System

Abstract

A system is provided.SOLUTION: A system comprising: means for inputting a retrieval keyword; means for retrieving a related web page based on the inputted retrieval keyword; means for activating a generation AI for analyzing each web page obtained as a retrieval result; means for evaluating the use of technical terms and the reliability of information sources using the generation AI; means for calculating a reliability score of each web page based on the evaluation results; means for integrating the reliability scores into the retrieval result; and means for displaying the integrated retrieval result.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] When gathering information online, it is difficult for non-expert users, especially, to determine the reliability of information displayed in search results. Currently, users can only judge reliability based on superficial information such as the website operator's title or visual design, leaving them powerless to protect against misrepresented information. It is particularly difficult to determine the reliability of websites containing specialized content (e.g., medical or financial information), which increases the risk of making decisions based on inaccurate information. There is a need for a method that solves these problems and allows users to easily obtain reliable information. [Means for solving the problem]

[0005] The present invention provides a web page credibility evaluation system using a generative AI. Specifically, the system includes the following means:

[0006] 1. How to enter search keywords

[0007] 2. A way to search for related web pages based on the search keywords entered

[0008] 3. A means to trigger a generative AI to analyze each web page retrieved as a search result.

[0009] 4. Using generative AI to assess terminology and source credibility

[0010] 5. A method for calculating the credibility score of each web page based on the evaluation results

[0011] 6. A way to integrate trustworthiness scores into search results

[0012] 7. A way to display consolidated search results

[0013] This allows users to easily check the reliability of web pages displayed in search results and protect themselves from fraudulent or incorrect information. The system also enables more accurate reliability judgments by using credibility assessments based on official institutions and academic papers.

[0014] A "search keyword" is a word or phrase that a user enters into a search engine that triggers a search for related information.

[0015] "Related web pages" are a set of websites retrieved by a search engine based on the entered search keywords, which may contain the information the user is looking for.

[0016] "Generative AI" is a type of artificial intelligence technology that uses specific algorithms to analyze data and generate results.

[0017] "Terminology" refers to words or phrases used in a particular field of expertise that have a common understanding within that field.

[0018] A "source" is the place or source from which particular information comes; it is a document that is quoted, linked to, or referenced within a web page.

[0019] A "trustworthiness score" is an indicator that indicates the reliability of information within a web page using numbers or letters, and is calculated based on the results of the evaluation by the generation AI.

[0020] "Structural analysis" is the process of analyzing the HTML content of a web page to understand its internal structure (e.g., tags and sections).

[0021] The "main content" refers to the information portion of a web page that is considered to be the most important, and is the information that users refer to most often.

[0022] "Metadata" is supplemental information related to a web page, including information such as the page's title, description, keywords, and creation date.

[0023] A "user" is a person who uses this system to search for information and is the ultimate recipient of the information.

[0024] "Search results" are the list of relevant web pages provided by a search engine after a user enters search keywords.

[0025] A "source" is the original source from which information within a web page was originally referenced. [Brief explanation of the drawings]

[0026] [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

[0027] 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.

[0028] First, the terms used in the following description will be explained.

[0029] 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).

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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."

[0034] [First embodiment]

[0035] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0036] 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.

[0037] 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).

[0038] 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.

[0039] 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.

[0040] 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.

[0041] 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.

[0042] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0043] 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.

[0044] 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.

[0045] 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.

[0046] 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."

[0047] The present invention provides a system for evaluating the reliability of web pages using generative AI, allowing users to easily select reliable information from search results.

[0048] System configuration

[0049] The system consists of the following main components:

[0050] 1. How to enter search keywords

[0051] 2. How to search for related web pages

[0052] 3. Web page analysis using generative AI

[0053] 4. Terminology and Source Credibility Assessment Tools

[0054] 5. How the reliability score is calculated

[0055] 6. Means of integrating trustworthiness scores into search results

[0056] 7. Display of integrated results

[0057] Program processing

[0058] Program Overview

[0059] The system operates through the following processes:

[0060] 1. Enter search keywords

[0061] 2. Obtaining search results

[0062] 3. Web page analysis

[0063] 4. Reliability evaluation

[0064] 5. Integration of evaluation results

[0065] 6. Display

[0066] The system begins with the user entering a search keyword. The device then sends the keyword to the server, which searches for relevant web pages. The resulting search results are analyzed by a generative AI to evaluate the trustworthiness of each web page. A trustworthiness score is then calculated and integrated into the search results. Finally, the search results with the associated trustworthiness score are displayed on the user's device.

[0067] Detailed program description

[0068] Enter search keywords

[0069] The user enters a keyword (e.g., "cold treatment") on the search screen and presses the search button, which starts the search process.

[0070] Getting search results

[0071] The device sends the input keywords to the server, which then searches for related web pages using existing search algorithms and temporarily stores the results.

[0072] Web page analysis

[0073] The server sends the URL of each web page retrieved as a search result to the generation AI module, which downloads the HTML content of each web page and analyzes it, including extracting the key content and metadata within the page.

[0074] Terminology and source credibility assessment

[0075] The generative AI evaluates the use of technical terms within a webpage and the reliability of the source. Specifically, it checks whether technical terms are used accurately and evaluates whether the cited sources are trustworthy. For example, if the source is a public institution or an academic paper, it will assign a high reliability score.

[0076] Calculating the reliability score

[0077] The server calculates a credibility score for each web page based on the results evaluated by the AI, which is expressed as a number or letter so that users can determine the trustworthiness of each search result at a glance.

[0078] Integrating trust scores into search results

[0079] The server integrates the authority score into the search results by adding the authority score below the title of each web page.

[0080] Viewing the integration results

[0081] The terminal displays the integrated search results received from the server to the user, who can then review the displayed search results and select which web page to view based on the reliability score.

[0082] Specific examples

[0083] For example, when searching for the keyword "cold treatment," the user first enters the keyword and presses the search button. The device sends the keyword to the server, which retrieves approximately 10 related web pages. The server then analyzes each web page using generative AI to evaluate the use of technical terms and the reliability of cited sources. As a result of the evaluation, a reliability score is calculated, and the server integrates the reliability score into the search results. Finally, the device displays the search results with the reliability score assigned to them to the user. The user can check the reliability score of each web page and select the most reliable page to view.

[0084] As described above, the system of the present invention helps users to easily and accurately determine the reliability of web pages, and can particularly increase the reliability of pages that contain specialized information.

[0085] The processing flow will be explained below.

[0086] Step 1: Enter search keywords

[0087] 1. A user enters a keyword (e.g., "cold treatment") into the search field of a search engine.

[0088] 2. The user presses the "Search" button.

[0089] Step 2: Getting search results

[0090] 1. The device sends the entered search keywords to the server.

[0091] 2. The server searches for relevant web pages based on the keywords.

[0092] 3. The server generates a list of URLs for the web pages retrieved as search results.

[0093] 4. The server temporarily stores the URL list.

[0094] Step 3: Analyze the web page

[0095] 1. The server retrieves the URL of each web page from the URL list in turn.

[0096] 2. The server downloads the HTML content of each web page.

[0097] 3. The server sends the downloaded HTML content to the generation AI module.

[0098] 4. Generative AI analyzes the page structure and extracts key content and metadata.

[0099] Step 4: Assess terminology and source credibility

[0100] 1. Analyze the key content extracted by the generative AI and evaluate the use of technical terms.

[0101] 2. Generative AI analyzes citations and links within web pages and evaluates the source of the citation or link.

[0102] 3. If the generating AI determines that the source is based on a public institution or academic paper, it will assign a higher credibility score to that source.

[0103] Step 5: Calculate the reliability score

[0104] 1. The server calculates the trustworthiness score for each web page based on the content evaluated by the AI.

[0105] 2. The server temporarily stores the credibility score for each web page.

[0106] Step 6: Integrating Trust Scores into Search Results

[0107] 1. The server integrates the confidence score into the search results.

[0108] 2. The server creates a data structure that adds an authority score to each web page below its title.

[0109] Step 7: Viewing the integration results

[0110] 1. The server sends the consolidated search results to the device.

[0111] 2. The terminal displays the search results received to the user.

[0112] 3. Users can see the credibility score under the title of each search result displayed.

[0113] Example 1

[0114] 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."

[0115] The amount of information on the Internet is enormous, making it difficult for users to quickly find reliable information. This problem is particularly pronounced in fields where reliability is essential, such as specialized or medical information. Furthermore, existing search engines lack the functionality to evaluate the reliability of search results, leaving users with insufficient means to select appropriate information.

[0116] 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.

[0117] In this invention, the server includes means for inputting search keywords, means for searching for related web pages based on the input search keywords, means for activating a generation AI to analyze each web page obtained as a search result, means for evaluating the use of technical terms and the reliability of information sources using the generation AI, means for calculating a reliability score for each web page based on the evaluation results, means for integrating the reliability scores into search results, means for displaying the integrated search results, means for transmitting the search keywords to the server and temporarily saving the obtained search results, means for the generation AI to download the HTML content of the web page and extract key content and metadata within the page, and means for the user's terminal to display the integrated search results received from the server, thereby enabling users to quickly identify reliable information and browse appropriate information sources.

[0118] A "search keyword" is a character string that a user inputs to search for information.

[0119] The "input means" is a device or interface for a user to input search keywords.

[0120] A "search tool" is a function or process that searches the Internet for relevant web pages based on input search keywords.

[0121] "Generative AI" is a system that uses artificial intelligence technology to analyze data and perform evaluation and generation.

[0122] The "analysis means" is a function that analyzes the acquired web page in detail and extracts information.

[0123] "Terminology" is a term that has a specialized meaning in a particular field.

[0124] A "source" is the source or citation of information contained within a web page.

[0125] The "trustworthiness assessment tool" is a function that uses generative AI to evaluate the use of terminology within a web page and the reliability of the source of information.

[0126] The "trustworthiness score" is an evaluation result that indicates the trustworthiness of each web page numerically or alphabetically.

[0127] The "calculation means" is a function that calculates a reliability score based on the evaluation results of the generating AI.

