthermore, while there is a demand for quickly optimizing ranking algorithms using user evaluation data and providing personalized information, there has been no system that combines sentiment analysis.

[0468] 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.

[0469] In this invention, the server includes means for collecting and normalizing information from a digital network using generative AI, means for analyzing the collected information using natural language processing and evaluating its reliability, means for ranking the analyzed and evaluated information based on criteria, means for displaying the information in a ranking format on the user's terminal, means for collecting user evaluation data and distributing revenue to rankers, means for analyzing user emotions in real time using a sentiment analysis engine, means for optimizing the ranking algorithm based on the collected sentiment data, and means for dynamically adjusting the content to be displayed. This makes it possible to provide efficient information taking user emotions into consideration, identify reliable information in a ranking format, and dynamically provide personalized information.

[0470] "Generative AI" is an artificial intelligence technology that collects information from digital networks and analyzes and generates data using natural language processing.

[0471] "Normalization" is the process of removing unnecessary elements from collected data and converting it into pure text data that is easy to analyze.

[0472] "Natural language processing" is a technology that allows computers to understand and analyze human language.

[0473] "Assessing credibility" is a method for determining the legitimacy or reliability of collected information.

[0474] "Ranking" refers to the process of ranking collected and analyzed information according to specific criteria.

[0475] An "emotion analysis engine" is a technology that analyzes user emotions in real time and collects them as data.

[0476] "Dynamic adjustment" is the process of instantly changing and updating displayed content based on data collected in real time.

[0477] "Distributing revenue" means distributing appropriate rewards to content providers and others based on user evaluation data.

[0478] This invention is a system that combines generative AI and a sentiment analysis engine to efficiently collect, analyze, assess credibility, rank, evaluate users' ratings, analyze sentiment, and dynamically adjust content. The following hardware and software are required for implementation:

[0479] Hardware and software used

[0480] Hardware: Camera (to capture the user's facial expressions), microphone (to capture the user's voice), user device (such as a smartphone or tablet).

[0481] Software: Generative AI model (for information analysis), TextBlob (for natural language processing), TENSORFLOW (registered trademark) / Keras (for sentiment analysis model), REST API (for information collection).

[0482] What the program does

[0483] The server uses generative AI to collect information from digital networks, obtains data through crawling and API access, and normalizes this information, removing unnecessary HTML tags and advertisements, converting it into pure text data.

[0484] Furthermore, the collected text data is analyzed using natural language processing software (such as TextBlob), important keywords and topics are extracted, and the reliability of the information is evaluated through a fact-checking API. The evaluated information is ranked according to specific criteria and displayed in a ranked format on the user's device.

[0485] User sentiment is analyzed in real time using a sentiment analysis engine (such as TensorFlow / Keras models), and the ranking algorithm is optimized based on the collected sentiment data. The content displayed is dynamically adjusted according to the user's interests and sentiment.

[0486] Specific examples

[0487] For example, consider a case where a user wants to gather information about the latest VR technology. The user enters keywords into the search box, and the system collects related information from the internet. The collected information is then analyzed using natural language processing to extract important keywords and topics. The reliability of the information is evaluated based on a fact-checking API, and the results are displayed to the user in a ranked format.

[0488] While users are browsing information, the sentiment analysis engine analyzes their facial expressions and voice in real time. Information that is determined to be of interest to the user is displayed preferentially, and related information is dynamically adjusted, allowing users to quickly and efficiently obtain the content that interests them most.

[0489] Prompt Sentence Examples

[0490] "Analyze the following text and extract key keywords and sentiment:"

[0491] Text to analyze: "I'm looking for information on the latest in VR technology."

[0492] Using this prompt, the generative AI model analyzes the text data and provides information according to the user's request, which is expected to significantly improve the user experience.

[0493] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0494] Step 1:

[0495] The user enters a search keyword from the terminal and sends a request.

[0496] Input: User's search keyword

[0497] Output: Search request

[0498] How it works: A user enters a keyword into the search box on their device and presses the submit button, which sends the keyword to the system.

[0499] Step 2:

[0500] The server uses generated AI to collect information related to the search keywords from the Internet.

[0501] Input: Search request

[0502] Output: Collected digital information (web pages, blogs, social media posts, etc.)

[0503] How it works: The server collects relevant information from various sources on the Internet (websites and APIs) and stores it in text format.

[0504] Step 3:

[0505] The server normalizes the collected information and removes unnecessary HTML tags and advertisements.

[0506] Input: Collected digital information

[0507] Output: Clean text data

[0508] How it works: The normalization process removes unnecessary elements such as HTML tags and advertisements, producing pure text data.

[0509] Step 4:

[0510] The server uses natural language processing to analyze the clean text data and extract important keywords and topics.

[0511] Input: clean text data

[0512] Output: Analysis results (important keywords, extracted topics)

[0513] What it does: It uses a natural language processing library such as TextBlob to parse the text and extract key keywords and topics.

[0514] Step 5:

[0515] The server uses a fact-checking API to assess the reliability of the collected information.

[0516] Input: clean text data

[0517] Output: Trustworthiness score

[0518] How it works: Scores the legitimacy and veracity of collected information through a fact-checking API.

[0519] Step 6:

[0520] The server analyzes and evaluates the information and then ranks it according to specific criteria.

[0521] Input: Analysis results, reliability evaluation score

[0522] Output: Ranked list of information

[0523] How it works: Applies an algorithm that ranks information based on criteria such as reliability, freshness, and relevance.

[0524] Step 7:

[0525] The terminal provides the user with information displayed in a ranking format.

[0526] Input: Ranked list of information

[0527] Output: Ranking information displayed in the user interface

[0528] How it works: The ranking results are displayed on the user's screen, allowing the user to visually confirm reliable information.

[0529] Step 8:

[0530] The device uses an emotion analysis engine to analyze the user's facial expressions and voice in real time.

[0531] Input: User's facial expression data, voice speech data

[0532] Output: Sentiment analysis results (e.g., interest, joy, dissatisfaction)

[0533] How it works: The camera and microphone are used to capture the user's facial expressions and voice, which are then analyzed in real time by an emotion analysis engine.

[0534] Step 9:

[0535] The server optimizes the ranking algorithm based on the collected sentiment data.

[0536] Input: Sentiment analysis results, ranking information list

[0537] Output: Optimized ranking information

[0538] How it works: Reflects user sentiment data and adjusts and optimizes the ranking algorithm.

[0539] Step 10:

[0540] The device dynamically adjusts the content displayed based on the user's interests and emotions.

[0541] Input: Optimized ranking information, sentiment analysis results

[0542] Output: Dynamically adjusted content

[0543] How it works: Changes the information shown based on your interests and automatically surfaces new relevant content and videos.

[0544] 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.

[0545] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0546] 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.

[0547] [Second embodiment]

[0548] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0549] 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.

[0550] 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).

[0551] 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.

[0552] 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.

[0553] 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).

[0554] 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.

[0555] 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.

[0556] 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.

[0557] 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.

[0558] 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.

[0559] 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."

[0560] This system uses generative AI to collect large amounts of data, analyzes it using natural language processing (NLP), and evaluates its reliability, allowing users to efficiently gather specific information in a short period of time and obtain highly reliable information in a ranked format.

[0561] Overall system flow

[0562] 1. Information gathering stage

[0563] On the device: The user enters specific information into a search box and submits a request.

[0564] Server: Based on the search request, they gather relevant information from websites, blogs, and social media across their digital network, using crawling and API access to obtain data.

[0565] 2. Data preprocessing stage

[0566] Server: Normalizes the collected information and removes unnecessary elements such as unnecessary HTML tags and advertisements, resulting in pure text data.

[0567] 3. Information Analysis Stage

[0568] Server: Using generative AI, natural language processing is performed to analyze the collected text, extracting important keywords and concepts, and evaluating the reliability of the information using a fact-checking API.

[0569] 4. Ranking Stage

[0570] Server: Analyzes and evaluates the information to generate a ranking according to specific criteria, such as reliability, freshness, and relevance.

[0571] 5. Display Stage

[0572] Device: The generated ranking results are displayed on the user's device, allowing the user to view the ranked, reliable information.

[0573] 6. Rating and Revenue Sharing Phase

[0574] Terminal: The user rates the displayed information, and the results are sent to the server.

[0575] Server: Distributes appropriate revenue to rankers based on user ratings. Collects rating data to help improve the algorithm.

[0576] Specific use cases

[0577] For example, if a user wants to gather information about "the most noteworthy smartphone technologies in 2023," the system works as follows:

[0578] 1. Information Collection:

[0579] A user types "2023 smartphone technology" into the search box and submits a request.

[0580] The server collects relevant information from the Internet.

[0581] 2. Data Preprocessing:

[0582] The server removes unnecessary HTML tags and advertisements from the collected information and converts it into pure text data.

[0583] 3. Information analysis:

[0584] The server uses generative AI to analyze the text and extract key keywords and topics, and it also uses fact-checking APIs to assess the reliability of the information.

[0585] 4. Ranking:

[0586] The server ranks the information based on the analysis and evaluation results, with the most important information being ranked at the top based on criteria such as reliability, recency, and relevance.

[0587] 5.Display:

[0588] The ranking information is displayed on the user's device, allowing the user to grasp important information at a glance.

[0589] 6. Ratings and Revenue Share:

[0590] The user rates the displayed information, and the results of this rating are sent to the server.

[0591] The server distributes revenue to rankers based on the evaluation data and uses it to optimize the algorithm.

[0592] This system allows users to efficiently gather the information they need, provides access to reliable information, and allows ranking creators to receive appropriate rewards.

[0593] The processing flow will be explained below.

[0594] Step 1:

[0595] Request reception

[0596] On the device: A user enters specific information into a search box and submits a search request, which includes search keywords, such as "smartphone technology in 2023."

[0597] Step 2:

[0598] Submitting a search query

[0599] Terminal: Sends the entered search query to the server. Here, the user's input is passed to the server as the search query.

[0600] Step 3:

[0601] Start gathering information

[0602] Server: Based on the received search query, the server launches a web crawler to collect relevant information from websites, blogs, social media, etc. on the Internet. It may also use specific APIs (such as news APIs) to obtain data.

[0603] Step 4:

[0604] Data normalization

[0605] Server: Normalizes the collected raw data. Specifically, it removes unnecessary elements such as unnecessary HTML tags, advertisements, and scripts from the information and extracts only the text information.

[0606] Step 5:

[0607] Text Cleaning

[0608] Server: Further cleans the normalized text, specifically removing line breaks and special characters, and arranging the text data in a consistent format, which improves the accuracy of text analysis.

[0609] Step 6:

[0610] Natural Language Processing

[0611] Server: Analyzes collected text information using generative AI, extracts important keywords and topics, and gains a deep understanding of the text content through grammatical and semantic analysis.

[0612] Step 7:

[0613] Reliability evaluation

[0614] Server: Using fact-checking APIs and proprietary algorithms, the reliability of collected information is evaluated. A reliability score is calculated and information with a low score is removed.

[0615] Step 8:

[0616] Ranking algorithm applied

[0617] Server: Applies a ranking algorithm to rank the information based on criteria such as reliability, recency, and relevance. Sorting is performed based on the score of each piece of information.

[0618] Step 9:

[0619] Saving ranking results

[0620] Server: Stores the generated ranking results in a database for later display and analysis.

[0621] Step 10:

[0622] Sending ranking results

[0623] Server: Sends the ranking results stored on the user's device.

[0624] Step 11:

[0625] User Interface Display

[0626] Terminal: The ranking results are displayed in a user interface designed to allow users to easily access reliable information.

[0627] Step 12:

[0628] Accepting user reviews

[0629] Device: The user rates the displayed information. Specifically, the rating is completed by clicking a rating button such as "useful" or "trustworthy."

[0630] Step 13:

[0631] Submitting evaluation data

[0632] Terminal: Sends user evaluation data to the server, which reflects the user's evaluation in the system.

[0633] Step 14:

[0634] Revenue Share Calculation

[0635] Server: Based on user rating data, distributes appropriate revenue to rankers. Advertising revenue and sponsorship income are calculated according to the ratings, and the corresponding points and revenue are awarded to rankers.

[0636] Step 15:

[0637] Collecting feedback

[0638] Server: Collects and analyzes user evaluation data and behavioral data, which helps identify areas for system improvement.

[0639] Step 16:

[0640] Algorithm Updates

[0641] Server: Optimizes the ranking algorithm based on the collected feedback data. Uses adaptive learning to improve the accuracy of ranking results from next time onwards.

[0642] This system allows users to efficiently gather the information they need and access reliable information, while also allowing rankers to receive appropriate rewards.

[0643] Example 1

[0644] 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."

[0645] Conventional information collection and analysis systems often lack the means to efficiently collect large amounts of information and evaluate it based on its reliability. As a result, it is difficult for users to quickly access the reliable information they need. Furthermore, there are issues with efficiency and accuracy in the process of verifying the accuracy of the collected information and generating rankings based on it. Furthermore, there is a lack of means to optimize the system itself using user evaluation data.

[0646] 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.

[0647] In this invention, the server includes: a means for a user to enter keywords into a search box on a terminal and send a request; a means for collecting information from a digital network based on the search request; a means for normalizing the collected information and removing unnecessary data to convert it into pure text data; a means for analyzing the collected text data using a generative AI model and extracting key keywords and concepts; a means for evaluating the reliability of information using a fact-checking API; a means for ranking information based on the analyzed and evaluated information; a means for displaying the ranking results on the user's terminal; and a means for collecting user evaluation data and distributing revenue to rankers. This enables efficient collection and analysis of large amounts of information, and evaluation and ranking based on reliability. Furthermore, utilizing user evaluation data enables system optimization and enables the provision of more accurate information.

[0648] "User" refers to an entity that uses the system to search for and evaluate information.

[0649] "Device" refers to a device used by a User, such as a computer, smartphone, or tablet.

[0650] A "search box" is an interface element that allows a user to enter search keywords.

[0651] A "request" refers to a search or operation request sent by a user from a terminal to a server.

[0652] A "server" is a collection of hardware and software that receives requests from users and collects and processes information.

[0653] "Digital network" refers to the entire system for electronic information exchange, including the Internet.

[0654] "Means of collecting information" refers to methods of obtaining the necessary data from the Internet using web crawlers or API access.

[0655] "Normalization" refers to the process of removing unnecessary elements from collected data and converting it into a unified format.

[0656] "Unnecessary data" refers to elements of collected information, such as HTML tags and advertisements, that may reduce the accuracy of search results.

[0657] "Pure text data" refers to information after it has been normalized and converted into a parsable text format.

[0658] "Generative AI model" refers to an artificial intelligence model for analyzing collected text data.

[0659] "Means for analyzing text data" refers to a method of using a generative AI model to analyze collected text data and extract key keywords and concepts.

[0660] "Major keywords and concepts" refer to information elements that are particularly important within the text data and serve as the basis for ranking.

[0661] A "fact-checking API" refers to an interface for external services that allow users to verify the reliability of information.

[0662] "Means for evaluating the reliability of information" refers to methods for scoring the veracity and reliability of collected information using fact-checking APIs, etc.

[0663] "Ranking" refers to the process of ranking information according to certain criteria (such as reliability, recency, relevance, etc.) based on evaluated information.

[0664] "Ranking results" refers to a collection of ranked information and a list of ranked information displayed to users.

[0665] "User rating data" refers to the rating a user gives to the displayed information, including feedback such as star ratings and comments.

[0666] "Ranker" refers to the person who creates the ranking or receives compensation based on the ranking.

[0667] "Means of revenue distribution" refers to a method of distributing appropriate rewards to rankers based on user evaluation data.

[0668] MODE FOR CARRYING OUT THE INVENTION

[0669] The system of the present invention is designed to enable users to efficiently collect reliable information. The system automatically performs a series of processes, from processing search requests to collecting, normalizing, analyzing, ranking, and displaying information to users. Furthermore, it collects user evaluation data and uses it for system optimization and revenue distribution.

[0670] Information collection and preprocessing

[0671] When a user enters a specific keyword (e.g., "Smartphone technology in 2023") into the device's search box and presses the search button, the server receives the request. Based on the request, the server uses a web crawler (e.g., Scrapy) to retrieve information from digital networks. It also collects the latest information from social media and search engines through API access (e.g., Twitter API, Google Custom Search API).

[0672] The collected information is normalized on the server using libraries such as Python's BeautifulSoup, which removes unnecessary HTML tags and advertisements and converts it into pure text data.

[0673] Information analysis

[0674] The normalized text data is analyzed by a generative AI model (e.g., OpenAI's GPT-4). The generative AI model is used to extract key keywords and concepts from the collected text data. At the same time, the reliability of the information is evaluated using a fact-checking API (e.g., Factmata API). The fact-checking API is used to score the veracity and reliability of the information.

[0675] Ranking and Display

[0676] Once the information has been analyzed and evaluated, it is ranked based on criteria such as reliability, recency, and relevance. In this ranking process, an overall score is assigned to each piece of information, and the information is sorted in descending order of its score. The ranking results are stored in a database on the server and displayed on the user's device. Users can use this ranking to quickly access the information they need.

[0677] To give a concrete example, let's say a user searches for information on "smartphone technologies to watch in 2023." Here's how the system works:

[0678] 1. A user types "2023 smartphone technology" into the search box and submits a request.

[0679] 2. The server collects relevant information from the Internet.

[0680] 3. The server removes unnecessary HTML tags and advertisements from the collected information and converts it into pure text data.

[0681] 4. The server analyzes the text using a generative AI model (e.g., GPT-4) to extract key keywords and topics, and also uses a fact-checking API to assess the reliability of the information.

[0682] 5. The server ranks the information based on the analysis and evaluation results. The most important information is ranked at the top based on criteria such as reliability, recency, and relevance.

[0683] 6. Ranking information is displayed on the user's device, allowing the user to grasp important information at a glance.

[0684] An example of a prompt sentence might be:

[0685] "Please explain the specific information gathering procedures for the system that allows users to gather information about 'notable smartphone technologies in 2023.'"