[0128] The "search result integration means" is a function that incorporates reliability scores into search results.

[0129] The "display means" is a function that displays the integrated search results on the user's terminal.

[0130] A "server" is a computer system for processing, storing, and distributing data.

[0131] A "terminal" is an electronic device that a user operates, and typically refers to a PC or smartphone.

[0132] "HTML content" is data written in a markup language that makes up a web page.

[0133] "Metadata" is data that provides additional information about the content of a web page.

[0134] DETAILED DESCRIPTION OF THE INVENTION The present invention is a system that utilizes a generative AI model to evaluate the trustworthiness of web pages, allowing users to easily select reliable information from search results.

[0135] System configuration

[0136] The system consists of the following main components:

[0137] 1. How to enter search keywords

[0138] 2. A way to find related web pages

[0139] 3. A means to trigger a generative AI to analyze each web page retrieved as a search result.

[0140] 4. Using generative AI to assess terminology and source credibility

[0141] 5. A method for calculating the credibility score of each web page based on the evaluation results

[0142] 6. A way to integrate trustworthiness scores into search results

[0143] 7. A way to display consolidated search results

[0144] 8. A means of sending search keywords to a server and temporarily storing the retrieved search results

[0145] 9. How generative AI downloads the HTML content of a webpage and extracts key content and metadata from the page.

[0146] 10. Means for displaying the integrated search results received by the user's device from the server

[0147] Hardware and software used

[0148] Client terminal: PC or smartphone used by the user

[0149] Server: A server (e.g., a cloud computing service) to host the search engine and the generative AI.

[0150] Generative AI models: Natural language processing models such as GPT-4

[0151] Program processing overview

[0152] When a user enters a specific keyword (e.g., "cold treatment") on the search screen and presses the search button, the device sends the keyword to the server. The server uses an existing search engine to search for related web pages and temporarily stores the retrieved results. The server then passes the temporarily stored URL list of search results to the generation AI module, which downloads and analyzes the HTML content of each web page. Based on the analyzed text content, the generation AI evaluates the use of technical terms and the reliability of cited sources. The server then calculates a reliability score for each web page based on the generation AI's evaluation results and integrates the reliability scores into the search results. Finally, the device displays the integrated search results received from the server to the user, who then selects which web page to view based on the reliability score.

[0153] Specific examples

[0154] For example, let's take a specific example of searching for the keyword "cold treatment." When a user enters the keyword and presses the search button, the device sends the keyword to the server, which retrieves approximately 10 related web pages. The server then uses generative AI to analyze each web page and evaluates the use of technical terms and the reliability of cited sources. As a result of the evaluation, a reliability score is calculated, and the server integrates this score into the search results. Finally, the device displays the search results with the assigned reliability score to the user. The user can check the reliability score of each web page and select the most reliable page to view.

[0155] Example prompts for generative AI models

[0156] "Evaluate the titles and content of web pages and calculate a credibility score based on the following criteria: accurate use of technical terminology, credibility of cited sources (e.g., official institutions, academic papers, news outlets, etc.). For example, if you are evaluating web pages resulting from a search for the keyword 'cold treatment methods,' rate each page using the criteria above."

[0157] The above describes the embodiments of the present invention, which enable a user to quickly identify reliable information and browse appropriate information sources.

[0158] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0159] Step 1:

[0160] The user enters a specific keyword (e.g., "cold treatment") on the search screen and presses the search button. The input here is the action of the user entering text into the search box, and this text becomes the input data for the next step. The device receives the user's input and prepares to send that information to the server. In concrete terms, the user opens a search form in a browser, enters "cold treatment" into the search box, and clicks the search button.

[0161] Step 2:

[0162] The device sends the entered keywords to the server, which then uses the entered keywords for the next search process. The server uses a wide range of search engines to search for related web pages based on the received keywords and temporarily stores the results. Specifically, the device sends the user's input information to the server using a communication protocol (e.g., HTTP), and the server uses Google's search API or similar to obtain a list of URLs for related pages. The obtained list of URLs is temporarily stored in memory.

[0163] Step 3:

[0164] The server passes the temporarily saved list of search result URLs to the generation AI module. This becomes the input data. The generation AI downloads the HTML content of each web page and analyzes it as text. Specifically, the server passes the URL list to the generation AI, which then uses an HTTP request to retrieve the HTML of each web page, analyzes it, and extracts the text content and metadata.

[0165] Step 4:

[0166] Based on the text content analyzed by the generative AI, the usage of technical terms and the reliability of cited sources are evaluated. This serves as input data for the next evaluation step. Specifically, technical terms are extracted from the analyzed text and checked to see if they are used accurately. The reliability of cited sources, such as public institutions and academic papers, is also evaluated. Specifically, the generative AI extracts technical terms from the text and checks the cited sources.

[0167] Step 5:

[0168] The server calculates a reliability score for each web page based on the evaluation results of the generation AI. This becomes the input data for the evaluation score used in the next step. The reliability score serves as an indicator when users check the results. Specifically, the server receives the evaluation data from the generation AI, inputs it into the scoring algorithm, and calculates the reliability score. The calculated score is generated in a form corresponding to the URL of each web page.

[0169] Step 6:

[0170] The server integrates the calculated reliability score into the search results, generating integrated data. This integration result becomes the input data for the next display step. The reliability score is integrated by adding it below the title of each web page. Specifically, the server compares the list of search result URLs with the calculated reliability score, adds the reliability score to the metadata of each URL, and generates the final list of search results.

[0171] Step 7:

[0172] The terminal displays the integrated search results received from the server to the user. This is the final output data. The user can check the displayed search results and select which web page to view based on the reliability score. Specifically, the terminal receives the response from the server, analyzes the response data, and converts it into a format to be displayed in the browser. The user then clicks on each web page to view it based on the reliability score.

[0173] (Application example 1)

[0174] 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."

[0175] In today's world, it is extremely important for users to verify the reliability of information they search for on the Internet. Particularly on online shopping sites, the reliability of product descriptions and reviews significantly influences users' purchasing decisions. However, current search result displays do not provide a concrete means for evaluating reliability, leaving users to make their own judgments about the quality of information. To address this issue, the present invention aims to provide a system that automatically assigns reliability scores to search results on online shopping sites, allowing users to easily select highly reliable information.

[0176] 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.

[0177] In this invention, the server includes means for inputting search keywords, means for searching for related web pages based on the input search keywords, means for activating a generation AI for analyzing each web page obtained as a search result, means for evaluating the use of technical terms and the reliability of information sources using the generation AI, means for calculating a reliability score for each web page based on the evaluation results, means for integrating the reliability score into the search results, means for displaying the integrated search results, means for analyzing the content of each product information page and evaluating the quality of product descriptions and reviews, and means for assigning a reliability score to the evaluated product information page. This allows users to easily select reliable product information pages and make purchasing decisions with confidence.

[0178] "Search keywords" are words or phrases that users enter to describe the information they are looking for.

[0179] A "web page" is a document containing information that can be accessed on the Internet and is written in a format such as HTML.

[0180] "Generative AI" is a system that uses artificial intelligence technology to generate text and analyze information.

[0181] "Jargon" is a specialized word or phrase used in a particular field or area.

[0182] A "source" is a source or data source from which particular information is obtained.

[0183] A "credibility score" is a number or indicator used to evaluate the reliability of a particular piece of information or source of information.

[0184] "Content" refers to the information contained in a web page or document, such as text, images, or video.

[0185] A "review" is an evaluation or comment written by a user or purchaser describing their own experience or opinion.

[0186] "Search results" are the list of relevant web pages or information that appears when a user enters keywords into a search engine.

[0187] A "product information page" is a web page that contains detailed descriptions, specifications, and reviews of a particular product.

[0188] This invention relates to a system that evaluates the reliability of each product information page when a user searches for a product on an online shopping site, allowing the user to make a purchasing decision based on reliable information. This system is realized by using generative AI to evaluate the use of technical terms and the reliability of information sources, from the input of search keywords to the display of search results, and calculates a reliability score.

[0189] System configuration

[0190] The system consists of the following main components:

[0191] 1. How to enter search keywords

[0192] 2. How to search for related web pages

[0193] 3. Web page analysis using generative AI

[0194] 4. Terminology and Source Credibility Assessment Tools

[0195] 5. How the reliability score is calculated

[0196] 6. A way to integrate trustworthiness scores into search results

[0197] 7. Display of integrated results

[0198] 8. A way to analyze the content of each product page and evaluate the quality of the product descriptions and reviews.

[0199] 9. A means of assigning a credibility score to rated product information pages

[0200] Program processing

[0201] 1. Enter search keywords

[0202] The user enters the product name into the smartphone app and presses the search button.

[0203] 2. Obtaining search results

[0204] The device sends the entered search keywords to the server, which retrieves related product information pages. The search results are temporarily saved.

[0205] 3. Web page analysis

[0206] The server sends the URL of each product information page retrieved as a search result to the generation AI module, which downloads the HTML content of each web page and analyzes it, including extracting the key content and metadata within the page.

[0207] 4. Terminology and source credibility assessment

[0208] The generative AI evaluates the use of technical terms on product information pages and the reliability of the sources. Specifically, it checks whether technical terms are used accurately and evaluates whether the cited sources are trustworthy. For example, if the source is a public institution or an academic paper, it will assign a high reliability score.

[0209] 5. Calculating the reliability score

[0210] The server calculates a reliability score for each product information page based on the results of the AI ​​generation. The reliability score is displayed as a number or letter, allowing users to determine the reliability of each product information page at a glance.

[0211] 6. Integrating Trust Scores into Search Results

[0212] The server integrates the reliability score into the search results by adding it below the title of each product information page.

[0213] 7. Displaying the integrated results

[0214] The terminal displays the integrated search results received from the server to the user, who can then review the displayed search results and select which information page to view based on the reliability score.

[0215] Hardware and software used

[0216] Hardware: Servers, smartphones

[0217] Software: Generative AI (GPT-4, etc.), Python, web frameworks (Django, Flask), databases (PostgreSQL)

[0218] Specific examples

[0219] For example, suppose a user enters "vitamin supplements" into a smartphone app and presses the search button. The device sends this search keyword to the server, which retrieves related product information pages. The server then uses generative AI to analyze each product information page and evaluates the use of technical terms and the reliability of the information source. A reliability score is calculated based on the evaluation results, and the server integrates the reliability score into the search results. The device then displays the search results with the reliability score to the user.

[0220] Prompt Sentence Examples

[0221] The prompt for the search "vitamin supplement" is:

[0222] Rate the trustworthiness of this product information page:

[0223] URL: https: / / example.com / vitamin-supplement

[0224] Key points: Accuracy of product description, quality of reviews, citation of reliable sources

[0225] The analysis results obtained from the generative AI model include numerical values ​​such as "content accuracy: 85%" and "source reliability: 90%." A reliability score is calculated based on this and displayed to the user. Users can then select products and make purchasing decisions with confidence based on the reliability score.

[0226] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0227] Step 1:

[0228] The user enters the product name into the smartphone app and presses the search button. The entered keywords are sent to the device as "search keywords." The entered data is the product name specified by the user, and this becomes the base data for subsequent processing.

[0229] Step 2:

[0230] The server receives the "search keywords" sent from the terminal. The server uses existing search engine functions to search for the relevant product information page. At this point, the only input is the "search keywords," and the output is a list of URLs for related product information pages.

[0231] Step 3:

[0232] The server sends a list of URLs for each product information page obtained as a search result to the generation AI module. The generation AI downloads the HTML content of each product and analyzes it. The analysis includes extracting the main content and metadata within the page. The input data is the list of URLs for the product information pages, and the output data is the analyzed HTML content.

[0233] Step 4:

[0234] The generative AI evaluates the use of technical terms on product information pages and the reliability of the sources. Specifically, it checks whether technical terms are used accurately and evaluates whether the cited sources are trustworthy. If highly reliable sources such as public institutions or academic papers are cited, the rating is higher. The input data is HTML content, and the output data is a reliability rating score.

[0235] Step 5:

[0236] The server calculates the reliability score of each product information page based on the results of the evaluation by the generation AI. The reliability score is calculated based on a specific algorithm and is displayed as a number or letter that allows users to determine the reliability of each product information page at a glance. The input data is the reliability evaluation score of each product information page, and the output data is the reliability score.

[0237] Step 6:

[0238] The server integrates the reliability scores into the search results. Specifically, it adds the reliability score below the title of each product information page. This process visually displays the reliability rating for all search results. The input data is the reliability score and a list of search results, and the output data is a list of search results with the reliability scores added.

[0239] Step 7:

[0240] The terminal displays the integrated search results with the reliability scores received from the server to the user. The user checks the displayed search results and selects which product information page to view based on the reliability scores. The input data are the search results with the reliability scores added, and the output data are the product information pages displayed to the user.

[0241] Through the above steps, the user can easily select highly reliable product information and make a purchase decision with confidence.

[0242] 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.

[0243] The present invention provides a system for providing more personalized search results by combining a conventional system that uses generative AI to evaluate the trustworthiness of web pages with an emotion engine that recognizes user emotions.

[0244] System configuration

[0245] The system consists of the following main components:

[0246] 1. How to enter search keywords

[0247] 2. How to search for related web pages

[0248] 3. Web page analysis using generative AI

[0249] 4. Terminology and Source Credibility Assessment Tools

[0250] 5. How the reliability score is calculated

[0251] 6. Means of integrating trustworthiness scores into search results

[0252] 7. Display of integrated results

[0253] 8. Emotion engine that recognizes user emotions

[0254] 9. Sentiment-Based Search Results Customization

[0255] Program processing

[0256] The system proceeds as follows:

[0257] Enter search keywords

[0258] A user enters a keyword (e.g., "cold treatment") into the search field of a search engine and presses the search button. This starts the search process, but at this point the emotion engine is also activated. The emotion engine analyzes the user's input method and speed, facial expressions, tone of voice, etc., to recognize the user's emotional state (e.g., impatience, anxiety, relaxation, etc.).

[0259] Getting search results

[0260] The device sends the emotion data recognized by the emotion engine along with the entered search keywords to the server, where the server searches for related web pages using existing search algorithms and temporarily stores the results.

[0261] Web page analysis

[0262] The server sends the URL of each web page retrieved as a search result to the generation AI module, which downloads the HTML content of each web page and analyzes it, including extracting the key content and metadata within the page.

[0263] Terminology and source credibility assessment

[0264] The generative AI evaluates the use of domain-specific terminology and determines the trustworthiness of cited sources. For example, it assigns a higher credibility score to sources from public institutions or academic papers. Furthermore, it can prioritize sources that inspire a sense of security based on the user's emotional state.

[0265] Calculating the reliability score

[0266] The server calculates a credibility score for each web page based on the evaluation results of the generation AI. In addition, it takes into account data from the emotion engine and adjusts the score according to the user's current emotional state. For example, if a user is feeling anxious, it assigns a higher score to information that is highly reliable and reassuring.

[0267] Integrating trust scores into search results

[0268] The server integrates the reliability score into the search results, and adjusts the display of search results based on the user's emotional state. For example, if a user is in a hurry, the server highlights the results so that the reliability can be seen at a glance.

[0269] Viewing the integration results

[0270] The server sends the integrated search results to the device, which then displays the results in a format that takes the user's emotions into consideration.The user can then review the displayed search results and select the most appropriate web page to view, with the reliability score and emotional engine having been adjusted accordingly.

[0271] Specific examples

[0272] For example, if the emotion engine recognizes that a user searching for the keyword "cold treatment" is feeling anxious or impatient when typing, the emotion engine sends this information to the server. The server retrieves search results based on that information, and the generation AI analyzes the webpage. It then evaluates the reliability of the terminology and information source, and calculates a reliability score taking into account the emotional data. Finally, when the search results incorporating the reliability score are sent to the user's device, reliable information that will reassure users who are feeling anxious is highlighted.

[0273] This allows users to easily access the most appropriate information based on their emotional state at the time, providing a more personalized search experience.

[0274] The processing flow will be explained below.

[0275] Step 1: Enter search keywords

[0276] 1. A user enters a keyword (e.g., "cold treatment") into the search field of a search engine.

[0277] 2. The user presses the "Search" button.

[0278] 3. The device monitors the user's input method (speed, rhythm, etc.), facial recognition camera, and tone of voice during voice input, and activates the emotion engine.

[0279] 4. The emotion engine analyzes the user's emotional state (e.g., impatience, anxiety, relaxation, etc.) and generates emotion data.

[0280] Step 2: Getting search results

[0281] 1. The device sends the entered search keywords and the emotion data recognized by the emotion engine to the server.

[0282] 2. The server searches for relevant web pages based on the keywords.

[0283] 3. The server generates a list of URLs of the web pages obtained as search results and temporarily stores them.

[0284] Step 3: Analyze the web page

[0285] 1. The server retrieves the URL of each web page from the URL list in turn.

[0286] 2. The server downloads the HTML content of each web page.

[0287] 3. The server sends the downloaded HTML content to the generation AI module.

[0288] 4. Generative AI analyzes the page structure and extracts key content and metadata.

[0289] Step 4: Assess terminology and source credibility

[0290] 1. Analyze the key content extracted by the generative AI and evaluate the use of technical terms.

[0291] 2. Generative AI analyzes citations and links within web pages and evaluates the source of the citation or link.

[0292] 3. If the generating AI determines that the source is based on a public institution or academic paper, it will assign a higher credibility score to that source.

[0293] 4. The generative AI uses data from the emotion engine to adjust the importance of information according to the user's emotional state. For example, if a user is feeling anxious, information that gives a sense of security will be given a high score.

[0294] Step 5: Calculate the reliability score

[0295] 1. The server calculates the trustworthiness score for each web page based on the content evaluated by the AI.

[0296] 2. The server also takes into account data from the emotion engine and adjusts the reliability score according to the user's emotional state.

[0297] 3. The server generates and temporarily stores a final credibility score for each web page.

[0298] Step 6: Integrating Trust Scores into Search Results

[0299] 1. The server integrates the confidence score into the search results.

[0300] 2. The server creates a data structure that adds an authority score to each web page below its title.

[0301] 3. The emotion engine determines how search results are displayed (e.g., highlighted, color-coded) depending on the user's emotional state.

[0302] Step 7: Viewing the integration results

[0303] 1. The server sends the consolidated search results to the device.

[0304] 2. The search results received by the device are displayed in a format that takes into account the user's emotional state.

[0305] 3. Users can see the credibility score under the title of each search result displayed and view customized information based on their emotional state.

[0306] Through the above steps, the system of the present invention provides personalized search results that adapt to the user's emotional state to enhance the user's search experience.

[0307] Example 2

[0308] 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."

[0309] Conventional search systems present relevant information based on keywords entered by the user, but because they do not take the user's emotional state into consideration, it is difficult to provide the optimal information the user is looking for. As a result, even if the user is feeling anxious or impatient, information that provides a sense of security is not prioritized, resulting in a poor user experience.

[0310] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for inputting a search keyword; means for searching for related web pages based on the input search keyword and user emotional data; means for activating an emotional engine and analyzing the user's emotional state; means for activating a generation AI for analyzing each web page obtained as a search result; means for evaluating the use of technical terms and the reliability of information sources using the generation AI; means for calculating a reliability score for each web page taking into account the user's emotional data; means for integrating the reliability score into the search result and adjusting the display method based on the user's emotional state; and means for displaying the integrated search result. This makes it possible to provide personalized search results that take into account the user's emotional state.

[0311] The "means for inputting search keywords" refers to a device or function that allows a user to input search keywords through the interface of a search engine.

[0312] "Means for searching for related web pages based on input search keywords and user emotion data" refers to a device or function for finding related web pages using search keywords input by the user and emotion data analyzed by the emotion engine.