[0686] This system allows users to efficiently gather the information they need and quickly access reliable information. Furthermore, by using user evaluation data, the system can be optimized to provide even higher quality information.

[0687] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0688] Step 1: Receiving a search request

[0689] Device: A user enters a specific keyword (e.g., "smartphone technology 2023") into the device's search box and presses the search button. This action generates a search request.

[0690] Input: The search keyword entered by the user.

[0691] Output: The search request sent to the server.

[0692] Step 2: Gather information

[0693] Server: Based on the received search request, launch a web crawler (e.g., Scrapy) to collect relevant information from the Internet, and then use API access (e.g., Twitter API, Google Custom Search API) to obtain relevant data.

[0694] Input: Keywords based on your search request.

[0695] Output: Collected data from web pages and APIs.

[0696] Step 3: Normalize the data

[0697] Server: To normalize the collected information, we use Python's BeautifulSoup library to remove unnecessary HTML tags and advertisements and convert it into pure text data.

[0698] Input: Raw data collected from web pages and APIs.

[0699] Output: Cleaned and normalized text data.

[0700] Step 4: Analyzing the text data

[0701] Server: Analyzes the normalized text data using a generative AI model (e.g., OpenAI's GPT-4), extracting key keywords and concepts.

[0702] Input: Normalized text data.

[0703] Output: Extracted keywords and concepts.

[0704] Step 5: Assess reliability

[0705] Server: Use a fact-checking API (e.g., Factmata API) to evaluate the reliability of the extracted information. Using this API, the truthfulness and reliability score of the information can be obtained.

[0706] Input: Keywords and concepts extracted by the generative AI model.

[0707] Output: Evaluation results including confidence scores.

[0708] Step 6: Ranking the information

[0709] Server: Based on the analyzed and evaluated information, the information is ranked based on specific criteria (reliability, freshness, relevance, etc.) and the rankings are stored in a SQL database or similar.

[0710] Input: Confidence score and evaluation results.

[0711] Output: A list of information in ranking format.

[0712] Step 7: View ranking information

[0713] Device: Ranking information is displayed on the user's device, allowing users to quickly access important information.

[0714] Input: A ranking list of information sent from the server.

[0715] Output: A viewable list of information displayed on the user's terminal.

[0716] Step 8: Collecting ratings and sharing revenue

[0717] Terminal: The user rates the displayed information, and the results are sent to the server.

[0718] Input: User rating data (star ratings, comments, etc.).

[0719] Output: The rating data sent to the server.

[0720] Server: Receives user evaluation data and distributes revenue to rankers based on that data. Additionally, the collected evaluation data is used to optimize the system.

[0721] Input: User rating data.

[0722] Output: Revenue distribution results to rankers and algorithm improvements using evaluation data.

[0723] In this way, the system performs specific data processing and calculations at each step, enabling users to access reliable information in a short time, and also optimizes the system itself by utilizing user evaluation data.

[0724] (Application example 1)

[0725] 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."

[0726] A major issue facing modern online shopping sites is the low reliability of product reviews. Many reviews are inaccurate or exaggerated, making it difficult for users to make appropriate decisions. This has created a need for users to efficiently find reliable reviews. Furthermore, existing ranking systems tend to rely on subjective evaluations and lack objective evaluations of reliability. This has led to major challenges in improving the reliability of reviews and user satisfaction.

[0727] 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.

[0728] In this invention, the server includes means for collecting and normalizing information from a digital network using a generative AI, means for analyzing the collected information using natural language processing and evaluating its reliability, means for ranking the analyzed and evaluated information based on criteria, means for displaying the information in a ranking format on the user's terminal, means for collecting user evaluation data and distributing revenue to rankers, and means for collecting product reviews and displaying highly reliable reviews at the top, thereby enabling users to efficiently access highly reliable product reviews.

[0729] "Generative AI" refers to generative artificial intelligence, a technology that generates new information and answers based on large amounts of data.

[0730] "Normalization" refers to the process of removing unnecessary information and formats from collected data.

[0731] "Natural language processing" refers to the technology that enables computers to understand, analyze, and process human language.

[0732] "Assessing trustworthiness" is the process of determining how accurate or relevant the information collected is.

[0733] "Ranking" refers to ranking data according to specific criteria.

[0734] "User's terminal" refers to a device used by a user, such as a smartphone or computer.

[0735] "User Ratings Data" means feedback and rating information provided by Users.

[0736] A "ranker" refers to the person or system that creates rankings based on collected information and evaluation data.

[0737] "Means for distributing revenue" refers to a mechanism for appropriately distributing revenue generated based on the generated information and rankings.

[0738] "Product reviews" refer to the impressions and evaluation data left by users after using a product.

[0739] A "reliable review" is a review that contains objective and accurate information and is deemed to be useful to other users.

[0740] This invention is a system that uses generative AI technology to collect reliable product reviews and provide them to users in a ranking format. This system consists of the following steps: information collection, data preprocessing, information analysis, reliability evaluation, ranking generation, display, user evaluation, and revenue distribution.

[0741] The server first uses generative AI to collect product review information from digital networks and normalizes it by removing unnecessary elements such as HTML tags and advertisements. Libraries such as BeautifulSoup and requests are used in this process. The collected data is then analyzed using natural language processing (NLP) to extract important keywords and topics. Software such as SpaCy and transformers are used in this stage.

[0742] The server evaluates the reliability of the reviews based on the analysis results and assigns them a score using a fact-checking API. Based on the obtained score, the reviews are ranked according to criteria. Data analysis tools such as scikit-learn are used to generate the rankings. Once the reliability evaluation and ranking are complete, the ranking results are sent to the user's device, where they can be viewed on their smartphone or other device.

[0743] Users rate the displayed ranking results, and the rating data is sent to the server. The server then optimizes the ranking algorithm based on the user rating data and distributes revenue to the rankers. As a concrete example, when a user searches for "latest smartphones," reliable reviews are displayed at the top. For example, specific reviews such as "This smartphone has a long battery life" are displayed in a prominent position.

[0744] An example of a prompt is, "Please collect reviews of the latest smartphones, rate their reliability, and rank them." By using such a prompt, users can efficiently obtain the information they need.

[0745] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0746] Step 1:

[0747] Information gathering

[0748] The server collects relevant review information from the digital network based on the keywords entered by the user in the search box. Specifically, it retrieves the information from the relevant product's web page or review site using the requests library. The retrieved information is saved on the server in HTML format.

[0749] Input: User's search keyword

[0750] Output: Collected review information in HTML format

[0751] Step 2:

[0752] Data Preprocessing

[0753] The server normalizes the collected HTML review information by using the BeautifulSoup library to remove unnecessary elements such as HTML tags and advertisements, converting it into pure text data, and removing line spacing, whitespace, and special characters.

[0754] Input: Collected review information in HTML format

[0755] Output: Normalized text data

[0756] Step 3:

[0757] Natural Language Processing (NLP) Analysis

[0758] The server then performs natural language processing on the normalized text data using SpaCy or transformers, extracting important keywords and topics from the reviews and calculating features (such as TF-IDF) for each review.

[0759] Input: normalized text data

[0760] Output: Extracted keywords, topics, and features

[0761] Step 4:

[0762] Reliability evaluation

[0763] The server uses a fact-checking API to score the credibility of each review. Specifically, it sends the analyzed data to the fact-checking API and receives a credibility score, which quantifies the credibility of each review.

[0764] Input: Extracted keywords, topics, and features

[0765] Output: Confidence score

[0766] Step 5:

[0767] Ranking Generation

[0768] The server ranks reviews based on their credibility score. The ranking algorithm uses factors such as credibility score, recency, and relevance. The ranking is generated using libraries such as scikit-learn.

[0769] Inputs: Confidence score, recency, relevance

[0770] Output: Ranked reviews

[0771] Step 6:

[0772] Displaying ranking results

[0773] The device displays the ranking results sent from the server to the user, who can then view the reviews ranked in order of reliability on their smartphone or computer screen.

[0774] Input: Ranked reviews

[0775] Output: Ranking results displayed on the user's device

[0776] Step 7:

[0777] User ratings and data collection

[0778] Users rate the displayed ranking results, and the rating data is sent from the device to the server, including feedback on the reliability and usefulness of the reviews.

[0779] Input: User Rating

[0780] Output: Rating data sent to the server

[0781] Step 8:

[0782] Revenue sharing and algorithm optimization

[0783] The server optimizes the ranking algorithm based on the evaluation data collected from users, and also executes the process of distributing profits to rankers. Appropriate data analysis is performed to improve the accuracy of the ranking algorithm and distribute profits.

[0784] Input: User rating data

[0785] Output: Optimized ranking algorithm, revenue sharing results

[0786] 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.

[0787] This invention provides a system that combines generative AI and an emotion engine to efficiently collect, analyze, assess credibility, rank, evaluate users, and distribute revenue. Furthermore, by analyzing user emotions, it optimizes the ranking algorithm and dynamically adjusts the content displayed.

[0788] Overall system flow

[0789] 1. Information gathering stage

[0790] On the device: A user types specific information (e.g., "smartphone technology in 2023") into a search box and submits a request.

[0791] Server: Based on the search request, they gather relevant information from websites, blogs, and social media across their digital network, using crawling and API access to obtain data.

[0792] 2. Data preprocessing stage

[0793] Server: Normalizes the collected information and removes unnecessary elements such as unnecessary HTML tags and advertisements, resulting in pure text data.

[0794] 3. Information Analysis Stage

[0795] Server: Using generative AI, natural language processing is performed to analyze the collected text, extracting important keywords and topics, and evaluating the reliability of the information using fact-checking APIs.

[0796] 4. Ranking Stage

[0797] Server: Analyzes and evaluates the information to generate a ranking according to specific criteria, such as reliability, freshness, and relevance.

[0798] 5. Display Stage

[0799] Device: The generated ranking results are displayed on the user's device, allowing the user to view the ranked, reliable information.

[0800] 6. Rating and Revenue Sharing Phase

[0801] Terminal: The user rates the displayed information. The rating data is sent to the server.

[0802] Server: Distributes appropriate revenue to rankers based on user rating data. Collects rating data to help improve the algorithm.

[0803] 7. Sentiment Analysis Stage

[0804] Device: An emotion engine that recognizes the user's emotions runs and collects the user's emotion data.

[0805] Example: An emotion engine analyzes in real time whether a user is very interested, angry, sad, etc.

[0806] 8. Sentiment data analysis and algorithm optimization

[0807] Server: The ranking algorithm is optimized based on the collected emotional data, such as giving priority to information where the user expressed positive emotions.

[0808] Example: If the emotion engine analyzes the user's emotion as "interesting," it will rank similarly interesting information higher.

[0809] 9. Dynamic Content Adjustment

[0810] On the device: Dynamically adjust the content displayed based on the analysis results of the emotion engine, prioritizing the display of topics and related information that interest the user.

[0811] Example: If a user is determined to be interested in smartphone technology, the app will automatically show them new related articles and videos.

[0812] Specific use cases

[0813] For example, if a user wants to gather information about "the most noteworthy smartphone technologies in 2023," the system works as follows:

[0814] 1. Information Collection:

[0815] A user types "2023 smartphone technology" into the search box and submits a request.

[0816] The server collects relevant information from the Internet.

[0817] 2. Data Preprocessing:

[0818] The server removes unnecessary HTML tags and advertisements from the collected information and converts it into pure text data.

[0819] 3. Information analysis:

[0820] The server uses generative AI to analyze the text and extract key keywords and topics, and it also uses fact-checking APIs to assess the reliability of the information.

[0821] 4. Emotion analysis:

[0822] While users browse information, the emotion engine analyzes their emotions in real time and changes relevant information if the user is interested or dissatisfied.

[0823] 5. Ranking:

[0824] The server ranks the information based on the analysis and evaluation results, incorporating factors such as reliability, recency, and relevance, as well as sentiment data, to rank the most important information at the top.

[0825] 6. Dynamic content adjustment:

[0826] The content displayed is dynamically adjusted based on the user's emotional data, prioritizing new and relevant information that will interest the user.

[0827] 7. Ratings and Revenue Share:

[0828] Users rate the displayed information, and the results are sent to the server, which distributes revenue to rankers based on the rating data and uses it to optimize the algorithm.

[0829] This system allows users to efficiently gather the information they need and access reliable information. By taking user sentiment into consideration, it also enables more personalized information to be provided, and rankers can receive appropriate rewards.

[0830] The processing flow will be explained below.

[0831] Step 1:

[0832] Request reception

[0833] On the device: A user enters specific information into a search box and submits a search request, which includes search keywords, such as "smartphone technology in 2023."

[0834] Example: A user searches for "smartphone technology 2023."

[0835] Step 2:

[0836] Submitting a search query

[0837] Terminal: Sends the entered search query to the server.

[0838] Step 3:

[0839] Start gathering information

[0840] Server: Based on the received search query, the server launches a web crawler to collect relevant information from websites, blogs, social media, etc. on the Internet. It may also use specific APIs (such as news APIs) to obtain data.

[0841] Example: The server crawls news sites and social media sites using the keyword "smartphone technology 2023" to collect related information.

[0842] Step 4:

[0843] Data normalization

[0844] Server: The collected raw data is normalized. Specifically, unnecessary elements such as unnecessary HTML tags, advertisements, and scripts are removed from the information, and only the text information is extracted.

[0845] Example: Removing advertisements and unnecessary HTML tags from aggregated news articles and blog posts.

[0846] Step 5:

[0847] Text Cleaning

[0848] Server: Further cleans the normalized text, specifically removing line breaks and special characters, and arranging the text data in a consistent format, which improves the accuracy of text analysis.

[0849] For example: cleaning text by removing unnecessary spaces and special characters.

[0850] Step 6:

[0851] Natural Language Processing

[0852] Server: Analyzes collected text information using generative AI, extracts important keywords and topics, and gains a deep understanding of the text content through grammatical and semantic analysis.

[0853] Example: Extracting keywords related to smartphone technology (5G, camera resolution, battery life, etc.).

[0854] Step 7:

[0855] Reliability evaluation

[0856] Server: Using fact-checking APIs and proprietary algorithms, the server evaluates the reliability of collected information, calculates a reliability score, and filters out information with a low score.

[0857] Example: Calculate a credibility score for each source and filter out information with a low score.

[0858] Step 8:

[0859] Ranking algorithm applied

[0860] Server: Applies a ranking algorithm to rank the information based on criteria such as reliability, recency, and relevance. Sorting is performed based on the score of each piece of information.

[0861] Example: Combining reliability, recency, and relevance scores and ranking the information with the highest score at the top.

[0862] Step 9:

[0863] Saving ranking results

[0864] Server: Stores the generated ranking results in a database for later display and analysis.

[0865] Example: Storing ranked smartphone technology information in a database.

[0866] Step 10:

[0867] Sending ranking results

[0868] Server: Sends the ranking results stored on the user's device.

[0869] Example: Sending information about ranked smartphone technologies to the user's device.

[0870] Step 11:

[0871] User Interface Display

[0872] Terminal: The ranking results are displayed on the user interface, allowing users to access reliable information through the displayed rankings.

[0873] Example: A smartphone screen displays a list of reliable smartphone technology information.

[0874] Step 12:

[0875] Accepting user reviews

[0876] Device: The user rates the displayed information. Specifically, the rating is completed by pressing a rating button such as "useful" or "trustworthy."

[0877] Example: A user rates this review as helpful.

[0878] Step 13:

[0879] Submitting evaluation data

[0880] Terminal: Sends user evaluation data to the server, which reflects the user's evaluation in the system.

[0881] Example: User rating data is sent to the server.

[0882] Step 14:

[0883] Revenue Share Calculation

[0884] Server: Based on user rating data, distributes appropriate revenue to rankers. Advertising revenue and sponsorship income are calculated according to the ratings, and the corresponding points and revenue are awarded to rankers.

[0885] Example: Revenues are distributed to rankers based on user ratings.

[0886] Step 15:

[0887] Collecting feedback

[0888] Server: Collects and analyzes user evaluation data and behavioral data, which helps identify areas for system improvement.

[0889] Example: Collecting evaluation data to help improve the system.

[0890] Step 16:

[0891] Algorithm Updates

[0892] Server: Optimizes the ranking algorithm based on the collected feedback data. Uses adaptive learning to improve the accuracy of ranking results from next time onwards.

[0893] Example: Learn the characteristics of highly rated articles and reflect them in future rankings.

[0894] Step 17:

[0895] Emotion analysis

[0896] On-device: Utilizing an emotion engine to recognize the user's emotional state in real time, including interest, excitement, anger, sadness, etc.

[0897] Example: An emotion engine analyzes user emotions in real time while the user is browsing information.

[0898] Step 18:

[0899] Sending emotional data

[0900] Terminal: Sends the user's emotional data to the server.

[0901] Example: The analyzed emotion data is sent to the server.

[0902] Step 19:

[0903] Sentiment data analysis and algorithm optimization

[0904] Server: The ranking algorithm is optimized based on the collected emotional data, such as giving priority to information where the user expressed positive emotions.

[0905] For example, if the emotion engine analyzes the user's emotion as "interesting," it will rank similarly interesting information higher.

[0906] Step 20:

[0907] Dynamic Content Adjustment

[0908] On the device: Dynamically adjust the content displayed based on the analysis results of the emotion engine, prioritizing the display of topics and related information that interest the user.

[0909] Example: If a user expresses an interest in smartphone technology, automatically show them new related articles and videos.

[0910] This system allows users to efficiently gather the information they need and access reliable information. By taking user sentiment into consideration, it also enables more personalized information to be provided, and rankers can receive appropriate rewards.

[0911] Example 2

[0912] 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."

[0913] Conventional information gathering and organization systems have a process for ranking collected information based on its reliability, recency, and relevance, but they lack optimization that takes into account user emotions and evaluation data, limiting their ability to improve user experience. In particular, they are unable to reflect users' emotional reactions to content evaluations in real time, making it difficult to provide personalized information. Furthermore, there are also insufficient means for automatically evaluating and scoring the reliability of collected information, creating a risk of misinformation being included.