[0313] "Means for activating an emotion engine and analyzing the user's emotional state" refers to a device or function for recognizing the user's emotions (e.g., impatience, anxiety, relaxation) by analyzing the user's input method, speed, facial expression, tone of voice, etc.

[0314] "Means for launching a generative AI to analyze each web page obtained as a search result" refers to a device or function that launches a generative AI model to analyze the content of each web page obtained as a search result.

[0315] "Means for using generative AI to evaluate terminology usage and source reliability" refers to a device or function that uses generative AI to analyze and evaluate the use of terminology within a web page and the reliability of cited sources.

[0316] The "means for calculating the reliability score of each web page taking into account the user's emotional data" refers to a device or function for calculating the reliability score of each web page taking into account the evaluation results of the generation AI and the user's emotional state.

[0317] "Means for integrating confidence scores into search results and adjusting the display based on the user's emotional state" refers to a device or function for incorporating calculated confidence scores into search results and changing the display of search results depending on the user's emotional state.

[0318] The "means for displaying integrated search results" is a device or function for displaying search results including reliability scores on a user's terminal so that the user can view them.

[0319] The following describes in detail the mode for carrying out the present invention: The system utilizes generative AI to evaluate the trustworthiness of web pages, and combines it with an emotion engine that recognizes user emotions to provide more personalized search results.

[0320] System configuration

[0321] The system consists of the following main elements:

[0322] 1. How to enter search keywords

[0323] 2. How to search for related web pages

[0324] 3. Web page analysis using generative AI

[0325] 4. Terminology and Source Credibility Assessment Tools

[0326] 5. How the reliability score is calculated

[0327] 6. Means of integrating trustworthiness scores into search results

[0328] 7. Display of integrated results

[0329] 8. Emotion engine that recognizes user emotions

[0330] 9. Sentiment-Based Search Results Customization

[0331] Enter search keywords

[0332] A user enters a keyword (e.g., "cold treatment") into the search field of a search engine and presses the search button. This action starts the search process and simultaneously activates the emotion engine. The emotion engine analyzes the user's input method, speed, facial expression, tone of voice, etc., to recognize the user's emotional state (e.g., impatience, anxiety, relaxation, etc.).

[0333] Getting search results

[0334] The device sends the entered search keywords and the emotion data recognized by the emotion engine to the server, which then uses existing search algorithms to search for related web pages and temporarily stores the results.

[0335] Web page analysis

[0336] The server sends the URL of each web page retrieved as a search result to the generative AI module. The generative AI downloads the HTML content of each web page, extracts key content and metadata, and analyzes it. The generative AI model can be implemented using Python libraries such as "BeautifulSoup" and "Selenium."

[0337] Terminology and source credibility assessment

[0338] Based on the analyzed content, the generative AI evaluates the use of technical terms and the reliability of cited sources. For example, if the source is a public institution or academic paper, it will give a high reliability score. It can also take into account the user's emotional state and prioritize sources that give a sense of security.

[0339] Calculating the reliability score

[0340] The server calculates a reliability score for each web page based on the evaluation results of the generated AI and data from the emotion engine. For example, if a user is feeling anxious, a high score can be assigned to information that is highly reliable and reassuring.

[0341] Integrating trust scores into search results

[0342] The server integrates the reliability score into search results, adjusting the presentation based on the user's emotional state—for example, highlighting results to make their reliability clear at a glance to users in a hurry.

[0343] Viewing the integration results

[0344] The server sends the integrated search results to the device, which then displays the search results in a format that takes the user's emotions into consideration.The user can review the displayed search results and select the most appropriate web page to view, with the results adjusted by the reliability score and emotion engine.

[0345] Specific examples

[0346] For example, if the emotion engine recognizes that a user searching for the keyword "cold treatment" is feeling anxious or impatient when typing, the emotion engine sends this emotional data to the server. The server retrieves search results based on that information, and the generation AI analyzes the webpage. The analysis results then evaluate the reliability of the terminology and information sources, and calculate a reliability score taking the emotional data into account. Finally, the search results, incorporating the reliability score, are sent to the user's device, and reliable information that will reassure anxious users is highlighted and displayed.

[0347] Example prompt

[0348] If the user enters the keyword "cold treatment" and the emotion engine recognizes that the user is feeling anxious, prioritize displaying reliable information that provides a sense of security.

[0349] This system allows users to easily obtain the most appropriate information based on their emotional state at the time, providing a more personalized search experience.

[0350] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0351] Step 1:

[0352] The user enters a search keyword. Specifically, the user enters "cold treatment" in the search field of the search engine and presses the search button. This input triggers the search process, obtaining "cold treatment" as the input and proceeding to the next step.

[0353] Step 2:

[0354] The emotion engine is activated. It collects and analyzes information such as the user's input method and speed, facial expressions, and tone of voice. For example, if the user's input speed is fast and they are putting a lot of force into the keys, it is likely that they are impatient. The emotion engine uses this data to recognize the user's emotional state (e.g., impatience, anxiety, relaxation). The user's behavioral data is taken as input, and the emotion analysis results are output.

[0355] Step 3:

[0356] The device sends the search keywords and emotion data to the server. Here, the device sends the keyword "cold treatment" and the emotion data (e.g., "impatience") analyzed by the emotion engine to the server. The search keywords and emotion data are sent to the server as input, and the next processing step is executed based on this.

[0357] Step 4:

[0358] The server searches for relevant web pages. Based on the received search keyword "cold treatment," the server uses an existing search algorithm to search for relevant web pages. At this time, emotion data is temporarily stored. The input is the search keyword, and the output is a list of relevant web pages.

[0359] Step 5:

[0360] The server sends the URLs of the web pages to the generation AI. The server then sends the URLs of each web page obtained as a search result to the generation AI module. The list of URLs obtained as input is sent, and analysis begins in the next step.

[0361] Step 6:

[0362] The generative AI analyzes the content of web pages. It downloads the HTML content of each web page and extracts key content and metadata. This analysis is performed using tools such as "BeautifulSoup" and "Selenium." The input is the HTML content, and the output is the parsed data.

[0363] Step 7:

[0364] Generative AI evaluates the reliability of terminology and sources. Based on the analyzed data, generative AI determines the use of terminology and the reliability of cited sources. For example, if the source is a public institution or academic paper, it will be assigned a high reliability score. The analyzed data is input, and a reliability evaluation score is generated as output.

[0365] Step 8:

[0366] The server calculates the reliability score. The server calculates the reliability score for each web page based on the evaluation results of the generation AI and data from the emotion engine. For example, if a user is feeling anxious, a high score is assigned to information that is highly reliable and gives a sense of security. The inputs are the evaluation results and emotion data, and the output is a reliability score.

[0367] Step 9:

[0368] The server integrates the reliability score into the search results. The calculated reliability score is incorporated into the search results and the display method is adjusted based on the user's emotional state. For example, for a user in a hurry, results are highlighted so that the reliability can be seen at a glance. The input is the reliability score and the search results, and the output is the adjusted search results.

[0369] Step 10:

[0370] The server sends the integrated results to the terminal. The server sends the integrated search results to the terminal, where they can be viewed by the user. The integrated search results are taken as input and sent to the terminal as output.

[0371] Step 11:

[0372] The device displays the search results. The device displays the search results in a format that takes the user's emotions into consideration. The user can review the displayed search results and select and view the optimal web page, adjusted by the reliability score and emotion engine. The input is the integrated results, and the output is the displayed search results.

[0373] (Application example 2)

[0374] 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."

[0375] Conventional navigation systems suggest routes without taking the user's emotional state into consideration, and therefore are unable to provide the optimal route, especially when the user is feeling anxious or impatient. Furthermore, they lack the reliability of search results and the ability to evaluate safety in real time, creating a need for a navigation system that users can use with peace of mind.

[0376] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for inputting search keywords, means for searching for related information based on the input search keywords, means for activating a generation AI for analyzing each piece of information obtained as a search result, means for evaluating the use of technical terms and the reliability of information sources using the generation AI, means for calculating a reliability score for each piece of information based on the evaluation results, means for integrating the reliability score into the search results, means for displaying the integrated search results, means for detecting the user's emotions using an emotion recognition device, and means for customizing the search results based on the user's emotion data. This enables personalized route selection according to the user's emotional state, thereby providing safe and reliable navigation.

[0377] A "search keyword" is a word or phrase that a user enters when searching for information.

[0378] "Search results" are a collection of related information obtained based on search keywords.

[0379] "Generative AI" is a system or program that uses artificial intelligence technology to analyze and evaluate information.

[0380] "Jargon" is specialized words and phrases used in a particular field.

[0381] "Source credibility" is a criterion for assessing the reliability of the source or source from which information is provided.

[0382] The "trustworthiness score" is a numerical representation of the reliability of information or sources evaluated by the generating AI.

[0383] An "emotion recognition device" is a device that uses a camera, a voice recognition system, etc. to detect a user's emotions in real time.

[0384] "Emotion data" is information that represents the emotional state of a user detected by an emotion recognition device.

[0385] A "means for customizing search results" is a method or apparatus that tailors and personalizes the search results obtained based on the user's emotional data.

[0386] The embodiment of the present invention will be described as a navigation system that operates within an autonomous vehicle. This system recognizes the user's emotions and uses generative AI to evaluate and provide road information and route reliability in real time. The following elements are required to implement this system:

[0387] System configuration

[0388] Enter search keywords

[0389] When a user inputs a destination into a navigation system, the system also recognizes the user's emotions.

[0390] User Emotion Recognition

[0391] The device is equipped with a camera and a voice recognition system that analyzes the user's facial expressions and voice in real time. This function uses a camera (e.g., Logitech C920) and a voice recognition system (e.g., Google Cloud Speech-to-Text). The recognized emotion data is sent to a server.

[0392] Find a route

[0393] The server searches for available routes based on the input destination, using traffic information services such as Google Maps API.

[0394] Analysis by generative AI

[0395] The route information obtained as a search result is analyzed by a generation AI on the server. The generation AI evaluates the latest road conditions and traffic information and quantifies the reliability of each route. This generation AI module uses models such as GPT-4.

[0396] Calculating the reliability score

[0397] Based on the data evaluated by the generative AI, a reliability score for each route is calculated using Scikit-learn's MLPClassifier or a custom function, and data from the emotion engine is also taken into account to adjust the score according to specific emotional states.

[0398] Customize route display

[0399] Based on the reliability score, the system displays the optimal route based on the user's emotional state on the device's navigation system screen. For example, it highlights safe and reliable routes for a user who is feeling anxious.

[0400] Specific examples

[0401] For example, if a user inputs the "safest route" into the navigation system of an autonomous vehicle, the system may recognize from the user's facial expressions and voice that they are feeling "anxiety" or "impatience." The emotion data and the searched route information are sent together to the server, and the generation AI analyzes these routes.

[0402] The generation AI calculates a reliability score for each route and uses prompts such as:

[0403] "If you want to avoid routes where there are traffic accidents, please provide a reliable route taking into account the current road conditions."

[0404] Based on this, the generation AI evaluates the most reliable route and displays a customized version that reflects the emotional data. Ultimately, route information that gives a sense of security to a user who is feeling anxious is displayed on the device, and the optimal route is suggested for the user.

[0405] In this way, the present invention provides a personalized navigation experience that responds to the user's emotional state.

[0406] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0407] Step 1:

[0408] The user inputs a destination into the navigation system inside the autonomous vehicle. The user's input is made via keyboard or voice. At this time, the camera and voice recognition system capture the user's facial expressions and voice in real time, and emotional data is acquired. The input destination data and emotional data are then sent to the terminal.

[0409] Step 2:

[0410] The device sends the acquired destination data and emotion data to the server. The server uses the destination data to search for multiple route information using the Google Maps API. The route information includes the distance, travel time, current traffic conditions, etc. for each route.

[0411] Server input: Destination data, emotion data

[0412] Server output: Route information

[0413] Step 3:

[0414] The server sends each route to the AI ​​generator, which analyzes the route and evaluates road conditions (traffic, accidents, construction, etc.) and other important factors. A generative AI model such as GPT-4 is used for the analysis. The evaluation results are quantified as a reliability score.

[0415] Input for the generation AI: Route information, prompt (e.g., "If I want to avoid a route where a traffic accident has occurred, please provide a reliable route taking into account the current road conditions.")

[0416] Generative AI output: Confidence score

[0417] Step 4:

[0418] The server adjusts the reliability score obtained by the AI ​​generator by taking into account emotional data. For example, if the user is in a hurry, it will give a higher score to safer and more reliable routes.

[0419] Server input: confidence score, sentiment data

[0420] Server output: Adjusted reliability score

[0421] Step 5:

[0422] The server reevaluates the reliability of each route based on the adjusted reliability score, selects the route that best suits the user's emotional state, and sends the selection result to the device.

[0423] Server Input: Adjusted Reliability Score

[0424] Server output: Optimal route information

[0425] Step 6:

[0426] The device displays the optimal route information sent from the server on the navigation system screen. When the user is in a hurry, reliable routes are visually highlighted, allowing the user to select them with confidence.

[0427] Input on device: Optimal route information

[0428] Device output: Route display on navigation screen

[0429] In this way, the present invention provides a navigation system that takes into account the emotional state of the user.

[0430] 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.

[0431] 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.

[0432] 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.

[0433] [Second embodiment]

[0434] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0435] 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.

[0436] 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).

[0437] 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.

[0438] 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.

[0439] 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).

[0440] 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.

[0441] 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.

[0442] 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.

[0443] 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.

[0444] 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.

[0445] 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."

[0446] The present invention provides a system for evaluating the reliability of web pages using generative AI, allowing users to easily select reliable information from search results.

[0447] System configuration

[0448] The system consists of the following main components:

[0449] 1. How to enter search keywords

[0450] 2. How to search for related web pages

[0451] 3. Web page analysis using generative AI

[0452] 4. Terminology and Source Credibility Assessment Tools

[0453] 5. How the reliability score is calculated

[0454] 6. Means of integrating trustworthiness scores into search results

[0455] 7. Display of integrated results

[0456] Program processing

[0457] Program Overview

[0458] The system operates through the following processes:

[0459] 1. Enter search keywords

[0460] 2. Obtaining search results

[0461] 3. Web page analysis

[0462] 4. Reliability evaluation

[0463] 5. Integration of evaluation results

[0464] 6. Display

[0465] The system begins with the user entering a search keyword. The device then sends the keyword to the server, which searches for relevant web pages. The resulting search results are analyzed by a generative AI to evaluate the trustworthiness of each web page. A trustworthiness score is then calculated and integrated into the search results. Finally, the search results with the associated trustworthiness score are displayed on the user's device.

[0466] Detailed program description

[0467] Enter search keywords

[0468] The user enters a keyword (e.g., "cold treatment") on the search screen and presses the search button, which starts the search process.

[0469] Getting search results

[0470] The device sends the input keywords to the server, which then searches for related web pages using existing search algorithms and temporarily stores the results.

[0471] Web page analysis

[0472] The server sends the URL of each web page retrieved as a search result to the generation AI module, which downloads the HTML content of each web page and analyzes it, including extracting the key content and metadata within the page.

[0473] Terminology and source credibility assessment

[0474] The generative AI evaluates the use of technical terms within a webpage and the reliability of the source. Specifically, it checks whether technical terms are used accurately and evaluates whether the cited sources are trustworthy. For example, if the source is a public institution or an academic paper, it will assign a high reliability score.

[0475] Calculating the reliability score

[0476] The server calculates a credibility score for each web page based on the results evaluated by the AI, which is expressed as a number or letter so that users can determine the trustworthiness of each search result at a glance.

[0477] Integrating trust scores into search results

[0478] The server integrates the authority score into the search results by adding the authority score below the title of each web page.

[0479] Viewing the integration results

[0480] The terminal displays the integrated search results received from the server to the user, who can then review the displayed search results and select which web page to view based on the reliability score.

[0481] Specific examples

[0482] For example, when searching for the keyword "cold treatment," the user first enters the keyword and presses the search button. The device sends the keyword to the server, which retrieves approximately 10 related web pages. The server then analyzes each web page using generative AI to evaluate the use of technical terms and the reliability of cited sources. As a result of the evaluation, a reliability score is calculated, and the server integrates the reliability score into the search results. Finally, the device displays the search results with the reliability score assigned to them to the user. The user can check the reliability score of each web page and select the most reliable page to view.

[0483] As described above, the system of the present invention helps users to easily and accurately determine the reliability of web pages, and can particularly increase the reliability of pages that contain specialized information.

[0484] The processing flow will be explained below.

[0485] Step 1: Enter search keywords

[0486] 1. A user enters a keyword (e.g., "cold treatment") into the search field of a search engine.

[0487] 2. The user presses the "Search" button.

[0488] Step 2: Getting search results

[0489] 1. The device sends the entered search keywords to the server.

[0490] 2. The server searches for relevant web pages based on the keywords.

[0491] 3. The server generates a list of URLs for the web pages retrieved as search results.

[0492] 4. The server temporarily stores the URL list.

[0493] Step 3: Analyze the web page

[0494] 1. The server retrieves the URL of each web page from the URL list in turn.

[0495] 2. The server downloads the HTML content of each web page.

[0496] 3. The server sends the downloaded HTML content to the generation AI module.

[0497] 4. Generative AI analyzes the page structure and extracts key content and metadata.

[0498] Step 4: Assess terminology and source credibility

[0499] 1. Analyze the key content extracted by the generative AI and evaluate the use of technical terms.

[0500] 2. Generative AI analyzes citations and links within web pages and evaluates the source of the citation or link.

[0501] 3. If the generating AI determines that the source is based on a public institution or academic paper, it will assign a higher credibility score to that source.

[0502] Step 5: Calculate the reliability score

[0503] 1. The server calculates the trustworthiness score for each web page based on the content evaluated by the AI.

[0504] 2. The server temporarily stores the credibility score for each web page.

[0505] Step 6: Integrating Trust Scores into Search Results

[0506] 1. The server integrates the confidence score into the search results.

[0507] 2. The server creates a data structure that adds an authority score to each web page below its title.

[0508] Step 7: Viewing the integration results

[0509] 1. The server sends the consolidated search results to the device.

[0510] 2. The terminal displays the search results received to the user.

[0511] 3. Users can see the credibility score under the title of each search result displayed.

[0512] Example 1

[0513] 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."

[0514] The amount of information on the Internet is enormous, making it difficult for users to quickly find reliable information. This problem is particularly pronounced in fields where reliability is essential, such as specialized or medical information. Furthermore, existing search engines lack the functionality to evaluate the reliability of search results, leaving users with insufficient means to select appropriate information.

[0515] 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.

[0516] In this invention, the server includes means for inputting search keywords, means for searching for related web pages based on the input search keywords, means for activating a generation AI to analyze each web page obtained as a search result, means for evaluating the use of technical terms and the reliability of information sources using the generation AI, means for calculating a reliability score for each web page based on the evaluation results, means for integrating the reliability scores into search results, means for displaying the integrated search results, means for transmitting the search keywords to the server and temporarily saving the obtained search results, means for the generation AI to download the HTML content of the web page and extract key content and metadata within the page, and means for the user's terminal to display the integrated search results received from the server, thereby enabling users to quickly identify reliable information and browse appropriate information sources.

[0517] A "search keyword" is a character string that a user inputs to search for information.

[0518] The "input means" is a device or interface for a user to input search keywords.

[0519] A "search tool" is a function or process that searches the Internet for relevant web pages based on input search keywords.

[0520] "Generative AI" is a system that uses artificial intelligence technology to analyze data and perform evaluation and generation.

[0521] The "analysis means" is a function that analyzes the acquired web page in detail and extracts information.

[0522] "Terminology" is a term that has a specialized meaning in a particular field.

[0523] A "source" is the source or citation of information contained within a web page.

[0524] The "trustworthiness assessment tool" is a function that uses generative AI to evaluate the use of terminology within a web page and the reliability of the source of information.