[0914] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0915] In this invention, the server includes means for collecting and normalizing information from a digital network using generative AI, means for analyzing the collected information using natural language processing and evaluating its reliability, means for ranking the analyzed and evaluated information based on criteria, means for displaying the information in a ranking format on the user's terminal, means for collecting user evaluation data and distributing revenue to rankers, means for analyzing user emotions using an emotion engine, means for optimizing the ranking algorithm based on the analyzed emotion data, and means for dynamically adjusting the content to be displayed based on the emotion data. This enables the provision of personalized information based on the user's emotion and evaluation data, resulting in more reliable information and an improved user experience.

[0916] "Generative AI" is a type of artificial intelligence technology used for natural language processing and data analysis, and has the ability to generate, convert, and analyze information from a wide range of data.

[0917] "Digital network" refers to the Internet and other digital communications networks that enable the collection, sharing, and transfer of information.

[0918] "Normalization" refers to the process of removing unnecessary data (e.g., HTML tags and advertisements) from collected information and converting it into pure text data.

[0919] "Natural language processing" refers to the technology of analyzing collected text data, extracting keywords and topics, and understanding the meaning and relationships of the information.

[0920] "Credibility assessment" is the process of assessing whether collected information is accurate, and includes assigning a credibility score using fact-checking APIs, etc.

[0921] "Ranking" refers to the process of analyzing and evaluating information and ranking it based on certain criteria (reliability, recency, relevance).

[0922] "User rating data" refers to data including ratings (e.g., star ratings, comments, etc.) made by users on information.

[0923] "Revenue sharing" refers to the process of distributing appropriate rewards to information providers based on users' rating data.

[0924] An "emotion engine" refers to technology for analyzing emotions from a user's facial expressions, voice, etc., and is used to collect and analyze emotional data in real time.

[0925] "Dynamic adjustment" refers to the process of changing and adjusting displayed content in real time based on the results of user sentiment analysis.

[0926] A "fact-checking API" is an application program interface used to assess the reliability of collected information and is a tool for fact-checking information.

[0927] This invention provides a system that combines generative AI and an emotion engine to efficiently collect, analyze, assess reliability, rank, and distribute revenues to users. Additionally, analyzing user emotions can optimize the ranking algorithm and dynamically adjust the content displayed. Below, we will explain each element of the system and how they are combined.

[0928] First, the device accepts the user's search request. The device has a search box, and the user enters the keyword of the information they want to find (e.g., "Smartphone technology in 2023") into this search box. The device then sends the search request to the server using the HTTP protocol.

[0929] Based on the search request received, the server collects relevant information from websites, blogs, social media, and other sources across the digital network, using crawling tools such as Apache Nutch and services such as the Twitter API.

[0930] The server then normalizes the collected information by using an HTML parser such as BeautifulSoup to remove unnecessary HTML tags and advertisements from the collected web pages and convert them into pure text data.

[0931] A generative AI model (e.g., GPT-4) runs on a server and analyzes the collected and normalized text data. During this analysis, natural language processing techniques are used to extract key keywords and topics, and fact-checking APIs (e.g., Snopes API) are used to assess the reliability of the information.

[0932] The server then generates a ranking based on the analyzed and evaluated information according to certain criteria (authority, recency, relevance), using a modified version of the PageRank algorithm or a recommendation system for this ranking process.

[0933] The ranking results are displayed on the user's device. The data is presented in a visually easy-to-read format using HTML and CSS. For example, the "latest articles about AI camera technology" could be displayed at the top.

[0934] Users rate the displayed information (e.g., star ratings or comments). The rating data is sent to the server, which then distributes appropriate revenue to the rankers based on the rating data. This rating data is also used to optimize the algorithm.

[0935] The user's device is equipped with an emotion engine, which uses technologies such as Emotion API to collect emotional data in real time from the user's facial expressions and voice. For example, it analyzes whether the user is expressing the emotion of "joy."

[0936] The server optimizes the ranking algorithm based on the collected sentiment data, for example, adjusting the algorithm to prioritize ranking information that users find "interesting," thereby improving the user experience.

[0937] Furthermore, the content displayed can be dynamically adjusted based on emotional data. For example, if it is determined that a user is interested in "AI camera technology," new related articles and videos will be dynamically displayed.

[0938] This system allows users to efficiently gather the information they need and access reliable information. It also enables personalized information provision that takes into account the user's emotions, and information providers can receive appropriate rewards.

[0939] Specific prompt examples:

[0940] "Please explain in detail the process flow for a user wanting to gather information about smartphone technology in 2023."

[0941] By combining generative AI models and emotion analysis technology, this system can respond quickly and appropriately to diverse user needs, significantly improving the quality and efficiency of information provision.

[0942] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0943] Step 1:

[0944] User: Enters "Smartphone technology in 2023" into the search box on the device and sends a search request, which generates search keywords as input data and sends them to the server using the HTTP protocol.

[0945] Input: Search keyword "Smartphone technology in 2023"

[0946] Output: Search request

[0947] Step 2:

[0948] Server: Based on the received search request, the server uses crawling tools such as Apache Nutch and Twitter APIs to collect relevant information from websites, blogs, and social media on the digital network. The collected information is stored as text data.

[0949] Input: Search request

[0950] Output: Collected information (text data)

[0951] Step 3:

[0952] Server: Normalize the collected information. Use an HTML parser such as BeautifulSoup to remove unnecessary HTML tags and advertisements and convert it into pure text data. This data processing results in clean text data.

[0953] Input: Collected information (text data)

[0954] Output: Normalized text data

[0955] Step 4:

[0956] Server: Uses a generative AI model (e.g., GPT-4) to extract key keywords and topics from the normalized text data, and uses fact-checking APIs (e.g., Snopes API) to assess the reliability of the information, resulting in analytics data and a reliability score.

[0957] Input: normalized text data

[0958] Output: Analysis data (major keywords, topics), reliability score

[0959] Step 5:

[0960] Server: Based on the analyzed information and the reliability score, it generates a ranking according to certain criteria (reliability, recency, relevance). This ranking is done using a modified version of the PageRank algorithm or a recommendation system.

[0961] Input: Analysis data (main keywords, topics), reliability score

[0962] Output: Ranking results

[0963] Step 6:

[0964] Device: The ranking results are visually formatted using HTML and CSS and displayed to the user. Users can view the ranking information on their devices.

[0965] Input: Ranking results

[0966] Output: Visualized ranking results (HTML, CSS)

[0967] Step 7:

[0968] User: Rates the displayed information (e.g., star ratings or comments). The rating data is sent to the server using the HTTP protocol.

[0969] Input: Visualized ranking results

[0970] Output: Rating data (star ratings, comments)

[0971] Step 8:

[0972] Server: Based on the collected evaluation data, the server distributes appropriate revenue to rankers. This evaluation data is also used to optimize the ranking algorithm.

[0973] Input: Evaluation data

[0974] Output: Revenue share data, optimized ranking algorithm

[0975] Step 9:

[0976] Device: Runs the emotion engine and collects emotional data from the user's facial expressions and voice. For example, it uses the Emotion API to analyze emotions such as "happiness," "interest," and "dissatisfaction" in real time.

[0977] Input: User's facial expressions and voice

[0978] Output: Emotion data

[0979] Step 10:

[0980] Server: The ranking algorithm is further optimized based on the data obtained from the sentiment analysis. For example, the algorithm is adjusted to prioritize ranking information that users find "interesting."

[0981] Input: Emotion data

[0982] Output: Optimized ranking algorithm

[0983] Step 11:

[0984] On-device: Dynamically adjusts content displayed based on sentiment analysis. For example, if a user is determined to be interested in "AI camera technology," relevant new articles and videos will be displayed.

[0985] Input: Sentiment analysis results

[0986] Output: Dynamically adjusted content placement

[0987] (Application example 2)

[0988] 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."

[0989] Conventional information gathering and ranking systems have had difficulty providing dynamic content that takes user sentiment into account, resulting in a lack of optimization for the user experience. Furthermore, while there is a demand for reliable information, there are limited methods for evaluating the accuracy and reliability of collected information in real time. Furthermore, while there is a demand for quickly optimizing ranking algorithms using user evaluation data and providing personalized information, there has been no system that combines sentiment analysis.

[0990] 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.

[0991] In this invention, the server includes means for collecting and normalizing information from a digital network using generative AI, means for analyzing the collected information using natural language processing and evaluating its reliability, means for ranking the analyzed and evaluated information based on criteria, means for displaying the information in a ranking format on the user's terminal, means for collecting user evaluation data and distributing revenue to rankers, means for analyzing user emotions in real time using a sentiment analysis engine, means for optimizing the ranking algorithm based on the collected sentiment data, and means for dynamically adjusting the content to be displayed. This makes it possible to provide efficient information taking user emotions into consideration, identify reliable information in a ranking format, and dynamically provide personalized information.

[0992] "Generative AI" is an artificial intelligence technology that collects information from digital networks and analyzes and generates data using natural language processing.

[0993] "Normalization" is the process of removing unnecessary elements from collected data and converting it into pure text data that is easy to analyze.

[0994] "Natural language processing" is a technology that allows computers to understand and analyze human language.

[0995] "Assessing credibility" is a method for determining the legitimacy or reliability of collected information.

[0996] "Ranking" refers to the process of ranking collected and analyzed information according to specific criteria.

[0997] An "emotion analysis engine" is a technology that analyzes user emotions in real time and collects them as data.

[0998] "Dynamic adjustment" is the process of instantly changing and updating displayed content based on data collected in real time.

[0999] "Distributing revenue" means distributing appropriate rewards to content providers and others based on user evaluation data.

[1000] This invention is a system that combines generative AI and a sentiment analysis engine to efficiently collect, analyze, assess credibility, rank, evaluate users' ratings, analyze sentiment, and dynamically adjust content. The following hardware and software are required for implementation:

[1001] Hardware and software used

[1002] Hardware: Camera (to capture the user's facial expressions), microphone (to capture the user's voice), user device (such as a smartphone or tablet).

[1003] Software: Generative AI model (for information analysis), TextBlob (for natural language processing), TensorFlow / Keras (for sentiment analysis model), REST API (for information collection).

[1004] What the program does

[1005] The server uses generative AI to collect information from digital networks, obtains data through crawling and API access, and normalizes this information, removing unnecessary HTML tags and advertisements, converting it into pure text data.

[1006] Furthermore, the collected text data is analyzed using natural language processing software (such as TextBlob), important keywords and topics are extracted, and the reliability of the information is evaluated through a fact-checking API. The evaluated information is ranked according to specific criteria and displayed in a ranked format on the user's device.

[1007] User sentiment is analyzed in real time using a sentiment analysis engine (such as TensorFlow / Keras models), and the ranking algorithm is optimized based on the collected sentiment data. The content displayed is dynamically adjusted according to the user's interests and sentiment.

[1008] Specific examples

[1009] For example, consider a case where a user wants to gather information about the latest VR technology. The user enters keywords into the search box, and the system collects related information from the internet. The collected information is then analyzed using natural language processing to extract important keywords and topics. The reliability of the information is evaluated based on a fact-checking API, and the results are displayed to the user in a ranked format.

[1010] While users are browsing information, the sentiment analysis engine analyzes their facial expressions and voice in real time. Information that is determined to be of interest to the user is displayed preferentially, and related information is dynamically adjusted, allowing users to quickly and efficiently obtain the content that interests them most.

[1011] Prompt Sentence Examples

[1012] "Analyze the following text and extract key keywords and sentiment:"

[1013] Text to analyze: "I'm looking for information on the latest in VR technology."

[1014] Using this prompt, the generative AI model analyzes the text data and provides information according to the user's request, which is expected to significantly improve the user experience.

[1015] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1016] Step 1:

[1017] The user enters a search keyword from the terminal and sends a request.

[1018] Input: User's search keyword

[1019] Output: Search request

[1020] How it works: A user enters a keyword into the search box on their device and presses the submit button, which sends the keyword to the system.

[1021] Step 2:

[1022] The server uses generated AI to collect information related to the search keywords from the Internet.

[1023] Input: Search request

[1024] Output: Collected digital information (web pages, blogs, social media posts, etc.)

[1025] How it works: The server collects relevant information from various sources on the Internet (websites and APIs) and stores it in text format.

[1026] Step 3:

[1027] The server normalizes the collected information and removes unnecessary HTML tags and advertisements.

[1028] Input: Collected digital information

[1029] Output: Clean text data

[1030] How it works: The normalization process removes unnecessary elements such as HTML tags and advertisements, producing pure text data.

[1031] Step 4:

[1032] The server uses natural language processing to analyze the clean text data and extract important keywords and topics.

[1033] Input: clean text data

[1034] Output: Analysis results (important keywords, extracted topics)

[1035] What it does: It uses a natural language processing library such as TextBlob to parse the text and extract key keywords and topics.

[1036] Step 5:

[1037] The server uses a fact-checking API to assess the reliability of the collected information.

[1038] Input: clean text data

[1039] Output: Trustworthiness score

[1040] How it works: Scores the legitimacy and veracity of collected information through a fact-checking API.

[1041] Step 6:

[1042] The server analyzes and evaluates the information and then ranks it according to specific criteria.

[1043] Input: Analysis results, reliability evaluation score

[1044] Output: Ranked list of information

[1045] How it works: Applies an algorithm that ranks information based on criteria such as reliability, freshness, and relevance.

[1046] Step 7:

[1047] The terminal provides the user with information displayed in a ranking format.

[1048] Input: Ranked list of information

[1049] Output: Ranking information displayed in the user interface

[1050] How it works: The ranking results are displayed on the user's screen, allowing the user to visually confirm reliable information.

[1051] Step 8:

[1052] The device uses an emotion analysis engine to analyze the user's facial expressions and voice in real time.

[1053] Input: User's facial expression data, voice speech data

[1054] Output: Sentiment analysis results (e.g., interest, joy, dissatisfaction)

[1055] How it works: The camera and microphone are used to capture the user's facial expressions and voice, which are then analyzed in real time by an emotion analysis engine.

[1056] Step 9:

[1057] The server optimizes the ranking algorithm based on the collected sentiment data.

[1058] Input: Sentiment analysis results, ranking information list

[1059] Output: Optimized ranking information

[1060] How it works: Reflects user sentiment data and adjusts and optimizes the ranking algorithm.

[1061] Step 10:

[1062] The device dynamically adjusts the content displayed based on the user's interests and emotions.

[1063] Input: Optimized ranking information, sentiment analysis results

[1064] Output: Dynamically adjusted content

[1065] How it works: Changes the information shown based on your interests and automatically surfaces new relevant content and videos.

[1066] 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.

[1067] 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.

[1068] 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.

[1069] [Third embodiment]

[1070] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1071] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1072] 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).

[1073] 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.

[1074] 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.

[1075] 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).

[1076] 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.

[1077] 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.

[1078] 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.

[1079] 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.

[1080] 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.

[1081] 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."

[1082] This system uses generative AI to collect large amounts of data, analyzes it using natural language processing (NLP), and evaluates its reliability, allowing users to efficiently gather specific information in a short period of time and obtain highly reliable information in a ranked format.

[1083] Overall system flow

[1084] 1. Information gathering stage

[1085] On the device: The user enters specific information into a search box and submits a request.

[1086] Server: Based on the search request, they gather relevant information from websites, blogs, and social media across their digital network, using crawling and API access to obtain data.

[1087] 2. Data preprocessing stage

[1088] Server: Normalizes the collected information and removes unnecessary elements such as unnecessary HTML tags and advertisements, resulting in pure text data.

[1089] 3. Information Analysis Stage

[1090] Server: Using generative AI, natural language processing is performed to analyze the collected text, extracting important keywords and concepts, and evaluating the reliability of the information using a fact-checking API.

[1091] 4. Ranking Stage

[1092] Server: Analyzes and evaluates the information to generate a ranking according to specific criteria, such as reliability, freshness, and relevance.

[1093] 5. Display Stage

[1094] Device: The generated ranking results are displayed on the user's device, allowing the user to view the ranked, reliable information.

[1095] 6. Rating and Revenue Sharing Phase

[1096] Terminal: The user rates the displayed information, and the results are sent to the server.

[1097] Server: Distributes appropriate revenue to rankers based on user ratings. Collects rating data to help improve the algorithm.

[1098] Specific use cases

[1099] For example, if a user wants to gather information about "the most noteworthy smartphone technologies in 2023," the system works as follows:

[1100] 1. Information Collection:

[1101] A user types "2023 smartphone technology" into the search box and submits a request.

[1102] The server collects relevant information from the Internet.

[1103] 2. Data Preprocessing:

[1104] The server removes unnecessary HTML tags and advertisements from the collected information and converts it into pure text data.

[1105] 3. Information analysis:

[1106] The server uses generative AI to analyze the text and extract key keywords and topics, and it also uses fact-checking APIs to assess the reliability of the information.

[1107] 4. Ranking:

[1108] The server ranks the information based on the analysis and evaluation results, with the most important information being ranked at the top based on criteria such as reliability, recency, and relevance.

[1109] 5.Display:

[1110] The ranking information is displayed on the user's device, allowing the user to grasp important information at a glance.

[1111] 6. Ratings and Revenue Share:

[1112] The user rates the displayed information, and the results of this rating are sent to the server.

[1113] The server distributes revenue to rankers based on the evaluation data and uses it to optimize the algorithm.

[1114] This system allows users to efficiently gather the information they need, provides access to reliable information, and allows ranking creators to receive appropriate rewards.

[1115] The processing flow will be explained below.

[1116] Step 1:

[1117] Request reception

[1118] On the device: A user enters specific information into a search box and submits a search request, which includes search keywords, such as "smartphone technology in 2023."

[1119] Step 2:

[1120] Submitting a search query

[1121] Terminal: Sends the entered search query to the server. Here, the user's input is passed to the server as the search query.