[0525] The "trustworthiness score" is an evaluation result that indicates the trustworthiness of each web page numerically or alphabetically.

[0526] The "calculation means" is a function that calculates a reliability score based on the evaluation results of the generating AI.

[0527] The "search result integration means" is a function that incorporates reliability scores into search results.

[0528] The "display means" is a function that displays the integrated search results on the user's terminal.

[0529] A "server" is a computer system for processing, storing, and distributing data.

[0530] A "terminal" is an electronic device that a user operates, and typically refers to a PC or smartphone.

[0531] "HTML content" is data written in a markup language that makes up a web page.

[0532] "Metadata" is data that provides additional information about the content of a web page.

[0533] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention is a system that utilizes a generative AI model to evaluate the trustworthiness of web pages, allowing users to easily select reliable information from search results.

[0534] System configuration

[0535] The system consists of the following main components:

[0536] 1. How to enter search keywords

[0537] 2. A way to find related web pages

[0538] 3. A means to trigger a generative AI to analyze each web page retrieved as a search result.

[0539] 4. Using generative AI to assess terminology and source credibility

[0540] 5. A method for calculating the credibility score of each web page based on the evaluation results

[0541] 6. A way to integrate trustworthiness scores into search results

[0542] 7. A way to display consolidated search results

[0543] 8. A means of sending search keywords to a server and temporarily storing the retrieved search results

[0544] 9. How generative AI downloads the HTML content of a webpage and extracts key content and metadata from the page.

[0545] 10. Means for displaying the integrated search results received by the user's device from the server

[0546] Hardware and software used

[0547] Client terminal: PC or smartphone used by the user

[0548] Server: A server (e.g., a cloud computing service) to host the search engine and the generative AI.

[0549] Generative AI models: Natural language processing models such as GPT-4

[0550] Program processing overview

[0551] When a user enters a specific keyword (e.g., "cold treatment") on the search screen and presses the search button, the device sends the keyword to the server. The server uses an existing search engine to search for related web pages and temporarily stores the retrieved results. The server then passes the temporarily stored URL list of search results to the generation AI module, which downloads and analyzes the HTML content of each web page. Based on the analyzed text content, the generation AI evaluates the use of technical terms and the reliability of cited sources. The server then calculates a reliability score for each web page based on the generation AI's evaluation results and integrates the reliability scores into the search results. Finally, the device displays the integrated search results received from the server to the user, who then selects which web page to view based on the reliability score.

[0552] Specific examples

[0553] For example, let's take a specific example of searching for the keyword "cold treatment." When a user enters the keyword and presses the search button, the device sends the keyword to the server, which retrieves approximately 10 related web pages. The server then uses generative AI to analyze each web page and evaluates the use of technical terms and the reliability of cited sources. As a result of the evaluation, a reliability score is calculated, and the server integrates this score into the search results. Finally, the device displays the search results with the assigned reliability score to the user. The user can check the reliability score of each web page and select the most reliable page to view.

[0554] Example prompts for generative AI models

[0555] "Evaluate the titles and content of web pages and calculate a credibility score based on the following criteria: accurate use of technical terminology, credibility of cited sources (e.g., official institutions, academic papers, news outlets, etc.). For example, if you are evaluating web pages resulting from a search for the keyword 'cold treatment methods,' rate each page using the criteria above."

[0556] The above describes the embodiments of the present invention, which enable a user to quickly identify reliable information and browse appropriate information sources.

[0557] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0558] Step 1:

[0559] The user enters a specific keyword (e.g., "cold treatment") on the search screen and presses the search button. The input here is the action of the user entering text into the search box, and this text becomes the input data for the next step. The device receives the user's input and prepares to send that information to the server. In concrete terms, the user opens a search form in a browser, enters "cold treatment" into the search box, and clicks the search button.

[0560] Step 2:

[0561] The device sends the entered keywords to the server, which then uses the entered keywords for the next search process. The server uses a wide range of search engines to search for related web pages based on the received keywords and temporarily stores the results. Specifically, the device sends the user's input information to the server using a communication protocol (e.g., HTTP), and the server uses Google's search API or similar to obtain a list of URLs for related pages. The obtained list of URLs is temporarily stored in memory.

[0562] Step 3:

[0563] The server passes the temporarily saved list of search result URLs to the generation AI module. This becomes the input data. The generation AI downloads the HTML content of each web page and analyzes it as text. Specifically, the server passes the URL list to the generation AI, which then uses an HTTP request to retrieve the HTML of each web page, analyzes it, and extracts the text content and metadata.

[0564] Step 4:

[0565] Based on the text content analyzed by the generative AI, the usage of technical terms and the reliability of cited sources are evaluated. This serves as input data for the next evaluation step. Specifically, technical terms are extracted from the analyzed text and checked to see if they are used accurately. The reliability of cited sources, such as public institutions and academic papers, is also evaluated. Specifically, the generative AI extracts technical terms from the text and checks the cited sources.

[0566] Step 5:

[0567] The server calculates a reliability score for each web page based on the evaluation results of the generation AI. This becomes the input data for the evaluation score used in the next step. The reliability score serves as an indicator when users check the results. Specifically, the server receives the evaluation data from the generation AI, inputs it into the scoring algorithm, and calculates the reliability score. The calculated score is generated in a form corresponding to the URL of each web page.

[0568] Step 6:

[0569] The server integrates the calculated reliability score into the search results, generating integrated data. This integration result becomes the input data for the next display step. The reliability score is integrated by adding it below the title of each web page. Specifically, the server compares the list of search result URLs with the calculated reliability score, adds the reliability score to the metadata of each URL, and generates the final list of search results.

[0570] Step 7:

[0571] The terminal displays the integrated search results received from the server to the user. This is the final output data. The user can check the displayed search results and select which web page to view based on the reliability score. Specifically, the terminal receives the response from the server, analyzes the response data, and converts it into a format to be displayed in the browser. The user then clicks on each web page to view it based on the reliability score.

[0572] (Application example 1)

[0573] 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."

[0574] In today's world, it is extremely important for users to verify the reliability of information they search for on the Internet. Particularly on online shopping sites, the reliability of product descriptions and reviews significantly influences users' purchasing decisions. However, current search result displays do not provide a concrete means for evaluating reliability, leaving users to make their own judgments about the quality of information. To address this issue, the present invention aims to provide a system that automatically assigns reliability scores to search results on online shopping sites, allowing users to easily select highly reliable information.

[0575] 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.

[0576] In this invention, the server includes means for inputting search keywords, means for searching for related web pages based on the input search keywords, means for activating a generation AI for analyzing each web page obtained as a search result, means for evaluating the use of technical terms and the reliability of information sources using the generation AI, means for calculating a reliability score for each web page based on the evaluation results, means for integrating the reliability score into the search results, means for displaying the integrated search results, means for analyzing the content of each product information page and evaluating the quality of product descriptions and reviews, and means for assigning a reliability score to the evaluated product information page. This allows users to easily select reliable product information pages and make purchasing decisions with confidence.

[0577] "Search keywords" are words or phrases that users enter to describe the information they are looking for.

[0578] A "web page" is a document containing information that can be accessed on the Internet and is written in a format such as HTML.

[0579] "Generative AI" is a system that uses artificial intelligence technology to generate text and analyze information.

[0580] "Jargon" is a specialized word or phrase used in a particular field or area.

[0581] A "source" is a source or data source from which particular information is obtained.

[0582] A "credibility score" is a number or indicator used to evaluate the reliability of a particular piece of information or source of information.

[0583] "Content" refers to the information contained in a web page or document, such as text, images, or video.

[0584] A "review" is an evaluation or comment written by a user or purchaser describing their own experience or opinion.

[0585] "Search results" are the list of relevant web pages or information that appears when a user enters keywords into a search engine.

[0586] A "product information page" is a web page that contains detailed descriptions, specifications, and reviews of a particular product.

[0587] This invention relates to a system that evaluates the reliability of each product information page when a user searches for a product on an online shopping site, allowing the user to make a purchasing decision based on reliable information. This system is realized by using generative AI to evaluate the use of technical terms and the reliability of information sources, from the input of search keywords to the display of search results, and calculates a reliability score.

[0588] System configuration

[0589] The system consists of the following main components:

[0590] 1. How to enter search keywords

[0591] 2. How to search for related web pages

[0592] 3. Web page analysis using generative AI

[0593] 4. Terminology and Source Credibility Assessment Tools

[0594] 5. How the reliability score is calculated

[0595] 6. A way to integrate trustworthiness scores into search results

[0596] 7. Display of integrated results

[0597] 8. A way to analyze the content of each product page and evaluate the quality of the product descriptions and reviews.

[0598] 9. A means of assigning a credibility score to rated product information pages

[0599] Program processing

[0600] 1. Enter search keywords

[0601] The user enters the product name into the smartphone app and presses the search button.

[0602] 2. Obtaining search results

[0603] The device sends the entered search keywords to the server, which retrieves related product information pages. The search results are temporarily saved.

[0604] 3. Web page analysis

[0605] The server sends the URL of each product information page retrieved as a search result to the generation AI module, which downloads the HTML content of each web page and analyzes it, including extracting the key content and metadata within the page.

[0606] 4. Terminology and source credibility assessment

[0607] The generative AI evaluates the use of technical terms on product information pages and the reliability of the sources. Specifically, it checks whether technical terms are used accurately and evaluates whether the cited sources are trustworthy. For example, if the source is a public institution or an academic paper, it will assign a high reliability score.

[0608] 5. Calculating the reliability score

[0609] The server calculates a reliability score for each product information page based on the results of the AI ​​generation. The reliability score is displayed as a number or letter, allowing users to determine the reliability of each product information page at a glance.

[0610] 6. Integrating Trust Scores into Search Results

[0611] The server integrates the reliability score into the search results by adding it below the title of each product information page.

[0612] 7. Displaying the integrated results

[0613] The terminal displays the integrated search results received from the server to the user, who can then review the displayed search results and select which information page to view based on the reliability score.

[0614] Hardware and software used

[0615] Hardware: Servers, smartphones

[0616] Software: Generative AI (GPT-4, etc.), Python, web frameworks (Django, Flask), databases (PostgreSQL)

[0617] Specific examples

[0618] For example, suppose a user enters "vitamin supplements" into a smartphone app and presses the search button. The device sends this search keyword to the server, which retrieves related product information pages. The server then uses generative AI to analyze each product information page and evaluates the use of technical terms and the reliability of the information source. A reliability score is calculated based on the evaluation results, and the server integrates the reliability score into the search results. The device then displays the search results with the reliability score to the user.

[0619] Prompt Sentence Examples

[0620] The prompt for the search "vitamin supplement" is:

[0621] Rate the trustworthiness of this product information page:

[0622] URL: https: / / example.com / vitamin-supplement

[0623] Key points: Accuracy of product description, quality of reviews, citation of reliable sources

[0624] The analysis results obtained from the generative AI model include numerical values ​​such as "content accuracy: 85%" and "source reliability: 90%." A reliability score is calculated based on this and displayed to the user. Users can then select products and make purchasing decisions with confidence based on the reliability score.

[0625] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0626] Step 1:

[0627] The user enters the product name into the smartphone app and presses the search button. The entered keywords are sent to the device as "search keywords." The entered data is the product name specified by the user, and this becomes the base data for subsequent processing.

[0628] Step 2:

[0629] The server receives the "search keywords" sent from the terminal. The server uses existing search engine functions to search for the relevant product information page. At this point, the only input is the "search keywords," and the output is a list of URLs for related product information pages.

[0630] Step 3:

[0631] The server sends a list of URLs for each product information page obtained as a search result to the generation AI module. The generation AI downloads the HTML content of each product and analyzes it. The analysis includes extracting the main content and metadata within the page. The input data is the list of URLs for the product information pages, and the output data is the analyzed HTML content.

[0632] Step 4:

[0633] The generative AI evaluates the use of technical terms on product information pages and the reliability of the sources. Specifically, it checks whether technical terms are used accurately and evaluates whether the cited sources are trustworthy. If highly reliable sources such as public institutions or academic papers are cited, the rating is higher. The input data is HTML content, and the output data is a reliability rating score.

[0634] Step 5:

[0635] The server calculates the reliability score of each product information page based on the results of the evaluation by the generation AI. The reliability score is calculated based on a specific algorithm and is displayed as a number or letter that allows users to determine the reliability of each product information page at a glance. The input data is the reliability evaluation score of each product information page, and the output data is the reliability score.

[0636] Step 6:

[0637] The server integrates the reliability scores into the search results. Specifically, it adds the reliability score below the title of each product information page. This process visually displays the reliability rating for all search results. The input data is the reliability score and a list of search results, and the output data is a list of search results with the reliability scores added.

[0638] Step 7:

[0639] The terminal displays the integrated search results with the reliability scores received from the server to the user. The user checks the displayed search results and selects which product information page to view based on the reliability scores. The input data are the search results with the reliability scores added, and the output data are the product information pages displayed to the user.

[0640] Through the above steps, the user can easily select highly reliable product information and make a purchase decision with confidence.

[0641] 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.

[0642] The present invention provides a system for providing more personalized search results by combining a conventional system that uses generative AI to evaluate the trustworthiness of web pages with an emotion engine that recognizes user emotions.

[0643] System configuration

[0644] The system consists of the following main components:

[0645] 1. How to enter search keywords

[0646] 2. How to search for related web pages

[0647] 3. Web page analysis using generative AI

[0648] 4. Terminology and Source Credibility Assessment Tools

[0649] 5. How the reliability score is calculated

[0650] 6. Means of integrating trustworthiness scores into search results

[0651] 7. Display of integrated results

[0652] 8. Emotion engine that recognizes user emotions

[0653] 9. Sentiment-Based Search Results Customization

[0654] Program processing

[0655] The system proceeds as follows:

[0656] Enter search keywords

[0657] A user enters a keyword (e.g., "cold treatment") into the search field of a search engine and presses the search button. This starts the search process, but at this point the emotion engine is also activated. The emotion engine analyzes the user's input method and speed, facial expressions, tone of voice, etc., to recognize the user's emotional state (e.g., impatience, anxiety, relaxation, etc.).

[0658] Getting search results

[0659] The device sends the emotion data recognized by the emotion engine along with the entered search keywords to the server, where the server searches for related web pages using existing search algorithms and temporarily stores the results.

[0660] Web page analysis

[0661] The server sends the URL of each web page retrieved as a search result to the generation AI module, which downloads the HTML content of each web page and analyzes it, including extracting the key content and metadata within the page.

[0662] Terminology and source credibility assessment

[0663] The generative AI evaluates the use of domain-specific terminology and determines the trustworthiness of cited sources. For example, it assigns a higher credibility score to sources from public institutions or academic papers. Furthermore, it can prioritize sources that inspire a sense of security based on the user's emotional state.

[0664] Calculating the reliability score

[0665] The server calculates a credibility score for each web page based on the evaluation results of the generation AI. In addition, it takes into account data from the emotion engine and adjusts the score according to the user's current emotional state. For example, if a user is feeling anxious, it assigns a higher score to information that is highly reliable and reassuring.

[0666] Integrating trust scores into search results

[0667] The server integrates the reliability score into the search results, and adjusts the display of search results based on the user's emotional state. For example, if a user is in a hurry, the server highlights the results so that the reliability can be seen at a glance.

[0668] Viewing the integration results

[0669] The server sends the integrated search results to the device, which then displays the results in a format that takes the user's emotions into consideration.The user can then review the displayed search results and select the most appropriate web page to view, with the reliability score and emotional engine having been adjusted accordingly.

[0670] Specific examples

[0671] For example, if the emotion engine recognizes that a user searching for the keyword "cold treatment" is feeling anxious or impatient when typing, the emotion engine sends this information to the server. The server retrieves search results based on that information, and the generation AI analyzes the webpage. It then evaluates the reliability of the terminology and information source, and calculates a reliability score taking into account the emotional data. Finally, when the search results incorporating the reliability score are sent to the user's device, reliable information that will reassure users who are feeling anxious is highlighted.

[0672] This allows users to easily access the most appropriate information based on their emotional state at the time, providing a more personalized search experience.

[0673] The processing flow will be explained below.

[0674] Step 1: Enter search keywords

[0675] 1. A user enters a keyword (e.g., "cold treatment") into the search field of a search engine.

[0676] 2. The user presses the "Search" button.

[0677] 3. The device monitors the user's input method (speed, rhythm, etc.), facial recognition camera, and tone of voice during voice input, and activates the emotion engine.

[0678] 4. The emotion engine analyzes the user's emotional state (e.g., impatience, anxiety, relaxation, etc.) and generates emotion data.

[0679] Step 2: Getting search results

[0680] 1. The device sends the entered search keywords and the emotion data recognized by the emotion engine to the server.

[0681] 2. The server searches for relevant web pages based on the keywords.

[0682] 3. The server generates a list of URLs of the web pages obtained as search results and temporarily stores them.

[0683] Step 3: Analyze the web page

[0684] 1. The server retrieves the URL of each web page from the URL list in turn.

[0685] 2. The server downloads the HTML content of each web page.

[0686] 3. The server sends the downloaded HTML content to the generation AI module.

[0687] 4. Generative AI analyzes the page structure and extracts key content and metadata.

[0688] Step 4: Assess terminology and source credibility

[0689] 1. Analyze the key content extracted by the generative AI and evaluate the use of technical terms.

[0690] 2. Generative AI analyzes citations and links within web pages and evaluates the source of the citation or link.

[0691] 3. If the generating AI determines that the source is based on a public institution or academic paper, it will assign a higher credibility score to that source.

[0692] 4. The generative AI uses data from the emotion engine to adjust the importance of information according to the user's emotional state. For example, if a user is feeling anxious, information that gives a sense of security will be given a high score.

[0693] Step 5: Calculate the reliability score

[0694] 1. The server calculates the trustworthiness score for each web page based on the content evaluated by the AI.

[0695] 2. The server also takes into account data from the emotion engine and adjusts the reliability score according to the user's emotional state.

[0696] 3. The server generates and temporarily stores a final credibility score for each web page.

[0697] Step 6: Integrating Trust Scores into Search Results

[0698] 1. The server integrates the confidence score into the search results.

[0699] 2. The server creates a data structure that adds an authority score to each web page below its title.

[0700] 3. The emotion engine determines how search results are displayed (e.g., highlighted, color-coded) depending on the user's emotional state.

[0701] Step 7: Viewing the integration results

[0702] 1. The server sends the consolidated search results to the device.

[0703] 2. The search results received by the device are displayed in a format that takes into account the user's emotional state.

[0704] 3. Users can see the credibility score under the title of each search result displayed and view customized information based on their emotional state.

[0705] Through the above steps, the system of the present invention provides personalized search results that adapt to the user's emotional state to enhance the user's search experience.

[0706] Example 2

[0707] 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."

[0708] Conventional search systems present relevant information based on keywords entered by the user, but because they do not take the user's emotional state into consideration, it is difficult to provide the optimal information the user is looking for. As a result, even if the user is feeling anxious or impatient, information that provides a sense of security is not prioritized, resulting in a poor user experience.

[0709] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for inputting a search keyword; means for searching for related web pages based on the input search keyword and user emotional data; means for activating an emotional engine and analyzing the user's emotional state; means for activating a generation AI for analyzing each web page obtained as a search result; means for evaluating the use of technical terms and the reliability of information sources using the generation AI; means for calculating a reliability score for each web page taking into account the user's emotional data; means for integrating the reliability score into the search result and adjusting the display method based on the user's emotional state; and means for displaying the integrated search result. This makes it possible to provide personalized search results that take into account the user's emotional state.

[0710] The "means for inputting search keywords" refers to a device or function that allows a user to input search keywords through the interface of a search engine.