[1122] Step 3:

[1123] Start gathering information

[1124] Server: Based on the received search query, the server launches a web crawler to collect relevant information from websites, blogs, social media, etc. on the Internet. It may also use specific APIs (such as news APIs) to obtain data.

[1125] Step 4:

[1126] Data normalization

[1127] Server: Normalizes the collected raw data. Specifically, it removes unnecessary elements such as unnecessary HTML tags, advertisements, and scripts from the information and extracts only the text information.

[1128] Step 5:

[1129] Text Cleaning

[1130] Server: Further cleans the normalized text, specifically removing line breaks and special characters, and arranging the text data in a consistent format, which improves the accuracy of text analysis.

[1131] Step 6:

[1132] Natural Language Processing

[1133] Server: Analyzes collected text information using generative AI, extracts important keywords and topics, and gains a deep understanding of the text content through grammatical and semantic analysis.

[1134] Step 7:

[1135] Reliability evaluation

[1136] Server: Using fact-checking APIs and proprietary algorithms, the reliability of collected information is evaluated. A reliability score is calculated and information with a low score is removed.

[1137] Step 8:

[1138] Ranking algorithm applied

[1139] Server: Applies a ranking algorithm to rank the information based on criteria such as reliability, recency, and relevance. Sorting is performed based on the score of each piece of information.

[1140] Step 9:

[1141] Saving ranking results

[1142] Server: Stores the generated ranking results in a database for later display and analysis.

[1143] Step 10:

[1144] Sending ranking results

[1145] Server: Sends the ranking results stored on the user's device.

[1146] Step 11:

[1147] User Interface Display

[1148] Terminal: The ranking results are displayed in a user interface designed to allow users to easily access reliable information.

[1149] Step 12:

[1150] Accepting user reviews

[1151] Device: The user rates the displayed information. Specifically, the rating is completed by clicking a rating button such as "useful" or "trustworthy."

[1152] Step 13:

[1153] Submitting evaluation data

[1154] Terminal: Sends user evaluation data to the server, which reflects the user's evaluation in the system.

[1155] Step 14:

[1156] Revenue Share Calculation

[1157] Server: Based on user rating data, distributes appropriate revenue to rankers. Advertising revenue and sponsorship income are calculated according to the ratings, and the corresponding points and revenue are awarded to rankers.

[1158] Step 15:

[1159] Collecting feedback

[1160] Server: Collects and analyzes user evaluation data and behavioral data, which helps identify areas for system improvement.

[1161] Step 16:

[1162] Algorithm Updates

[1163] Server: Optimizes the ranking algorithm based on the collected feedback data. Uses adaptive learning to improve the accuracy of ranking results from next time onwards.

[1164] This system allows users to efficiently gather the information they need and access reliable information, while also allowing rankers to receive appropriate rewards.

[1165] Example 1

[1166] 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."

[1167] Conventional information collection and analysis systems often lack the means to efficiently collect large amounts of information and evaluate it based on its reliability. As a result, it is difficult for users to quickly access the reliable information they need. Furthermore, there are issues with efficiency and accuracy in the process of verifying the accuracy of the collected information and generating rankings based on it. Furthermore, there is a lack of means to optimize the system itself using user evaluation data.

[1168] 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.

[1169] In this invention, the server includes: a means for a user to enter keywords into a search box on a terminal and send a request; a means for collecting information from a digital network based on the search request; a means for normalizing the collected information and removing unnecessary data to convert it into pure text data; a means for analyzing the collected text data using a generative AI model and extracting key keywords and concepts; a means for evaluating the reliability of information using a fact-checking API; a means for ranking information based on the analyzed and evaluated information; a means for displaying the ranking results on the user's terminal; and a means for collecting user evaluation data and distributing revenue to rankers. This enables efficient collection and analysis of large amounts of information, and evaluation and ranking based on reliability. Furthermore, utilizing user evaluation data enables system optimization and enables the provision of more accurate information.

[1170] "User" refers to an entity that uses the system to search for and evaluate information.

[1171] "Device" refers to a device used by a User, such as a computer, smartphone, or tablet.

[1172] A "search box" is an interface element that allows a user to enter search keywords.

[1173] A "request" refers to a search or operation request sent by a user from a terminal to a server.

[1174] A "server" is a collection of hardware and software that receives requests from users and collects and processes information.

[1175] "Digital network" refers to the entire system for electronic information exchange, including the Internet.

[1176] "Means of collecting information" refers to methods of obtaining the necessary data from the Internet using web crawlers or API access.

[1177] "Normalization" refers to the process of removing unnecessary elements from collected data and converting it into a unified format.

[1178] "Unnecessary data" refers to elements of collected information, such as HTML tags and advertisements, that may reduce the accuracy of search results.

[1179] "Pure text data" refers to information after it has been normalized and converted into a parsable text format.

[1180] "Generative AI model" refers to an artificial intelligence model for analyzing collected text data.

[1181] "Means for analyzing text data" refers to a method of using a generative AI model to analyze collected text data and extract key keywords and concepts.

[1182] "Major keywords and concepts" refer to information elements that are particularly important within the text data and serve as the basis for ranking.

[1183] A "fact-checking API" refers to an interface for external services that allow users to verify the reliability of information.

[1184] "Means for evaluating the reliability of information" refers to methods for scoring the veracity and reliability of collected information using fact-checking APIs, etc.

[1185] "Ranking" refers to the process of ranking information according to certain criteria (such as reliability, recency, relevance, etc.) based on evaluated information.

[1186] "Ranking results" refers to a collection of ranked information and a list of ranked information displayed to users.

[1187] "User rating data" refers to the rating a user gives to the displayed information, including feedback such as star ratings and comments.

[1188] "Ranker" refers to the person who creates the ranking or receives compensation based on the ranking.

[1189] "Means of revenue distribution" refers to a method of distributing appropriate rewards to rankers based on user evaluation data.

[1190] MODE FOR CARRYING OUT THE INVENTION

[1191] The system of the present invention is designed to enable users to efficiently collect reliable information. The system automatically performs a series of processes, from processing search requests to collecting, normalizing, analyzing, ranking, and displaying information to users. Furthermore, it collects user evaluation data and uses it for system optimization and revenue distribution.

[1192] Information collection and preprocessing

[1193] When a user enters a specific keyword (e.g., "Smartphone technology in 2023") into the device's search box and presses the search button, the server receives the request. Based on the request, the server uses a web crawler (e.g., Scrapy) to retrieve information from digital networks. It also collects the latest information from social media and search engines through API access (e.g., Twitter API, Google Custom Search API).

[1194] The collected information is normalized on the server using libraries such as Python's BeautifulSoup, which removes unnecessary HTML tags and advertisements and converts it into pure text data.

[1195] Information analysis

[1196] The normalized text data is analyzed by a generative AI model (e.g., OpenAI's GPT-4). The generative AI model is used to extract key keywords and concepts from the collected text data. At the same time, the reliability of the information is evaluated using a fact-checking API (e.g., Factmata API). The fact-checking API is used to score the veracity and reliability of the information.

[1197] Ranking and Display

[1198] Once the information has been analyzed and evaluated, it is ranked based on criteria such as reliability, recency, and relevance. In this ranking process, an overall score is assigned to each piece of information, and the information is sorted in descending order of its score. The ranking results are stored in a database on the server and displayed on the user's device. Users can use this ranking to quickly access the information they need.

[1199] To give a concrete example, let's say a user searches for information on "smartphone technologies to watch in 2023." Here's how the system works:

[1200] 1. A user types "2023 smartphone technology" into the search box and submits a request.

[1201] 2. The server collects relevant information from the Internet.

[1202] 3. The server removes unnecessary HTML tags and advertisements from the collected information and converts it into pure text data.

[1203] 4. The server analyzes the text using a generative AI model (e.g., GPT-4) to extract key keywords and topics, and also uses a fact-checking API to assess the reliability of the information.

[1204] 5. The server ranks the information based on the analysis and evaluation results. The most important information is ranked at the top based on criteria such as reliability, recency, and relevance.

[1205] 6. Ranking information is displayed on the user's device, allowing the user to grasp important information at a glance.

[1206] An example of a prompt sentence might be:

[1207] "Please explain the specific information gathering procedures for the system that allows users to gather information about 'notable smartphone technologies in 2023.'"

[1208] This system allows users to efficiently gather the information they need and quickly access reliable information. Furthermore, by using user evaluation data, the system can be optimized to provide even higher quality information.

[1209] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1210] Step 1: Receiving a search request

[1211] Device: A user enters a specific keyword (e.g., "smartphone technology 2023") into the device's search box and presses the search button. This action generates a search request.

[1212] Input: The search keyword entered by the user.

[1213] Output: The search request sent to the server.

[1214] Step 2: Gather information

[1215] Server: Based on the received search request, launch a web crawler (e.g., Scrapy) to collect relevant information from the Internet, and then use API access (e.g., Twitter API, Google Custom Search API) to obtain relevant data.

[1216] Input: Keywords based on your search request.

[1217] Output: Collected data from web pages and APIs.

[1218] Step 3: Normalize the data

[1219] Server: To normalize the collected information, we use Python's BeautifulSoup library to remove unnecessary HTML tags and advertisements and convert it into pure text data.

[1220] Input: Raw data collected from web pages and APIs.

[1221] Output: Cleaned and normalized text data.

[1222] Step 4: Analyzing the text data

[1223] Server: Analyzes the normalized text data using a generative AI model (e.g., OpenAI's GPT-4), extracting key keywords and concepts.

[1224] Input: Normalized text data.

[1225] Output: Extracted keywords and concepts.

[1226] Step 5: Assess reliability

[1227] Server: Use a fact-checking API (e.g., Factmata API) to evaluate the reliability of the extracted information. Using this API, the truthfulness and reliability score of the information can be obtained.

[1228] Input: Keywords and concepts extracted by the generative AI model.

[1229] Output: Evaluation results including confidence scores.

[1230] Step 6: Ranking the information

[1231] Server: Based on the analyzed and evaluated information, the information is ranked based on specific criteria (reliability, freshness, relevance, etc.) and the rankings are stored in a SQL database or similar.

[1232] Input: Confidence score and evaluation results.

[1233] Output: A list of information in ranking format.

[1234] Step 7: View ranking information

[1235] Device: Ranking information is displayed on the user's device, allowing users to quickly access important information.

[1236] Input: A ranking list of information sent from the server.

[1237] Output: A viewable list of information displayed on the user's terminal.

[1238] Step 8: Collecting ratings and sharing revenue

[1239] Terminal: The user rates the displayed information, and the results are sent to the server.

[1240] Input: User rating data (star ratings, comments, etc.).

[1241] Output: The rating data sent to the server.

[1242] Server: Receives user evaluation data and distributes revenue to rankers based on that data. Additionally, the collected evaluation data is used to optimize the system.

[1243] Input: User rating data.

[1244] Output: Revenue distribution results to rankers and algorithm improvements using evaluation data.

[1245] In this way, the system performs specific data processing and calculations at each step, enabling users to access reliable information in a short time, and also optimizes the system itself by utilizing user evaluation data.

[1246] (Application example 1)

[1247] 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."

[1248] A major issue facing modern online shopping sites is the low reliability of product reviews. Many reviews are inaccurate or exaggerated, making it difficult for users to make appropriate decisions. This has created a need for users to efficiently find reliable reviews. Furthermore, existing ranking systems tend to rely on subjective evaluations and lack objective evaluations of reliability. This has led to major challenges in improving the reliability of reviews and user satisfaction.

[1249] 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.

[1250] In this invention, the server includes means for collecting and normalizing information from a digital network using a generative AI, means for analyzing the collected information using natural language processing and evaluating its reliability, means for ranking the analyzed and evaluated information based on criteria, means for displaying the information in a ranking format on the user's terminal, means for collecting user evaluation data and distributing revenue to rankers, and means for collecting product reviews and displaying highly reliable reviews at the top, thereby enabling users to efficiently access highly reliable product reviews.

[1251] "Generative AI" refers to generative artificial intelligence, a technology that generates new information and answers based on large amounts of data.

[1252] "Normalization" refers to the process of removing unnecessary information and formats from collected data.

[1253] "Natural language processing" refers to the technology that enables computers to understand, analyze, and process human language.

[1254] "Assessing trustworthiness" is the process of determining how accurate or relevant the information collected is.

[1255] "Ranking" refers to ranking data according to specific criteria.

[1256] "User's terminal" refers to a device used by a user, such as a smartphone or computer.

[1257] "User Ratings Data" means feedback and rating information provided by Users.

[1258] A "ranker" refers to the person or system that creates rankings based on collected information and evaluation data.

[1259] "Means for distributing revenue" refers to a mechanism for appropriately distributing revenue generated based on the generated information and rankings.

[1260] "Product reviews" refer to the impressions and evaluation data left by users after using a product.

[1261] A "reliable review" is a review that contains objective and accurate information and is deemed to be useful to other users.

[1262] This invention is a system that uses generative AI technology to collect reliable product reviews and provide them to users in a ranking format. This system consists of the following steps: information collection, data preprocessing, information analysis, reliability evaluation, ranking generation, display, user evaluation, and revenue distribution.

[1263] The server first uses generative AI to collect product review information from digital networks and normalizes it by removing unnecessary elements such as HTML tags and advertisements. Libraries such as BeautifulSoup and requests are used in this process. The collected data is then analyzed using natural language processing (NLP) to extract important keywords and topics. Software such as SpaCy and transformers are used in this stage.

[1264] The server evaluates the reliability of the reviews based on the analysis results and assigns them a score using a fact-checking API. Based on the obtained score, the reviews are ranked according to criteria. Data analysis tools such as scikit-learn are used to generate the rankings. Once the reliability evaluation and ranking are complete, the ranking results are sent to the user's device, where they can be viewed on their smartphone or other device.

[1265] Users rate the displayed ranking results, and the rating data is sent to the server. The server then optimizes the ranking algorithm based on the user rating data and distributes revenue to the rankers. As a concrete example, when a user searches for "latest smartphones," reliable reviews are displayed at the top. For example, specific reviews such as "This smartphone has a long battery life" are displayed in a prominent position.

[1266] An example of a prompt is, "Please collect reviews of the latest smartphones, rate their reliability, and rank them." By using such a prompt, users can efficiently obtain the information they need.

[1267] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1268] Step 1:

[1269] Information gathering

[1270] The server collects relevant review information from the digital network based on the keywords entered by the user in the search box. Specifically, it retrieves the information from the relevant product's web page or review site using the requests library. The retrieved information is saved on the server in HTML format.

[1271] Input: User's search keyword

[1272] Output: Collected review information in HTML format

[1273] Step 2:

[1274] Data Preprocessing

[1275] The server normalizes the collected HTML review information by using the BeautifulSoup library to remove unnecessary elements such as HTML tags and advertisements, converting it into pure text data, and removing line spacing, whitespace, and special characters.

[1276] Input: Collected review information in HTML format

[1277] Output: Normalized text data

[1278] Step 3:

[1279] Natural Language Processing (NLP) Analysis

[1280] The server then performs natural language processing on the normalized text data using SpaCy or transformers, extracting important keywords and topics from the reviews and calculating features (such as TF-IDF) for each review.

[1281] Input: normalized text data

[1282] Output: Extracted keywords, topics, and features

[1283] Step 4:

[1284] Reliability evaluation

[1285] The server uses a fact-checking API to score the credibility of each review. Specifically, it sends the analyzed data to the fact-checking API and receives a credibility score, which quantifies the credibility of each review.

[1286] Input: Extracted keywords, topics, and features

[1287] Output: Confidence score

[1288] Step 5:

[1289] Ranking Generation

[1290] The server ranks reviews based on their credibility score. The ranking algorithm uses factors such as credibility score, recency, and relevance. The ranking is generated using libraries such as scikit-learn.

[1291] Inputs: Confidence score, recency, relevance

[1292] Output: Ranked reviews

[1293] Step 6:

[1294] Displaying ranking results

[1295] The device displays the ranking results sent from the server to the user, who can then view the reviews ranked in order of reliability on their smartphone or computer screen.

[1296] Input: Ranked reviews

[1297] Output: Ranking results displayed on the user's device

[1298] Step 7:

[1299] User ratings and data collection

[1300] Users rate the displayed ranking results, and the rating data is sent from the device to the server, including feedback on the reliability and usefulness of the reviews.

[1301] Input: User Rating

[1302] Output: Rating data sent to the server

[1303] Step 8:

[1304] Revenue sharing and algorithm optimization

[1305] The server optimizes the ranking algorithm based on the evaluation data collected from users, and also executes the process of distributing profits to rankers. Appropriate data analysis is performed to improve the accuracy of the ranking algorithm and distribute profits.

[1306] Input: User rating data

[1307] Output: Optimized ranking algorithm, revenue sharing results

[1308] 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.

[1309] This invention provides a system that combines generative AI and an emotion engine to efficiently collect, analyze, assess credibility, rank, evaluate users, and distribute revenue. Furthermore, by analyzing user emotions, it optimizes the ranking algorithm and dynamically adjusts the content displayed.

[1310] Overall system flow

[1311] 1. Information gathering stage

[1312] On the device: A user types specific information (e.g., "smartphone technology in 2023") into a search box and submits a request.

[1313] Server: Based on the search request, they gather relevant information from websites, blogs, and social media across their digital network, using crawling and API access to obtain data.

[1314] 2. Data preprocessing stage

[1315] Server: Normalizes the collected information and removes unnecessary elements such as unnecessary HTML tags and advertisements, resulting in pure text data.

[1316] 3. Information Analysis Stage

[1317] Server: Using generative AI, natural language processing is performed to analyze the collected text, extracting important keywords and topics, and evaluating the reliability of the information using fact-checking APIs.

[1318] 4. Ranking Stage

[1319] Server: Analyzes and evaluates the information to generate a ranking according to specific criteria, such as reliability, freshness, and relevance.

[1320] 5. Display Stage

[1321] Device: The generated ranking results are displayed on the user's device, allowing the user to view the ranked, reliable information.

[1322] 6. Rating and Revenue Sharing Phase

[1323] Terminal: The user rates the displayed information. The rating data is sent to the server.

[1324] Server: Distributes appropriate revenue to rankers based on user rating data. Collects rating data to help improve the algorithm.

[1325] 7. Sentiment Analysis Stage

[1326] Device: An emotion engine that recognizes the user's emotions runs and collects the user's emotion data.

[1327] Example: An emotion engine analyzes in real time whether a user is very interested, angry, sad, etc.

[1328] 8. Sentiment data analysis and algorithm optimization

[1329] Server: The ranking algorithm is optimized based on the collected emotional data, such as giving priority to information where the user expressed positive emotions.

[1330] Example: If the emotion engine analyzes the user's emotion as "interesting," it will rank similarly interesting information higher.

[1331] 9. Dynamic Content Adjustment

[1332] On the device: Dynamically adjust the content displayed based on the analysis results of the emotion engine, prioritizing the display of topics and related information that interest the user.

[1333] Example: If a user is determined to be interested in smartphone technology, the app will automatically show them new related articles and videos.

[1334] Specific use cases

[1335] For example, if a user wants to gather information about "the most noteworthy smartphone technologies in 2023," the system works as follows:

[1336] 1. Information Collection:

[1337] A user types "2023 smartphone technology" into the search box and submits a request.

[1338] The server collects relevant information from the Internet.

[1339] 2. Data Preprocessing:

[1340] The server removes unnecessary HTML tags and advertisements from the collected information and converts it into pure text data.

[1341] 3. Information analysis:

[1342] The server uses generative AI to analyze the text and extract key keywords and topics, and it also uses fact-checking APIs to assess the reliability of the information.

[1343] 4. Emotion analysis:

[1344] While users browse information, the emotion engine analyzes their emotions in real time and changes relevant information if the user is interested or dissatisfied.

[1345] 5. Ranking:

[1346] The server ranks the information based on the analysis and evaluation results, incorporating factors such as reliability, recency, and relevance, as well as sentiment data, to rank the most important information at the top.

[1347] 6. Dynamic content adjustment:

[1348] The content displayed is dynamically adjusted based on the user's emotional data, prioritizing new and relevant information that will interest the user.

[1349] 7. Ratings and Revenue Share:

[1350] Users rate the displayed information, and the results are sent to the server, which distributes revenue to rankers based on the rating data and uses it to optimize the algorithm.

[1351] This system allows users to efficiently gather the information they need and access reliable information. By taking user sentiment into consideration, it also enables more personalized information to be provided, and rankers can receive appropriate rewards.

[1352] The processing flow will be explained below.

[1353] Step 1:

[1354] Request reception

[1355] On the device: A user enters specific information into a search box and submits a search request, which includes search keywords, such as "smartphone technology in 2023."

[1356] Example: A user searches for "smartphone technology 2023."

[1357] Step 2:

[1358] Submitting a search query

[1359] Terminal: Sends the entered search query to the server.

[1360] Step 3:

[1361] Start gathering information

[1362] Server: Based on the received search query, the server launches a web crawler to collect relevant information from websites, blogs, social media, etc. on the Internet. It may also use specific APIs (such as news APIs) to obtain data.

[1363] Example: The server crawls news sites and social media sites using the keyword "smartphone technology 2023" to collect related information.

[1364] Step 4:

[1365] Data normalization

[1366] Server: The collected raw data is normalized. Specifically, unnecessary elements such as unnecessary HTML tags, advertisements, and scripts are removed from the information, and only the text information is extracted.

[1367] Example: Removing advertisements and unnecessary HTML tags from aggregated news articles and blog posts.

[1368] Step 5:

[1369] Text Cleaning

[1370] Server: Further cleans the normalized text, specifically removing line breaks and special characters, and arranging the text data in a consistent format, which improves the accuracy of text analysis.

[1371] For example: cleaning text by removing unnecessary spaces and special characters.

[1372] Step 6:

[1373] Natural Language Processing

[1374] Server: Analyzes collected text information using generative AI, extracts important keywords and topics, and gains a deep understanding of the text content through grammatical and semantic analysis.

[1375] Example: Extracting keywords related to smartphone technology (5G, camera resolution, battery life, etc.).

[1376] Step 7:

[1377] Reliability evaluation

[1378] Server: Using fact-checking APIs and proprietary algorithms, the server evaluates the reliability of collected information, calculates a reliability score, and filters out information with a low score.

[1379] Example: Calculate a credibility score for each source and filter out information with a low score.

[1380] Step 8:

[1381] Ranking algorithm applied

[1382] Server: Applies a ranking algorithm to rank the information based on criteria such as reliability, recency, and relevance. Sorting is performed based on the score of each piece of information.

[1383] Example: Combining reliability, recency, and relevance scores and ranking the information with the highest score at the top.

[1384] Step 9:

[1385] Saving ranking results

[1386] Server: Stores the generated ranking results in a database for later display and analysis.

[1387] Example: Storing ranked smartphone technology information in a database.

[1388] Step 10:

[1389] Sending ranking results

[1390] Server: Sends the ranking results stored on the user's device.

[1391] Example: Sending information about ranked smartphone technologies to the user's device.

[1392] Step 11:

[1393] User Interface Display

[1394] Terminal: The ranking results are displayed on the user interface, allowing users to access reliable information through the displayed rankings.

[1395] Example: A smartphone screen displays a list of reliable smartphone technology information.

[1396] Step 12:

[1397] Accepting user reviews

[1398] Device: The user rates the displayed information. Specifically, the rating is completed by pressing a rating button such as "useful" or "trustworthy."

[1399] Example: A user rates this review as helpful.

[1400] Step 13:

[1401] Submitting evaluation data

[1402] Terminal: Sends user evaluation data to the server, which reflects the user's evaluation in the system.

[1403] Example: User rating data is sent to the server.

[1404] Step 14:

[1405] Revenue Share Calculation

[1406] Server: Based on user rating data, distributes appropriate revenue to rankers. Advertising revenue and sponsorship income are calculated according to the ratings, and the corresponding points and revenue are awarded to rankers.

[1407] Example: Revenues are distributed to rankers based on user ratings.

[1408] Step 15:

[1409] Collecting feedback

[1410] Server: Collects and analyzes user evaluation data and behavioral data, which helps identify areas for system improvement.

[1411] Example: Collecting evaluation data to help improve the system.

[1412] Step 16:

[1413] Algorithm Updates

[1414] Server: Optimizes the ranking algorithm based on the collected feedback data. Uses adaptive learning to improve the accuracy of ranking results from next time onwards.

[1415] Example: Learn the characteristics of highly rated articles and reflect them in future rankings.

[1416] Step 17:

[1417] Emotion analysis

[1418] On-device: Utilizing an emotion engine to recognize the user's emotional state in real time, including interest, excitement, anger, sadness, etc.

[1419] Example: An emotion engine analyzes user emotions in real time while the user is browsing information.

[1420] Step 18:

[1421] Sending emotional data

[1422] Terminal: Sends the user's emotional data to the server.

[1423] Example: The analyzed emotion data is sent to the server.

[1424] Step 19:

[1425] Sentiment data analysis and algorithm optimization

[1426] Server: The ranking algorithm is optimized based on the collected emotional data, such as giving priority to information where the user expressed positive emotions.

[1427] For example, if the emotion engine analyzes the user's emotion as "interesting," it will rank similarly interesting information higher.

[1428] Step 20:

[1429] Dynamic Content Adjustment

[1430] On the device: Dynamically adjust the content displayed based on the analysis results of the emotion engine, prioritizing the display of topics and related information that interest the user.

[1431] Example: If a user expresses an interest in smartphone technology, automatically show them new related articles and videos.

[1432] This system allows users to efficiently gather the information they need and access reliable information. By taking user sentiment into consideration, it also enables more personalized information to be provided, and rankers can receive appropriate rewards.

[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 headset type terminal 314 will be referred to as a "terminal."

[1435] Conventional information gathering and organization systems have a process for ranking collected information based on its reliability, recency, and relevance, but they lack optimization that takes into account user emotions and evaluation data, limiting their ability to improve user experience. In particular, they are unable to reflect users' emotional reactions to content evaluations in real time, making it difficult to provide personalized information. Furthermore, there are also insufficient means for automatically evaluating and scoring the reliability of collected information, creating a risk of misinformation being included.

[1436] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1437] In this invention, the server includes means for collecting and normalizing information from a digital network using generative AI, means for analyzing the collected information using natural language processing and evaluating its reliability, means for ranking the analyzed and evaluated information based on criteria, means for displaying the information in a ranking format on the user's terminal, means for collecting user evaluation data and distributing revenue to rankers, means for analyzing user emotions using an emotion engine, means for optimizing the ranking algorithm based on the analyzed emotion data, and means for dynamically adjusting the content to be displayed based on the emotion data. This enables the provision of personalized information based on the user's emotion and evaluation data, resulting in more reliable information and an improved user experience.

[1438] "Generative AI" is a type of artificial intelligence technology used for natural language processing and data analysis, and has the ability to generate, convert, and analyze information from a wide range of data.

[1439] "Digital network" refers to the Internet and other digital communications networks that enable the collection, sharing, and transfer of information.

[1440] "Normalization" refers to the process of removing unnecessary data (e.g., HTML tags and advertisements) from collected information and converting it into pure text data.

[1441] "Natural language processing" refers to the technology of analyzing collected text data, extracting keywords and topics, and understanding the meaning and relationships of the information.

[1442] "Credibility assessment" is the process of assessing whether collected information is accurate, and includes assigning a credibility score using fact-checking APIs, etc.

[1443] "Ranking" refers to the process of analyzing and evaluating information and ranking it based on certain criteria (reliability, recency, relevance).

[1444] "User rating data" refers to data including ratings (e.g., star ratings, comments, etc.) made by users on information.

[1445] "Revenue sharing" refers to the process of distributing appropriate rewards to information providers based on users' rating data.

[1446] An "emotion engine" refers to technology for analyzing emotions from a user's facial expressions, voice, etc., and is used to collect and analyze emotional data in real time.

[1447] "Dynamic adjustment" refers to the process of changing and adjusting displayed content in real time based on the results of user sentiment analysis.

[1448] A "fact-checking API" is an application program interface used to assess the reliability of collected information and is a tool for fact-checking information.

[1449] This invention provides a system that combines generative AI and an emotion engine to efficiently collect, analyze, assess reliability, rank, and distribute revenues to users. Additionally, analyzing user emotions can optimize the ranking algorithm and dynamically adjust the content displayed. Below, we will explain each element of the system and how they are combined.

[1450] First, the device accepts the user's search request. The device has a search box, and the user enters the keyword of the information they want to find (e.g., "Smartphone technology in 2023") into this search box. The device then sends the search request to the server using the HTTP protocol.

[1451] Based on the search request received, the server collects relevant information from websites, blogs, social media, and other sources across the digital network, using crawling tools such as Apache Nutch and services such as the Twitter API.

[1452] The server then normalizes the collected information by using an HTML parser such as BeautifulSoup to remove unnecessary HTML tags and advertisements from the collected web pages and convert them into pure text data.

[1453] A generative AI model (e.g., GPT-4) runs on a server and analyzes the collected and normalized text data. During this analysis, natural language processing techniques are used to extract key keywords and topics, and fact-checking APIs (e.g., Snopes API) are used to assess the reliability of the information.

[1454] The server then generates a ranking based on the analyzed and evaluated information according to certain criteria (authority, recency, relevance), using a modified version of the PageRank algorithm or a recommendation system for this ranking process.

[1455] The ranking results are displayed on the user's device. The data is presented in a visually easy-to-read format using HTML and CSS. For example, the "latest articles about AI camera technology" could be displayed at the top.

[1456] Users rate the displayed information (e.g., star ratings or comments). The rating data is sent to the server, which then distributes appropriate revenue to the rankers based on the rating data. This rating data is also used to optimize the algorithm.

[1457] The user's device is equipped with an emotion engine, which uses technologies such as Emotion API to collect emotional data in real time from the user's facial expressions and voice. For example, it analyzes whether the user is expressing the emotion of "joy."

[1458] The server optimizes the ranking algorithm based on the collected sentiment data, for example, adjusting the algorithm to prioritize ranking information that users find "interesting," thereby improving the user experience.

[1459] Furthermore, the content displayed can be dynamically adjusted based on emotional data. For example, if it is determined that a user is interested in "AI camera technology," new related articles and videos will be dynamically displayed.

[1460] This system allows users to efficiently gather the information they need and access reliable information. It also enables personalized information provision that takes into account the user's emotions, and information providers can receive appropriate rewards.

[1461] Specific prompt examples:

[1462] "Please explain in detail the process flow for a user wanting to gather information about smartphone technology in 2023."

[1463] By combining generative AI models and emotion analysis technology, this system can respond quickly and appropriately to diverse user needs, significantly improving the quality and efficiency of information provision.

[1464] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1465] Step 1:

[1466] User: Enters "Smartphone technology in 2023" into the search box on the device and sends a search request, which generates search keywords as input data and sends them to the server using the HTTP protocol.

[1467] Input: Search keyword "Smartphone technology in 2023"

[1468] Output: Search request

[1469] Step 2:

[1470] Server: Based on the received search request, the server uses crawling tools such as Apache Nutch and Twitter APIs to collect relevant information from websites, blogs, and social media on the digital network. The collected information is stored as text data.

[1471] Input: Search request

[1472] Output: Collected information (text data)

[1473] Step 3:

[1474] Server: Normalize the collected information. Use an HTML parser such as BeautifulSoup to remove unnecessary HTML tags and advertisements and convert it into pure text data. This data processing results in clean text data.

[1475] Input: Collected information (text data)

[1476] Output: Normalized text data

[1477] Step 4:

[1478] Server: Uses a generative AI model (e.g., GPT-4) to extract key keywords and topics from the normalized text data, and uses fact-checking APIs (e.g., Snopes API) to assess the reliability of the information, resulting in analytics data and a reliability score.

[1479] Input: normalized text data

[1480] Output: Analysis data (major keywords, topics), reliability score

[1481] Step 5:

[1482] Server: Based on the analyzed information and the reliability score, it generates a ranking according to certain criteria (reliability, recency, relevance). This ranking is done using a modified version of the PageRank algorithm or a recommendation system.

[1483] Input: Analysis data (main keywords, topics), reliability score

[1484] Output: Ranking results

[1485] Step 6:

[1486] Device: The ranking results are visually formatted using HTML and CSS and displayed to the user. Users can view the ranking information on their devices.

[1487] Input: Ranking results

[1488] Output: Visualized ranking results (HTML, CSS)

[1489] Step 7:

[1490] User: Rates the displayed information (e.g., star ratings or comments). The rating data is sent to the server using the HTTP protocol.

[1491] Input: Visualized ranking results

[1492] Output: Rating data (star ratings, comments)

[1493] Step 8:

[1494] Server: Based on the collected evaluation data, the server distributes appropriate revenue to rankers. This evaluation data is also used to optimize the ranking algorithm.

[1495] Input: Evaluation data

[1496] Output: Revenue share data, optimized ranking algorithm

[1497] Step 9:

[1498] Device: Runs the emotion engine and collects emotional data from the user's facial expressions and voice. For example, it uses the Emotion API to analyze emotions such as "happiness," "interest," and "dissatisfaction" in real time.

[1499] Input: User's facial expressions and voice

[1500] Output: Emotion data

[1501] Step 10:

[1502] Server: The ranking algorithm is further optimized based on the data obtained from the sentiment analysis. For example, the algorithm is adjusted to prioritize ranking information that users find "interesting."

[1503] Input: Emotion data

[1504] Output: Optimized ranking algorithm

[1505] Step 11:

[1506] On-device: Dynamically adjusts content displayed based on sentiment analysis. For example, if a user is determined to be interested in "AI camera technology," relevant new articles and videos will be displayed.

[1507] Input: Sentiment analysis results

[1508] Output: Dynamically adjusted content placement

[1509] (Application example 2)

[1510] 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."

[1511] Conventional information gathering and ranking systems have had difficulty providing dynamic content that takes user sentiment into account, resulting in a lack of optimization for the user experience. Furthermore, while there is a demand for reliable information, there are limited methods for evaluating the accuracy and reliability of collected information in real time. Furthermore, while there is a demand for quickly optimizing ranking algorithms using user evaluation data and providing personalized information, there has been no system that combines sentiment analysis.

[1512] 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.

[1513] In this invention, the server includes means for collecting and normalizing information from a digital network using generative AI, means for analyzing the collected information using natural language processing and evaluating its reliability, means for ranking the analyzed and evaluated information based on criteria, means for displaying the information in a ranking format on the user's terminal, means for collecting user evaluation data and distributing revenue to rankers, means for analyzing user emotions in real time using a sentiment analysis engine, means for optimizing the ranking algorithm based on the collected sentiment data, and means for dynamically adjusting the content to be displayed. This makes it possible to provide efficient information taking user emotions into consideration, identify reliable information in a ranking format, and dynamically provide personalized information.

[1514] "Generative AI" is an artificial intelligence technology that collects information from digital networks and analyzes and generates data using natural language processing.

[1515] "Normalization" is the process of removing unnecessary elements from collected data and converting it into pure text data that is easy to analyze.

[1516] "Natural language processing" is a technology that allows computers to understand and analyze human language.

[1517] "Assessing credibility" is a method for determining the legitimacy or reliability of collected information.

[1518] "Ranking" refers to the process of ranking collected and analyzed information according to specific criteria.

[1519] An "emotion analysis engine" is a technology that analyzes user emotions in real time and collects them as data.

[1520] "Dynamic adjustment" is the process of instantly changing and updating displayed content based on data collected in real time.

[1521] "Distributing revenue" means distributing appropriate rewards to content providers and others based on user evaluation data.

[1522] This invention is a system that combines generative AI and a sentiment analysis engine to efficiently collect, analyze, assess credibility, rank, evaluate users' ratings, analyze sentiment, and dynamically adjust content. The following hardware and software are required for implementation:

[1523] Hardware and software used

[1524] Hardware: Camera (to capture the user's facial expressions), microphone (to capture the user's voice), user device (such as a smartphone or tablet).

[1525] Software: Generative AI model (for information analysis), TextBlob (for natural language processing), TensorFlow / Keras (for sentiment analysis model), REST API (for information collection).

[1526] What the program does

[1527] The server uses generative AI to collect information from digital networks, obtains data through crawling and API access, and normalizes this information, removing unnecessary HTML tags and advertisements, converting it into pure text data.

[1528] Furthermore, the collected text data is analyzed using natural language processing software (such as TextBlob), important keywords and topics are extracted, and the reliability of the information is evaluated through a fact-checking API. The evaluated information is ranked according to specific criteria and displayed in a ranked format on the user's device.

[1529] User sentiment is analyzed in real time using a sentiment analysis engine (such as TensorFlow / Keras models), and the ranking algorithm is optimized based on the collected sentiment data. The content displayed is dynamically adjusted according to the user's interests and sentiment.

[1530] Specific examples

[1531] For example, consider a case where a user wants to gather information about the latest VR technology. The user enters keywords into the search box, and the system collects related information from the internet. The collected information is then analyzed using natural language processing to extract important keywords and topics. The reliability of the information is evaluated based on a fact-checking API, and the results are displayed to the user in a ranked format.

[1532] While users are browsing information, the sentiment analysis engine analyzes their facial expressions and voice in real time. Information that is determined to be of interest to the user is displayed preferentially, and related information is dynamically adjusted, allowing users to quickly and efficiently obtain the content that interests them most.

[1533] Prompt Sentence Examples

[1534] "Analyze the following text and extract key keywords and sentiment:"

[1535] Text to analyze: "I'm looking for information on the latest in VR technology."

[1536] Using this prompt, the generative AI model analyzes the text data and provides information according to the user's request, which is expected to significantly improve the user experience.

[1537] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1538] Step 1:

[1539] The user enters a search keyword from the terminal and sends a request.

[1540] Input: User's search keyword

[1541] Output: Search request

[1542] How it works: A user enters a keyword into the search box on their device and presses the submit button, which sends the keyword to the system.

[1543] Step 2:

[1544] The server uses generated AI to collect information related to the search keywords from the Internet.

[1545] Input: Search request

[1546] Output: Collected digital information (web pages, blogs, social media posts, etc.)

[1547] How it works: The server collects relevant information from various sources on the Internet (websites and APIs) and stores it in text format.

[1548] Step 3:

[1549] The server normalizes the collected information and removes unnecessary HTML tags and advertisements.

[1550] Input: Collected digital information

[1551] Output: Clean text data

[1552] How it works: The normalization process removes unnecessary elements such as HTML tags and advertisements, producing pure text data.

[1553] Step 4:

[1554] The server uses natural language processing to analyze the clean text data and extract important keywords and topics.

[1555] Input: clean text data

[1556] Output: Analysis results (important keywords, extracted topics)

[1557] What it does: It uses a natural language processing library such as TextBlob to parse the text and extract key keywords and topics.

[1558] Step 5:

[1559] The server uses a fact-checking API to assess the reliability of the collected information.

[1560] Input: clean text data

[1561] Output: Trustworthiness score

[1562] How it works: Scores the legitimacy and veracity of collected information through a fact-checking API.

[1563] Step 6:

[1564] The server analyzes and evaluates the information and then ranks it according to specific criteria.

[1565] Input: Analysis results, reliability evaluation score

[1566] Output: Ranked list of information

[1567] How it works: Applies an algorithm that ranks information based on criteria such as reliability, freshness, and relevance.

[1568] Step 7:

[1569] The terminal provides the user with information displayed in a ranking format.

[1570] Input: Ranked list of information

[1571] Output: Ranking information displayed in the user interface

[1572] How it works: The ranking results are displayed on the user's screen, allowing the user to visually confirm reliable information.

[1573] Step 8:

[1574] The device uses an emotion analysis engine to analyze the user's facial expressions and voice in real time.

[1575] Input: User's facial expression data, voice speech data

[1576] Output: Sentiment analysis results (e.g., interest, joy, dissatisfaction)

[1577] How it works: The camera and microphone are used to capture the user's facial expressions and voice, which are then analyzed in real time by an emotion analysis engine.

[1578] Step 9:

[1579] The server optimizes the ranking algorithm based on the collected sentiment data.

[1580] Input: Sentiment analysis results, ranking information list

[1581] Output: Optimized ranking information

[1582] How it works: Reflects user sentiment data and adjusts and optimizes the ranking algorithm.

[1583] Step 10:

[1584] The device dynamically adjusts the content displayed based on the user's interests and emotions.

[1585] Input: Optimized ranking information, sentiment analysis results

[1586] Output: Dynamically adjusted content

[1587] How it works: Changes the information shown based on your interests and automatically surfaces new relevant content and videos.

[1588] 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.

[1589] 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.

[1590] 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.

[1591] [Fourth embodiment]

[1592] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1593] 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.

[1594] 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).

[1595] 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.

[1596] 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.

[1597] 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).

[1598] 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.

[1599] 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.

[1600] 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.

[1601] 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.

[1602] 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.

[1603] 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.

[1604] 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."

[1605] This system uses generative AI to collect large amounts of data, analyzes it using natural language processing (NLP), and evaluates its reliability, allowing users to efficiently gather specific information in a short period of time and obtain highly reliable information in a ranked format.

[1606] Overall system flow

[1607] 1. Information gathering stage

[1608] On the device: The user enters specific information into a search box and submits a request.

[1609] Server: Based on the search request, they gather relevant information from websites, blogs, and social media across their digital network, using crawling and API access to obtain data.

[1610] 2. Data preprocessing stage

[1611] Server: Normalizes the collected information and removes unnecessary elements such as unnecessary HTML tags and advertisements, resulting in pure text data.

[1612] 3. Information Analysis Stage

[1613] Server: Using generative AI, natural language processing is performed to analyze the collected text, extracting important keywords and concepts, and evaluating the reliability of the information using a fact-checking API.

[1614] 4. Ranking Stage

[1615] Server: Analyzes and evaluates the information to generate a ranking according to specific criteria, such as reliability, freshness, and relevance.

[1616] 5. Display Stage

[1617] Device: The generated ranking results are displayed on the user's device, allowing the user to view the ranked, reliable information.

[1618] 6. Rating and Revenue Sharing Phase

[1619] Terminal: The user rates the displayed information, and the results are sent to the server.

[1620] Server: Distributes appropriate revenue to rankers based on user ratings. Collects rating data to help improve the algorithm.

[1621] Specific use cases

[1622] For example, if a user wants to gather information about "the most noteworthy smartphone technologies in 2023," the system works as follows:

[1623] 1. Information Collection:

[1624] A user types "2023 smartphone technology" into the search box and submits a request.

[1625] The server collects relevant information from the Internet.

[1626] 2. Data Preprocessing:

[1627] The server removes unnecessary HTML tags and advertisements from the collected information and converts it into pure text data.

[1628] 3. Information analysis:

[1629] The server uses generative AI to analyze the text and extract key keywords and topics, and it also uses fact-checking APIs to assess the reliability of the information.

[1630] 4. Ranking:

[1631] The server ranks the information based on the analysis and evaluation results, with the most important information being ranked at the top based on criteria such as reliability, recency, and relevance.

[1632] 5.Display:

[1633] The ranking information is displayed on the user's device, allowing the user to grasp important information at a glance.

[1634] 6. Ratings and Revenue Share:

[1635] The user rates the displayed information, and the results of this rating are sent to the server.

[1636] The server distributes revenue to rankers based on the evaluation data and uses it to optimize the algorithm.

[1637] This system allows users to efficiently gather the information they need, provides access to reliable information, and allows ranking creators to receive appropriate rewards.

[1638] The processing flow will be explained below.

[1639] Step 1:

[1640] Request reception

[1641] On the device: A user enters specific information into a search box and submits a search request, which includes search keywords, such as "smartphone technology in 2023."

[1642] Step 2:

[1643] Submitting a search query

[1644] Terminal: Sends the entered search query to the server. Here, the user's input is passed to the server as the search query.

[1645] Step 3:

[1646] Start gathering information

[1647] Server: Based on the received search query, the server launches a web crawler to collect relevant information from websites, blogs, social media, etc. on the Internet. It may also use specific APIs (such as news APIs) to obtain data.

[1648] Step 4:

[1649] Data normalization

[1650] Server: Normalizes the collected raw data. Specifically, it removes unnecessary elements such as unnecessary HTML tags, advertisements, and scripts from the information and extracts only the text information.

[1651] Step 5:

[1652] Text Cleaning

[1653] Server: Further cleans the normalized text, specifically removing line breaks and special characters, and arranging the text data in a consistent format, which improves the accuracy of text analysis.

[1654] Step 6:

[1655] Natural Language Processing

[1656] Server: Analyzes collected text information using generative AI, extracts important keywords and topics, and gains a deep understanding of the text content through grammatical and semantic analysis.

[1657] Step 7:

[1658] Reliability evaluation

[1659] Server: Using fact-checking APIs and proprietary algorithms, the reliability of collected information is evaluated. A reliability score is calculated and information with a low score is removed.

[1660] Step 8:

[1661] Ranking algorithm applied

[1662] Server: Applies a ranking algorithm to rank the information based on criteria such as reliability, recency, and relevance. Sorting is performed based on the score of each piece of information.

[1663] Step 9:

[1664] Saving ranking results

[1665] Server: Stores the generated ranking results in a database for later display and analysis.

[1666] Step 10:

[1667] Sending ranking results

[1668] Server: Sends the ranking results stored on the user's device.

[1669] Step 11:

[1670] User Interface Display

[1671] Terminal: The ranking results are displayed in a user interface designed to allow users to easily access reliable information.

[1672] Step 12:

[1673] Accepting user reviews

[1674] Device: The user rates the displayed information. Specifically, the rating is completed by clicking a rating button such as "useful" or "trustworthy."

[1675] Step 13:

[1676] Submitting evaluation data

[1677] Terminal: Sends user evaluation data to the server, which reflects the user's evaluation in the system.

[1678] Step 14:

[1679] Revenue Share Calculation

[1680] Server: Based on user rating data, distributes appropriate revenue to rankers. Advertising revenue and sponsorship income are calculated according to the ratings, and the corresponding points and revenue are awarded to rankers.

[1681] Step 15:

[1682] Collecting feedback

[1683] Server: Collects and analyzes user evaluation data and behavioral data, which helps identify areas for system improvement.

[1684] Step 16:

[1685] Algorithm Updates

[1686] Server: Optimizes the ranking algorithm based on the collected feedback data. Uses adaptive learning to improve the accuracy of ranking results from next time onwards.

[1687] This system allows users to efficiently gather the information they need and access reliable information, while also allowing rankers to receive appropriate rewards.

[1688] Example 1

[1689] 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."

[1690] Conventional information collection and analysis systems often lack the means to efficiently collect large amounts of information and evaluate it based on its reliability. As a result, it is difficult for users to quickly access the reliable information they need. Furthermore, there are issues with efficiency and accuracy in the process of verifying the accuracy of the collected information and generating rankings based on it. Furthermore, there is a lack of means to optimize the system itself using user evaluation data.

[1691] 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.

[1692] In this invention, the server includes: a means for a user to enter keywords into a search box on a terminal and send a request; a means for collecting information from a digital network based on the search request; a means for normalizing the collected information and removing unnecessary data to convert it into pure text data; a means for analyzing the collected text data using a generative AI model and extracting key keywords and concepts; a means for evaluating the reliability of information using a fact-checking API; a means for ranking information based on the analyzed and evaluated information; a means for displaying the ranking results on the user's terminal; and a means for collecting user evaluation data and distributing revenue to rankers. This enables efficient collection and analysis of large amounts of information, and evaluation and ranking based on reliability. Furthermore, utilizing user evaluation data enables system optimization and enables the provision of more accurate information.

[1693] "User" refers to an entity that uses the system to search for and evaluate information.

[1694] "Device" refers to a device used by a User, such as a computer, smartphone, or tablet.

[1695] A "search box" is an interface element that allows a user to enter search keywords.

[1696] A "request" refers to a search or operation request sent by a user from a terminal to a server.

[1697] A "server" is a collection of hardware and software that receives requests from users and collects and processes information.

[1698] "Digital network" refers to the entire system for electronic information exchange, including the Internet.

[1699] "Means of collecting information" refers to methods of obtaining the necessary data from the Internet using web crawlers or API access.

[1700] "Normalization" refers to the process of removing unnecessary elements from collected data and converting it into a unified format.

[1701] "Unnecessary data" refers to elements of collected information, such as HTML tags and advertisements, that may reduce the accuracy of search results.

[1702] "Pure text data" refers to information after it has been normalized and converted into a parsable text format.

[1703] "Generative AI model" refers to an artificial intelligence model for analyzing collected text data.

[1704] "Means for analyzing text data" refers to a method of using a generative AI model to analyze collected text data and extract key keywords and concepts.

[1705] "Major keywords and concepts" refer to information elements that are particularly important within the text data and serve as the basis for ranking.

[1706] A "fact-checking API" refers to an interface for external services that allow users to verify the reliability of information.

[1707] "Means for evaluating the reliability of information" refers to methods for scoring the veracity and reliability of collected information using fact-checking APIs, etc.

[1708] "Ranking" refers to the process of ranking information according to certain criteria (such as reliability, recency, relevance, etc.) based on evaluated information.

[1709] "Ranking results" refers to a collection of ranked information and a list of ranked information displayed to users.

[1710] "User rating data" refers to the rating a user gives to the displayed information, including feedback such as star ratings and comments.

[1711] "Ranker" refers to the person who creates the ranking or receives compensation based on the ranking.

[1712] "Means of revenue distribution" refers to a method of distributing appropriate rewards to rankers based on user evaluation data.

[1713] MODE FOR CARRYING OUT THE INVENTION

[1714] The system of the present invention is designed to enable users to efficiently collect reliable information. The system automatically performs a series of processes, from processing search requests to collecting, normalizing, analyzing, ranking, and displaying information to users. Furthermore, it collects user evaluation data and uses it for system optimization and revenue distribution.

[1715] Information collection and preprocessing

[1716] When a user enters a specific keyword (e.g., "Smartphone technology in 2023") into the device's search box and presses the search button, the server receives the request. Based on the request, the server uses a web crawler (e.g., Scrapy) to retrieve information from digital networks. It also collects the latest information from social media and search engines through API access (e.g., Twitter API, Google Custom Search API).

[1717] The collected information is normalized on the server using libraries such as Python's BeautifulSoup, which removes unnecessary HTML tags and advertisements and converts it into pure text data.

[1718] Information analysis

[1719] The normalized text data is analyzed by a generative AI model (e.g., OpenAI's GPT-4). The generative AI model is used to extract key keywords and concepts from the collected text data. At the same time, the reliability of the information is evaluated using a fact-checking API (e.g., Factmata API). The fact-checking API is used to score the veracity and reliability of the information.

[1720] Ranking and Display

[1721] Once the information has been analyzed and evaluated, it is ranked based on criteria such as reliability, recency, and relevance. In this ranking process, an overall score is assigned to each piece of information, and the information is sorted in descending order of its score. The ranking results are stored in a database on the server and displayed on the user's device. Users can use this ranking to quickly access the information they need.

[1722] To give a concrete example, let's say a user searches for information on "smartphone technologies to watch in 2023." Here's how the system works:

[1723] 1. A user types "2023 smartphone technology" into the search box and submits a request.

[1724] 2. The server collects relevant information from the Internet.

[1725] 3. The server removes unnecessary HTML tags and advertisements from the collected information and converts it into pure text data.

[1726] 4. The server analyzes the text using a generative AI model (e.g., GPT-4) to extract key keywords and topics, and also uses a fact-checking API to assess the reliability of the information.

[1727] 5. The server ranks the information based on the analysis and evaluation results. The most important information is ranked at the top based on criteria such as reliability, recency, and relevance.

[1728] 6. Ranking information is displayed on the user's device, allowing the user to grasp important information at a glance.

[1729] An example of a prompt sentence might be:

[1730] "Please explain the specific information gathering procedures for the system that allows users to gather information about 'notable smartphone technologies in 2023.'"

[1731] This system allows users to efficiently gather the information they need and quickly access reliable information. Furthermore, by using user evaluation data, the system can be optimized to provide even higher quality information.

[1732] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1733] Step 1: Receiving a search request

[1734] Device: A user enters a specific keyword (e.g., "smartphone technology 2023") into the device's search box and presses the search button. This action generates a search request.

[1735] Input: The search keyword entered by the user.

[1736] Output: The search request sent to the server.

[1737] Step 2: Gather information

[1738] Server: Based on the received search request, launch a web crawler (e.g., Scrapy) to collect relevant information from the Internet, and then use API access (e.g., Twitter API, Google Custom Search API) to obtain relevant data.

[1739] Input: Keywords based on your search request.

[1740] Output: Collected data from web pages and APIs.

[1741] Step 3: Normalize the data

[1742] Server: To normalize the collected information, we use Python's BeautifulSoup library to remove unnecessary HTML tags and advertisements and convert it into pure text data.

[1743] Input: Raw data collected from web pages and APIs.

[1744] Output: Cleaned and normalized text data.

[1745] Step 4: Analyzing the text data

[1746] Server: Analyzes the normalized text data using a generative AI model (e.g., OpenAI's GPT-4), extracting key keywords and concepts.

[1747] Input: Normalized text data.

[1748] Output: Extracted keywords and concepts.

[1749] Step 5: Assess reliability

[1750] Server: Use a fact-checking API (e.g., Factmata API) to evaluate the reliability of the extracted information. Using this API, the truthfulness and reliability score of the information can be obtained.

[1751] Input: Keywords and concepts extracted by the generative AI model.

[1752] Output: Evaluation results including confidence scores.

[1753] Step 6: Ranking the information

[1754] Server: Based on the analyzed and evaluated information, the information is ranked based on specific criteria (reliability, freshness, relevance, etc.) and the rankings are stored in a SQL database or similar.

[1755] Input: Confidence score and evaluation results.

[1756] Output: A list of information in ranking format.

[1757] Step 7: View ranking information

[1758] Device: Ranking information is displayed on the user's device, allowing users to quickly access important information.

[1759] Input: A ranking list of information sent from the server.

[1760] Output: A viewable list of information displayed on the user's terminal.

[1761] Step 8: Collecting ratings and sharing revenue

[1762] Terminal: The user rates the displayed information, and the results are sent to the server.

[1763] Input: User rating data (star ratings, comments, etc.).

[1764] Output: The rating data sent to the server.

[1765] Server: Receives user evaluation data and distributes revenue to rankers based on that data. Additionally, the collected evaluation data is used to optimize the system.

[1766] Input: User rating data.

[1767] Output: Revenue distribution results to rankers and algorithm improvements using evaluation data.

[1768] In this way, the system performs specific data processing and calculations at each step, enabling users to access reliable information in a short time, and also optimizes the system itself by utilizing user evaluation data.

[1769] (Application example 1)

[1770] 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."

[1771] A major issue facing modern online shopping sites is the low reliability of product reviews. Many reviews are inaccurate or exaggerated, making it difficult for users to make appropriate decisions. This has created a need for users to efficiently find reliable reviews. Furthermore, existing ranking systems tend to rely on subjective evaluations and lack objective evaluations of reliability. This has led to major challenges in improving the reliability of reviews and user satisfaction.

[1772] 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.

[1773] In this invention, the server includes means for collecting and normalizing information from a digital network using a generative AI, means for analyzing the collected information using natural language processing and evaluating its reliability, means for ranking the analyzed and evaluated information based on criteria, means for displaying the information in a ranking format on the user's terminal, means for collecting user evaluation data and distributing revenue to rankers, and means for collecting product reviews and displaying highly reliable reviews at the top, thereby enabling users to efficiently access highly reliable product reviews.

[1774] "Generative AI" refers to generative artificial intelligence, a technology that generates new information and answers based on large amounts of data.

[1775] "Normalization" refers to the process of removing unnecessary information and formats from collected data.

[1776] "Natural language processing" refers to the technology that enables computers to understand, analyze, and process human language.

[1777] "Assessing trustworthiness" is the process of determining how accurate or relevant the information collected is.

[1778] "Ranking" refers to ranking data according to specific criteria.

[1779] "User's terminal" refers to a device used by a user, such as a smartphone or computer.

[1780] "User Ratings Data" means feedback and rating information provided by Users.

[1781] A "ranker" refers to the person or system that creates rankings based on collected information and evaluation data.

[1782] "Means for distributing revenue" refers to a mechanism for appropriately distributing revenue generated based on the generated information and rankings.

[1783] "Product reviews" refer to the impressions and evaluation data left by users after using a product.

[1784] A "reliable review" is a review that contains objective and accurate information and is deemed to be useful to other users.

[1785] This invention is a system that uses generative AI technology to collect reliable product reviews and provide them to users in a ranking format. This system consists of the following steps: information collection, data preprocessing, information analysis, reliability evaluation, ranking generation, display, user evaluation, and revenue distribution.

[1786] The server first uses generative AI to collect product review information from digital networks and normalizes it by removing unnecessary elements such as HTML tags and advertisements. Libraries such as BeautifulSoup and requests are used in this process. The collected data is then analyzed using natural language processing (NLP) to extract important keywords and topics. Software such as SpaCy and transformers are used in this stage.

[1787] The server evaluates the reliability of the reviews based on the analysis results and assigns them a score using a fact-checking API. Based on the obtained score, the reviews are ranked according to criteria. Data analysis tools such as scikit-learn are used to generate the rankings. Once the reliability evaluation and ranking are complete, the ranking results are sent to the user's device, where they can be viewed on their smartphone or other device.

[1788] Users rate the displayed ranking results, and the rating data is sent to the server. The server then optimizes the ranking algorithm based on the user rating data and distributes revenue to the rankers. As a concrete example, when a user searches for "latest smartphones," reliable reviews are displayed at the top. For example, specific reviews such as "This smartphone has a long battery life" are displayed in a prominent position.

[1789] An example of a prompt is, "Please collect reviews of the latest smartphones, rate their reliability, and rank them." By using such a prompt, users can efficiently obtain the information they need.

[1790] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1791] Step 1:

[1792] Information gathering

[1793] The server collects relevant review information from the digital network based on the keywords entered by the user in the search box. Specifically, it retrieves the information from the relevant product's web page or review site using the requests library. The retrieved information is saved on the server in HTML format.

[1794] Input: User's search keyword

[1795] Output: Collected review information in HTML format

[1796] Step 2:

[1797] Data Preprocessing

[1798] The server normalizes the collected HTML review information by using the BeautifulSoup library to remove unnecessary elements such as HTML tags and advertisements, converting it into pure text data, and removing line spacing, whitespace, and special characters.

[1799] Input: Collected review information in HTML format

[1800] Output: Normalized text data

[1801] Step 3:

[1802] Natural Language Processing (NLP) Analysis

[1803] The server then performs natural language processing on the normalized text data using SpaCy or transformers, extracting important keywords and topics from the reviews and calculating features (such as TF-IDF) for each review.

[1804] Input: normalized text data

[1805] Output: Extracted keywords, topics, and features

[1806] Step 4:

[1807] Reliability evaluation

[1808] The server uses a fact-checking API to score the credibility of each review. Specifically, it sends the analyzed data to the fact-checking API and receives a credibility score, which quantifies the credibility of each review.

[1809] Input: Extracted keywords, topics, and features

[1810] Output: Confidence score

[1811] Step 5:

[1812] Ranking Generation

[1813] The server ranks reviews based on their credibility score. The ranking algorithm uses factors such as credibility score, recency, and relevance. The ranking is generated using libraries such as scikit-learn.

[1814] Inputs: Confidence score, recency, relevance

[1815] Output: Ranked reviews

[1816] Step 6:

[1817] Displaying ranking results

[1818] The device displays the ranking results sent from the server to the user, who can then view the reviews ranked in order of reliability on their smartphone or computer screen.

[1819] Input: Ranked reviews

[1820] Output: Ranking results displayed on the user's device

[1821] Step 7:

[1822] User ratings and data collection

[1823] Users rate the displayed ranking results, and the rating data is sent from the device to the server, including feedback on the reliability and usefulness of the reviews.

[1824] Input: User Rating

[1825] Output: Rating data sent to the server

[1826] Step 8:

[1827] Revenue sharing and algorithm optimization

[1828] The server optimizes the ranking algorithm based on the evaluation data collected from users, and also executes the process of distributing profits to rankers. Appropriate data analysis is performed to improve the accuracy of the ranking algorithm and distribute profits.

[1829] Input: User rating data

[1830] Output: Optimized ranking algorithm, revenue sharing results

[1831] 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.

[1832] This invention provides a system that combines generative AI and an emotion engine to efficiently collect, analyze, assess credibility, rank, evaluate users, and distribute revenue. Furthermore, by analyzing user emotions, it optimizes the ranking algorithm and dynamically adjusts the content displayed.

[1833] Overall system flow

[1834] 1. Information gathering stage

[1835] On the device: A user types specific information (e.g., "smartphone technology in 2023") into a search box and submits a request.

[1836] Server: Based on the search request, they gather relevant information from websites, blogs, and social media across their digital network, using crawling and API access to obtain data.

[1837] 2. Data preprocessing stage

[1838] Server: Normalizes the collected information and removes unnecessary elements such as unnecessary HTML tags and advertisements, resulting in pure text data.

[1839] 3. Information Analysis Stage

[1840] Server: Using generative AI, natural language processing is performed to analyze the collected text, extracting important keywords and topics, and evaluating the reliability of the information using fact-checking APIs.

[1841] 4. Ranking Stage

[1842] Server: Analyzes and evaluates the information to generate a ranking according to specific criteria, such as reliability, freshness, and relevance.

[1843] 5. Display Stage

[1844] Device: The generated ranking results are displayed on the user's device, allowing the user to view the ranked, reliable information.

[1845] 6. Rating and Revenue Sharing Phase

[1846] Terminal: The user rates the displayed information. The rating data is sent to the server.

[1847] Server: Distributes appropriate revenue to rankers based on user rating data. Collects rating data to help improve the algorithm.

[1848] 7. Sentiment Analysis Stage

[1849] Device: An emotion engine that recognizes the user's emotions runs and collects the user's emotion data.

[1850] Example: An emotion engine analyzes in real time whether a user is very interested, angry, sad, etc.

[1851] 8. Sentiment data analysis and algorithm optimization

[1852] Server: The ranking algorithm is optimized based on the collected emotional data, such as giving priority to information where the user expressed positive emotions.

[1853] Example: If the emotion engine analyzes the user's emotion as "interesting," it will rank similarly interesting information higher.

[1854] 9. Dynamic Content Adjustment

[1855] On the device: Dynamically adjust the content displayed based on the analysis results of the emotion engine, prioritizing the display of topics and related information that interest the user.

[1856] Example: If a user is determined to be interested in smartphone technology, the app will automatically show them new related articles and videos.

[1857] Specific use cases

[1858] For example, if a user wants to gather information about "the most noteworthy smartphone technologies in 2023," the system works as follows:

[1859] 1. Information Collection:

[1860] A user types "2023 smartphone technology" into the search box and submits a request.

[1861] The server collects relevant information from the Internet.

[1862] 2. Data Preprocessing:

[1863] The server removes unnecessary HTML tags and advertisements from the collected information and converts it into pure text data.

[1864] 3. Information analysis:

[1865] The server uses generative AI to analyze the text and extract key keywords and topics, and it also uses fact-checking APIs to assess the reliability of the information.

[1866] 4. Emotion analysis:

[1867] While users browse information, the emotion engine analyzes their emotions in real time and changes relevant information if the user is interested or dissatisfied.

[1868] 5. Ranking:

[1869] The server ranks the information based on the analysis and evaluation results, incorporating factors such as reliability, recency, and relevance, as well as sentiment data, to rank the most important information at the top.

[1870] 6. Dynamic content adjustment:

[1871] The content displayed is dynamically adjusted based on the user's emotional data, prioritizing new and relevant information that will interest the user.

[1872] 7. Ratings and Revenue Share:

[1873] Users rate the displayed information, and the results are sent to the server, which distributes revenue to rankers based on the rating data and uses it to optimize the algorithm.

[1874] This system allows users to efficiently gather the information they need and access reliable information. By taking user sentiment into consideration, it also enables more personalized information to be provided, and rankers can receive appropriate rewards.

[1875] The processing flow will be explained below.

[1876] Step 1:

[1877] Request reception

[1878] On the device: A user enters specific information into a search box and submits a search request, which includes search keywords, such as "smartphone technology in 2023."

[1879] Example: A user searches for "smartphone technology 2023."

[1880] Step 2:

[1881] Submitting a search query

[1882] Terminal: Sends the entered search query to the server.

[1883] Step 3:

[1884] Start gathering information

[1885] Server: Based on the received search query, the server launches a web crawler to collect relevant information from websites, blogs, social media, etc. on the Internet. It may also use specific APIs (such as news APIs) to obtain data.

[1886] Example: The server crawls news sites and social media sites using the keyword "smartphone technology 2023" to collect related information.

[1887] Step 4:

[1888] Data normalization

[1889] Server: The collected raw data is normalized. Specifically, unnecessary elements such as unnecessary HTML tags, advertisements, and scripts are removed from the information, and only the text information is extracted.

[1890] Example: Removing advertisements and unnecessary HTML tags from aggregated news articles and blog posts.

[1891] Step 5:

[1892] Text Cleaning

[1893] Server: Further cleans the normalized text, specifically removing line breaks and special characters, and arranging the text data in a consistent format, which improves the accuracy of text analysis.

[1894] For example: cleaning text by removing unnecessary spaces and special characters.

[1895] Step 6:

[1896] Natural Language Processing

[1897] Server: Analyzes collected text information using generative AI, extracts important keywords and topics, and gains a deep understanding of the text content through grammatical and semantic analysis.

[1898] Example: Extracting keywords related to smartphone technology (5G, camera resolution, battery life, etc.).

[1899] Step 7:

[1900] Reliability evaluation

[1901] Server: Using fact-checking APIs and proprietary algorithms, the server evaluates the reliability of collected information, calculates a reliability score, and filters out information with a low score.

[1902] Example: Calculate a credibility score for each source and filter out information with a low score.

[1903] Step 8:

[1904] Ranking algorithm applied

[1905] Server: Applies a ranking algorithm to rank the information based on criteria such as reliability, recency, and relevance. Sorting is performed based on the score of each piece of information.

[1906] Example: Combining reliability, recency, and relevance scores and ranking the information with the highest score at the top.

[1907] Step 9:

[1908] Saving ranking results

[1909] Server: Stores the generated ranking results in a database for later display and analysis.

[1910] Example: Storing ranked smartphone technology information in a database.

[1911] Step 10:

[1912] Sending ranking results

[1913] Server: Sends the ranking results stored on the user's device.

[1914] Example: Sending information about ranked smartphone technologies to the user's device.

[1915] Step 11:

[1916] User Interface Display

[1917] Terminal: The ranking results are displayed on the user interface, allowing users to access reliable information through the displayed rankings.

[1918] Example: A smartphone screen displays a list of reliable smartphone technology information.

[1919] Step 12:

[1920] Accepting user reviews

[1921] Device: The user rates the displayed information. Specifically, the rating is completed by pressing a rating button such as "useful" or "trustworthy."

[1922] Example: A user rates this review as helpful.

[1923] Step 13:

[1924] Submitting evaluation data

[1925] Terminal: Sends user evaluation data to the server, which reflects the user's evaluation in the system.

[1926] Example: User rating data is sent to the server.

[1927] Step 14:

[1928] Revenue Share Calculation

[1929] Server: Based on user rating data, distributes appropriate revenue to rankers. Advertising revenue and sponsorship income are calculated according to the ratings, and the corresponding points and revenue are awarded to rankers.

[1930] Example: Revenues are distributed to rankers based on user ratings.

[1931] Step 15:

[1932] Collecting feedback

[1933] Server: Collects and analyzes user evaluation data and behavioral data, which helps identify areas for system improvement.

[1934] Example: Collecting evaluation data to help improve the system.

[1935] Step 16:

[1936] Algorithm Updates

[1937] Server: Optimizes the ranking algorithm based on the collected feedback data. Uses adaptive learning to improve the accuracy of ranking results from next time onwards.

[1938] Example: Learn the characteristics of highly rated articles and reflect them in future rankings.

[1939] Step 17:

[1940] Emotion analysis

[1941] On-device: Utilizing an emotion engine to recognize the user's emotional state in real time, including interest, excitement, anger, sadness, etc.

[1942] Example: An emotion engine analyzes user emotions in real time while the user is browsing information.

[1943] Step 18:

[1944] Sending emotional data

[1945] Terminal: Sends the user's emotional data to the server.

[1946] Example: The analyzed emotion data is sent to the server.

[1947] Step 19:

[1948] Sentiment data analysis and algorithm optimization

[1949] Server: The ranking algorithm is optimized based on the collected emotional data, such as giving priority to information where the user expressed positive emotions.

[1950] For example, if the emotion engine analyzes the user's emotion as "interesting," it will rank similarly interesting information higher.

[1951] Step 20:

[1952] Dynamic Content Adjustment

[1953] On the device: Dynamically adjust the content displayed based on the analysis results of the emotion engine, prioritizing the display of topics and related information that interest the user.

[1954] Example: If a user expresses an interest in smartphone technology, automatically show them new related articles and videos.

[1955] This system allows users to efficiently gather the information they need and access reliable information. By taking user sentiment into consideration, it also enables more personalized information to be provided, and rankers can receive appropriate rewards.

[1956] Example 2

[1957] 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."

[1958] Conventional information gathering and organization systems have a process for ranking collected information based on its reliability, recency, and relevance, but they lack optimization that takes into account user emotions and evaluation data, limiting their ability to improve user experience. In particular, they are unable to reflect users' emotional reactions to content evaluations in real time, making it difficult to provide personalized information. Furthermore, there are also insufficient means for automatically evaluating and scoring the reliability of collected information, creating a risk of misinformation being included.

[1959] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1960] In this invention, the server includes means for collecting and normalizing information from a digital network using generative AI, means for analyzing the collected information using natural language processing and evaluating its reliability, means for ranking the analyzed and evaluated information based on criteria, means for displaying the information in a ranking format on the user's terminal, means for collecting user evaluation data and distributing revenue to rankers, means for analyzing user emotions using an emotion engine, means for optimizing the ranking algorithm based on the analyzed emotion data, and means for dynamically adjusting the content to be displayed based on the emotion data. This enables the provision of personalized information based on the user's emotion and evaluation data, resulting in more reliable information and an improved user experience.

[1961] "Generative AI" is a type of artificial intelligence technology used for natural language processing and data analysis, and has the ability to generate, convert, and analyze information from a wide range of data.

[1962] "Digital network" refers to the Internet and other digital communications networks that enable the collection, sharing, and transfer of information.

[1963] "Normalization" refers to the process of removing unnecessary data (e.g., HTML tags and advertisements) from collected information and converting it into pure text data.

[1964] "Natural language processing" refers to the technology of analyzing collected text data, extracting keywords and topics, and understanding the meaning and relationships of the information.

[1965] "Credibility assessment" is the process of assessing whether collected information is accurate, and includes assigning a credibility score using fact-checking APIs, etc.

[1966] "Ranking" refers to the process of analyzing and evaluating information and ranking it based on certain criteria (reliability, recency, relevance).

[1967] "User rating data" refers to data including ratings (e.g., star ratings, comments, etc.) made by users on information.

[1968] "Revenue sharing" refers to the process of distributing appropriate rewards to information providers based on users' rating data.

[1969] An "emotion engine" refers to technology for analyzing emotions from a user's facial expressions, voice, etc., and is used to collect and analyze emotional data in real time.

[1970] "Dynamic adjustment" refers to the process of changing and adjusting displayed content in real time based on the results of user sentiment analysis.

[1971] A "fact-checking API" is an application program interface used to assess the reliability of collected information and is a tool for fact-checking information.

[1972] This invention provides a system that combines generative AI and an emotion engine to efficiently collect, analyze, assess reliability, rank, and distribute revenues to users. Additionally, analyzing user emotions can optimize the ranking algorithm and dynamically adjust the content displayed. Below, we will explain each element of the system and how they are combined.

[1973] First, the device accepts the user's search request. The device has a search box, and the user enters the keyword of the information they want to find (e.g., "Smartphone technology in 2023") into this search box. The device then sends the search request to the server using the HTTP protocol.

[1974] Based on the search request received, the server collects relevant information from websites, blogs, social media, and other sources across the digital network, using crawling tools such as Apache Nutch and services such as the Twitter API.

[1975] The server then normalizes the collected information by using an HTML parser such as BeautifulSoup to remove unnecessary HTML tags and advertisements from the collected web pages and convert them into pure text data.

[1976] A generative AI model (e.g., GPT-4) runs on a server and analyzes the collected and normalized text data. During this analysis, natural language processing techniques are used to extract key keywords and topics, and fact-checking APIs (e.g., Snopes API) are used to assess the reliability of the information.

[1977] The server then generates a ranking based on the analyzed and evaluated information according to certain criteria (authority, recency, relevance), using a modified version of the PageRank algorithm or a recommendation system for this ranking process.

[1978] The ranking results are displayed on the user's device. The data is presented in a visually easy-to-read format using HTML and CSS. For example, the "latest articles about AI camera technology" could be displayed at the top.

[1979] Users rate the displayed information (e.g., star ratings or comments). The rating data is sent to the server, which then distributes appropriate revenue to the rankers based on the rating data. This rating data is also used to optimize the algorithm.

[1980] The user's device is equipped with an emotion engine, which uses technologies such as Emotion API to collect emotional data in real time from the user's facial expressions and voice. For example, it analyzes whether the user is expressing the emotion of "joy."

[1981] The server optimizes the ranking algorithm based on the collected sentiment data, for example, adjusting the algorithm to prioritize ranking information that users find "interesting," thereby improving the user experience.

[1982] Furthermore, the content displayed can be dynamically adjusted based on emotional data. For example, if it is determined that a user is interested in "AI camera technology," new related articles and videos will be dynamically displayed.

[1983] This system allows users to efficiently gather the information they need and access reliable information. It also enables personalized information provision that takes into account the user's emotions, and information providers can receive appropriate rewards.

[1984] Specific prompt examples:

[1985] "Please explain in detail the process flow for a user wanting to gather information about smartphone technology in 2023."

[1986] By combining generative AI models and emotion analysis technology, this system can respond quickly and appropriately to diverse user needs, significantly improving the quality and efficiency of information provision.

[1987] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1988] Step 1:

[1989] User: Enters "Smartphone technology in 2023" into the search box on the device and sends a search request, which generates search keywords as input data and sends them to the server using the HTTP protocol.

[1990] Input: Search keyword "Smartphone technology in 2023"

[1991] Output: Search request

[1992] Step 2:

[1993] Server: Based on the received search request, the server uses crawling tools such as Apache Nutch and Twitter APIs to collect relevant information from websites, blogs, and social media on the digital network. The collected information is stored as text data.

[1994] Input: Search request

[1995] Output: Collected information (text data)

[1996] Step 3:

[1997] Server: Normalize the collected information. Use an HTML parser such as BeautifulSoup to remove unnecessary HTML tags and advertisements and convert it into pure text data. This data processing results in clean text data.

[1998] Input: Collected information (text data)

[1999] Output: Normalized text data

[2000] Step 4:

[2001] Server: Uses a generative AI model (e.g., GPT-4) to extract key keywords and topics from the normalized text data, and uses fact-checking APIs (e.g., Snopes API) to assess the reliability of the information, resulting in analytics data and a reliability score.

[2002] Input: normalized text data

[2003] Output: Analysis data (major keywords, topics), reliability score

[2004] Step 5:

[2005] Server: Based on the analyzed information and the reliability score, it generates a ranking according to certain criteria (reliability, recency, relevance). This ranking is done using a modified version of the PageRank algorithm or a recommendation system.

[2006] Input: Analysis data (main keywords, topics), reliability score

[2007] Output: Ranking results

[2008] Step 6:

[2009] Device: The ranking results are visually formatted using HTML and CSS and displayed to the user. Users can view the ranking information on their devices.

[2010] Input: Ranking results

[2011] Output: Visualized ranking results (HTML, CSS)

[2012] Step 7:

[2013] User: Rates the displayed information (e.g., star ratings or comments). The rating data is sent to the server using the HTTP protocol.

[2014] Input: Visualized ranking results

[2015] Output: Rating data (star ratings, comments)

[2016] Step 8:

[2017] Server: Based on the collected evaluation data, the server distributes appropriate revenue to rankers. This evaluation data is also used to optimize the ranking algorithm.

[2018] Input: Evaluation data

[2019] Output: Revenue share data, optimized ranking algorithm

[2020] Step 9:

[2021] Device: Runs the emotion engine and collects emotional data from the user's facial expressions and voice. For example, it uses the Emotion API to analyze emotions such as "happiness," "interest," and "dissatisfaction" in real time.

[2022] Input: User's facial expressions and voice

[2023] Output: Emotion data

[2024] Step 10:

[2025] Server: The ranking algorithm is further optimized based on the data obtained from the sentiment analysis. For example, the algorithm is adjusted to prioritize ranking information that users find "interesting."

[2026] Input: Emotion data

[2027] Output: Optimized ranking algorithm

[2028] Step 11:

[2029] On-device: Dynamically adjusts content displayed based on sentiment analysis. For example, if a user is determined to be interested in "AI camera technology," relevant new articles and videos will be displayed.

[2030] Input: Sentiment analysis results

[2031] Output: Dynamically adjusted content placement

[2032] (Application example 2)

[2033] 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."

[2034] Conventional information gathering and ranking systems have had difficulty providing dynamic content that takes user sentiment into account, resulting in a lack of optimization for the user experience. Furthermore, while there is a demand for reliable information, there are limited methods for evaluating the accuracy and reliability of collected information in real time. Furthermore, while there is a demand for quickly optimizing ranking algorithms using user evaluation data and providing personalized information, there has been no system that combines sentiment analysis.

[2035] 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.

[2036] In this invention, the server includes means for collecting and normalizing information from a digital network using generative AI, means for analyzing the collected information using natural language processing and evaluating its reliability, means for ranking the analyzed and evaluated information based on criteria, means for displaying the information in a ranking format on the user's terminal, means for collecting user evaluation data and distributing revenue to rankers, means for analyzing user emotions in real time using a sentiment analysis engine, means for optimizing the ranking algorithm based on the collected sentiment data, and means for dynamically adjusting the content to be displayed. This makes it possible to provide efficient information taking user emotions into consideration, identify reliable information in a ranking format, and dynamically provide personalized information.

[2037] "Generative AI" is an artificial intelligence technology that collects information from digital networks and analyzes and generates data using natural language processing.

[2038] "Normalization" is the process of removing unnecessary elements from collected data and converting it into pure text data that is easy to analyze.

[2039] "Natural language processing" is a technology that allows computers to understand and analyze human language.

[2040] "Assessing credibility" is a method for determining the legitimacy or reliability of collected information.

[2041] "Ranking" refers to the process of ranking collected and analyzed information according to specific criteria.

[2042] An "emotion analysis engine" is a technology that analyzes user emotions in real time and collects them as data.

[2043] "Dynamic adjustment" is the process of instantly changing and updating displayed content based on data collected in real time.

[2044] "Distributing revenue" means distributing appropriate rewards to content providers and others based on user evaluation data.

[2045] This invention is a system that combines generative AI and a sentiment analysis engine to efficiently collect, analyze, assess credibility, rank, evaluate users' ratings, analyze sentiment, and dynamically adjust content. The following hardware and software are required for implementation:

[2046] Hardware and software used

[2047] Hardware: Camera (to capture the user's facial expressions), microphone (to capture the user's voice), user device (such as a smartphone or tablet).

[2048] Software: Generative AI model (for information analysis), TextBlob (for natural language processing), TensorFlow / Keras (for sentiment analysis model), REST API (for information collection).

[2049] What the program does

[2050] The server uses generative AI to collect information from digital networks, obtains data through crawling and API access, and normalizes this information, removing unnecessary HTML tags and advertisements, converting it into pure text data.

[2051] Furthermore, the collected text data is analyzed using natural language processing software (such as TextBlob), important keywords and topics are extracted, and the reliability of the information is evaluated through a fact-checking API. The evaluated information is ranked according to specific criteria and displayed in a ranked format on the user's device.

[2052] User sentiment is analyzed in real time using a sentiment analysis engine (such as TensorFlow / Keras models), and the ranking algorithm is optimized based on the collected sentiment data. The content displayed is dynamically adjusted according to the user's interests and sentiment.

[2053] Specific examples

[2054] For example, consider a case where a user wants to gather information about the latest VR technology. The user enters keywords into the search box, and the system collects related information from the internet. The collected information is then analyzed using natural language processing to extract important keywords and topics. The reliability of the information is evaluated based on a fact-checking API, and the results are displayed to the user in a ranked format.

[2055] While users are browsing information, the sentiment analysis engine analyzes their facial expressions and voice in real time. Information that is determined to be of interest to the user is displayed preferentially, and related information is dynamically adjusted, allowing users to quickly and efficiently obtain the content that interests them most.

[2056] Prompt Sentence Examples

[2057] "Analyze the following text and extract key keywords and sentiment:"

[2058] Text to analyze: "I'm looking for information on the latest in VR technology."

[2059] Using this prompt, the generative AI model analyzes the text data and provides information according to the user's request, which is expected to significantly improve the user experience.

[2060] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2061] Step 1:

[2062] The user enters a search keyword from the terminal and sends a request.

[2063] Input: User's search keyword

[2064] Output: Search request

[2065] How it works: A user enters a keyword into the search box on their device and presses the submit button, which sends the keyword to the system.

[2066] Step 2:

[2067] The server uses generated AI to collect information related to the search keywords from the Internet.

[2068] Input: Search request

[2069] Output: Collected digital information (web pages, blogs, social media posts, etc.)

[2070] How it works: The server collects relevant information from various sources on the Internet (websites and APIs) and stores it in text format.

[2071] Step 3:

[2072] The server normalizes the collected information and removes unnecessary HTML tags and advertisements.

[2073] Input: Collected digital information

[2074] Output: Clean text data

[2075] How it works: The normalization process removes unnecessary elements such as HTML tags and advertisements, producing pure text data.

[2076] Step 4:

[2077] The server uses natural language processing to analyze the clean text data and extract important keywords and topics.

[2078] Input: clean text data

[2079] Output: Analysis results (important keywords, extracted topics)

[2080] What it does: It uses a natural language processing library such as TextBlob to parse the text and extract key keywords and topics.

[2081] Step 5:

[2082] The server uses a fact-checking API to assess the reliability of the collected information.

[2083] Input: clean text data

[2084] Output: Trustworthiness score

[2085] How it works: Scores the legitimacy and veracity of collected information through a fact-checking API.

[2086] Step 6:

[2087] The server analyzes and evaluates the information and then ranks it according to specific criteria.

[2088] Input: Analysis results, reliability evaluation score

[2089] Output: Ranked list of information

[2090] How it works: Applies an algorithm that ranks information based on criteria such as reliability, freshness, and relevance.

[2091] Step 7:

[2092] The terminal provides the user with information displayed in a ranking format.

[2093] Input: Ranked list of information

[2094] Output: Ranking information displayed in the user interface

[2095] How it works: The ranking results are displayed on the user's screen, allowing the user to visually confirm reliable information.

[2096] Step 8:

[2097] The device uses an emotion analysis engine to analyze the user's facial expressions and voice in real time.

[2098] Input: User's facial expression data, voice speech data

[2099] Output: Sentiment analysis results (e.g., interest, joy, dissatisfaction)

[2100] How it works: The camera and microphone are used to capture the user's facial expressions and voice, which are then analyzed in real time by an emotion analysis engine.

[2101] Step 9:

[2102] The server optimizes the ranking algorithm based on the collected sentiment data.

[2103] Input: Sentiment analysis results, ranking information list

[2104] Output: Optimized ranking information

[2105] How it works: Reflects user sentiment data and adjusts and optimizes the ranking algorithm.

[2106] Step 10:

[2107] The device dynamically adjusts the content displayed based on the user's interests and emotions.

[2108] Input: Optimized ranking information, sentiment analysis results

[2109] Output: Dynamically adjusted content

[2110] How it works: Changes the information shown based on your interests and automatically surfaces new relevant content and videos.

[2111] 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.

[2112] 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.

[2113] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[2114] 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.

[2115] FIG. 9 illustrates 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 behaviors 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.

[2116] 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.

[2117] 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).

[2118] 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.

[2119] 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."

[2120] 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.

[2121] 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).

[2122] 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.

[2123] 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.

[2124] 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.

[2125] 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.

[2126] 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.

[2127] 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.

[2128] 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.

[2129] 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.

[2130] 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.

[2131] 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.

[2132] The following is further disclosed regarding the above embodiment.

[2133] (Claim 1)

[2134] A means of collecting and normalizing information from digital networks using generative AI;

[2135] A means for analyzing collected information using natural language processing and evaluating its reliability;

[2136] a means for ranking the analyzed and evaluated information based on criteria;

[2137] A way to display the rankings on the user's device,

[2138] A means of collecting user evaluation data and distributing revenue to rankers,

[2139] A system including:

[2140] (Claim 2)

[2141] 10. The system of claim 1, further comprising means for optimizing the ranking algorithm based on user rating data.

[2142] (Claim 3)

[2143] 10. The system of claim 1, further comprising means for scoring the reliability of information collected using a fact-checking API.

[2144] "Example 1"

[2145] (Claim 1)

[2146] A means for users to enter keywords into the device's search box and submit a request;

[2147] a means for collecting information from a digital network based on a search request;

[2148] A means for normalizing the collected information and converting it into pure text data by removing unnecessary data;

[2149] A means of analyzing collected text data using a generative AI model and extracting key keywords and concepts;

[2150] A means to assess the reliability of information using fact-checking APIs; and

[2151] a means for ranking the information based on the analyzed and evaluated information;

[2152] A means for displaying the ranking results on the user's device;

[2153] A means of collecting user evaluation data and distributing revenue to rankers,

[2154] A system including:

[2155] (Claim 2)

[2156] 10. The system of claim 1, further comprising means for optimizing the ranking algorithm based on user rating data.

[2157] (Claim 3)

[2158] 10. The system of claim 1, further comprising means for scoring the reliability of information collected using a fact-checking API.

[2159] "Application Example 1"

[2160] (Claim 1)

[2161] A means of collecting and normalizing information from digital networks using generative AI;

[2162] A means for analyzing collected information using natural language processing and evaluating its reliability;

[2163] a means for ranking the analyzed and evaluated information based on criteria;

[2164] A way to display the rankings on the user's device,

[2165] A means of collecting user evaluation data and distributing revenue to rankers,

[2166] A means to collect product reviews and display reliable reviews at the top,

[2167] A system including:

[2168] (Claim 2)

[2169] 10. The system of claim 1, further comprising means for optimizing the ranking algorithm based on user rating data.

[2170] (Claim 3)

[2171] 10. The system of claim 1, further comprising means for scoring the reliability of information collected using a fact-checking API.

[2172] "Example 2: Combining Emotion Engines"

[2173] (Claim 1)

[2174] A means of collecting and normalizing information from digital networks using generative AI;

[2175] A means for analyzing collected information using natural language processing and evaluating its reliability;

[2176] a means for ranking the analyzed and evaluated information based on criteria;

[2177] A means for displaying the rankings on a user's device;

[2178] A means for collecting user evaluation data and distributing revenue to rankers;

[2179] means for analyzing a user's emotions using an emotion engine;

[2180] a means for optimizing a ranking algorithm based on the analyzed sentiment data;

[2181] means for dynamically adjusting displayed content based on the emotion data;

[2182] A system including:

[2183] (Claim 2)

[2184] A means for optimizing a ranking algorithm based on user evaluation data;

[2185] 10. The system of claim 1, further comprising means for optimizing a ranking algorithm based on sentiment analysis data.

[2186] (Claim 3)

[2187] 10. The system of claim 1, further comprising means for scoring the reliability of information collected using a fact-checking API.

[2188] "Application example 2 when combining emotion engines"

[2189] (Claim 1)

[2190] A means of collecting and normalizing information from digital networks using generative AI;

[2191] A means for analyzing collected information using natural language processing and evaluating its reliability;

[2192] a means for ranking the analyzed and evaluated information based on criteria;

[2193] A way to display the rankings on the user's device,

[2194] A means of collecting user evaluation data and distributing revenue to rankers,

[2195] A means for analyzing user emotions in real time using an emotion analysis engine;

[2196] a means for optimizing a ranking algorithm based on the collected sentiment data;

[2197] a means for dynamically adjusting the displayed content;

[2198] A system including:

[2199] (Claim 2)

[2200] 10. The system of claim 1, further comprising means for optimizing the ranking algorithm based on user rating data.

[2201] (Claim 3)

[2202] 10. The system of claim 1, further comprising means for scoring the reliability of information collected using a fact-checking API. [Explanation of symbols]

[2203] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of collecting and normalizing information from digital networks using generative AI; A means for analyzing collected information using natural language processing and evaluating its reliability; a means for ranking the analyzed and evaluated information based on criteria; A way to display the rankings on the user's device, A means of collecting user evaluation data and distributing revenue to rankers, A system including:

2. 10. The system of claim 1, further comprising means for optimizing the ranking algorithm based on user rating data.

3. The system of claim 1 , further comprising means for scoring the reliability of information collected using a fact-checking API.

Images