[0711] "Means for searching for related web pages based on input search keywords and user emotion data" refers to a device or function for finding related web pages using search keywords input by the user and emotion data analyzed by the emotion engine.

[0712] "Means for activating an emotion engine and analyzing the user's emotional state" refers to a device or function for recognizing the user's emotions (e.g., impatience, anxiety, relaxation) by analyzing the user's input method, speed, facial expression, tone of voice, etc.

[0713] "Means for launching a generative AI to analyze each web page obtained as a search result" refers to a device or function that launches a generative AI model to analyze the content of each web page obtained as a search result.

[0714] "Means for using generative AI to evaluate terminology usage and source reliability" refers to a device or function that uses generative AI to analyze and evaluate the use of terminology within a web page and the reliability of cited sources.

[0715] The "means for calculating the reliability score of each web page taking into account the user's emotional data" refers to a device or function for calculating the reliability score of each web page taking into account the evaluation results of the generation AI and the user's emotional state.

[0716] "Means for integrating confidence scores into search results and adjusting the display based on the user's emotional state" refers to a device or function for incorporating calculated confidence scores into search results and changing the display of search results depending on the user's emotional state.

[0717] The "means for displaying integrated search results" is a device or function for displaying search results including reliability scores on a user's terminal so that the user can view them.

[0718] The following describes in detail the mode for carrying out the present invention: The system utilizes generative AI to evaluate the trustworthiness of web pages, and combines it with an emotion engine that recognizes user emotions to provide more personalized search results.

[0719] System configuration

[0720] The system consists of the following main elements:

[0721] 1. How to enter search keywords

[0722] 2. How to search for related web pages

[0723] 3. Web page analysis using generative AI

[0724] 4. Terminology and Source Credibility Assessment Tools

[0725] 5. How the reliability score is calculated

[0726] 6. Means of integrating trustworthiness scores into search results

[0727] 7. Display of integrated results

[0728] 8. Emotion engine that recognizes user emotions

[0729] 9. Sentiment-Based Search Results Customization

[0730] Enter search keywords

[0731] A user enters a keyword (e.g., "cold treatment") into the search field of a search engine and presses the search button. This action starts the search process and simultaneously activates the emotion engine. The emotion engine analyzes the user's input method, speed, facial expression, tone of voice, etc., to recognize the user's emotional state (e.g., impatience, anxiety, relaxation, etc.).

[0732] Getting search results

[0733] The device sends the entered search keywords and the emotion data recognized by the emotion engine to the server, which then uses existing search algorithms to search for related web pages and temporarily stores the results.

[0734] Web page analysis

[0735] The server sends the URL of each web page retrieved as a search result to the generative AI module. The generative AI downloads the HTML content of each web page, extracts key content and metadata, and analyzes it. The generative AI model can be implemented using Python libraries such as "BeautifulSoup" and "Selenium."

[0736] Terminology and source credibility assessment

[0737] Based on the analyzed content, the generative AI evaluates the use of technical terms and the reliability of cited sources. For example, if the source is a public institution or academic paper, it will give a high reliability score. It can also take into account the user's emotional state and prioritize sources that give a sense of security.

[0738] Calculating the reliability score

[0739] The server calculates a reliability score for each web page based on the evaluation results of the generated AI and data from the emotion engine. For example, if a user is feeling anxious, a high score can be assigned to information that is highly reliable and reassuring.

[0740] Integrating trust scores into search results

[0741] The server integrates the reliability score into search results, adjusting the presentation based on the user's emotional state—for example, highlighting results to make their reliability clear at a glance to users in a hurry.

[0742] Viewing the integration results

[0743] The server sends the integrated search results to the device, which then displays the search results in a format that takes the user's emotions into consideration.The user can review the displayed search results and select the most appropriate web page to view, with the results adjusted by the reliability score and emotion engine.

[0744] Specific examples

[0745] For example, if the emotion engine recognizes that a user searching for the keyword "cold treatment" is feeling anxious or impatient when typing, the emotion engine sends this emotional data to the server. The server retrieves search results based on that information, and the generation AI analyzes the webpage. The analysis results then evaluate the reliability of the terminology and information sources, and calculate a reliability score taking the emotional data into account. Finally, the search results, incorporating the reliability score, are sent to the user's device, and reliable information that will reassure anxious users is highlighted and displayed.

[0746] Example prompt

[0747] If the user enters the keyword "cold treatment" and the emotion engine recognizes that the user is feeling anxious, prioritize displaying reliable information that provides a sense of security.

[0748] This system allows users to easily obtain the most appropriate information based on their emotional state at the time, providing a more personalized search experience.

[0749] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0750] Step 1:

[0751] The user enters a search keyword. Specifically, the user enters "cold treatment" in the search field of the search engine and presses the search button. This input triggers the search process, obtaining "cold treatment" as the input and proceeding to the next step.

[0752] Step 2:

[0753] The emotion engine is activated. It collects and analyzes information such as the user's input method and speed, facial expressions, and tone of voice. For example, if the user's input speed is fast and they are putting a lot of force into the keys, it is likely that they are impatient. The emotion engine uses this data to recognize the user's emotional state (e.g., impatience, anxiety, relaxation). The user's behavioral data is taken as input, and the emotion analysis results are output.

[0754] Step 3:

[0755] The device sends the search keywords and emotion data to the server. Here, the device sends the keyword "cold treatment" and the emotion data (e.g., "impatience") analyzed by the emotion engine to the server. The search keywords and emotion data are sent to the server as input, and the next processing step is executed based on this.

[0756] Step 4:

[0757] The server searches for relevant web pages. Based on the received search keyword "cold treatment," the server uses an existing search algorithm to search for relevant web pages. At this time, emotion data is temporarily stored. The input is the search keyword, and the output is a list of relevant web pages.

[0758] Step 5:

[0759] The server sends the URLs of the web pages to the generation AI. The server then sends the URLs of each web page obtained as a search result to the generation AI module. The list of URLs obtained as input is sent, and analysis begins in the next step.

[0760] Step 6:

[0761] The generative AI analyzes the content of web pages. It downloads the HTML content of each web page and extracts key content and metadata. This analysis is performed using tools such as "BeautifulSoup" and "Selenium." The input is the HTML content, and the output is the parsed data.

[0762] Step 7:

[0763] Generative AI evaluates the reliability of terminology and sources. Based on the analyzed data, generative AI determines the use of terminology and the reliability of cited sources. For example, if the source is a public institution or academic paper, it will be assigned a high reliability score. The analyzed data is input, and a reliability evaluation score is generated as output.

[0764] Step 8:

[0765] The server calculates the reliability score. The server calculates the reliability score for each web page based on the evaluation results of the generation AI and data from the emotion engine. For example, if a user is feeling anxious, a high score is assigned to information that is highly reliable and gives a sense of security. The inputs are the evaluation results and emotion data, and the output is a reliability score.

[0766] Step 9:

[0767] The server integrates the reliability score into the search results. The calculated reliability score is incorporated into the search results and the display method is adjusted based on the user's emotional state. For example, for a user in a hurry, results are highlighted so that the reliability can be seen at a glance. The input is the reliability score and the search results, and the output is the adjusted search results.

[0768] Step 10:

[0769] The server sends the integrated results to the terminal. The server sends the integrated search results to the terminal, where they can be viewed by the user. The integrated search results are taken as input and sent to the terminal as output.

[0770] Step 11:

[0771] The device displays the search results. The device displays the search results in a format that takes the user's emotions into consideration. The user can review the displayed search results and select and view the optimal web page, adjusted by the reliability score and emotion engine. The input is the integrated results, and the output is the displayed search results.

[0772] (Application example 2)

[0773] 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."

[0774] Conventional navigation systems suggest routes without taking the user's emotional state into consideration, and therefore are unable to provide the optimal route, especially when the user is feeling anxious or impatient. Furthermore, they lack the reliability of search results and the ability to evaluate safety in real time, creating a need for a navigation system that users can use with peace of mind.

[0775] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for inputting search keywords, means for searching for related information based on the input search keywords, means for activating a generation AI for analyzing each piece of information obtained as a search result, means for evaluating the use of technical terms and the reliability of information sources using the generation AI, means for calculating a reliability score for each piece of information based on the evaluation results, means for integrating the reliability score into the search results, means for displaying the integrated search results, means for detecting the user's emotions using an emotion recognition device, and means for customizing the search results based on the user's emotion data. This enables personalized route selection according to the user's emotional state, thereby providing safe and reliable navigation.

[0776] A "search keyword" is a word or phrase that a user enters when searching for information.

[0777] "Search results" are a collection of related information obtained based on search keywords.

[0778] "Generative AI" is a system or program that uses artificial intelligence technology to analyze and evaluate information.

[0779] "Jargon" is specialized words and phrases used in a particular field.

[0780] "Source credibility" is a criterion for assessing the reliability of the source or source from which information is provided.

[0781] The "trustworthiness score" is a numerical representation of the reliability of information or sources evaluated by the generating AI.

[0782] An "emotion recognition device" is a device that uses a camera, a voice recognition system, etc. to detect a user's emotions in real time.

[0783] "Emotion data" is information that represents the emotional state of a user detected by an emotion recognition device.

[0784] A "means for customizing search results" is a method or apparatus that tailors and personalizes the search results obtained based on the user's emotional data.

[0785] The embodiment of the present invention will be described as a navigation system that operates within an autonomous vehicle. This system recognizes the user's emotions and uses generative AI to evaluate and provide road information and route reliability in real time. The following elements are required to implement this system:

[0786] System configuration

[0787] Enter search keywords

[0788] When a user inputs a destination into a navigation system, the system also recognizes the user's emotions.

[0789] User Emotion Recognition

[0790] The device is equipped with a camera and a voice recognition system that analyzes the user's facial expressions and voice in real time. This function uses a camera (e.g., Logitech C920) and a voice recognition system (e.g., Google Cloud Speech-to-Text). The recognized emotion data is sent to a server.

[0791] Find a route

[0792] The server searches for available routes based on the input destination, using traffic information services such as Google Maps API.

[0793] Analysis by generative AI

[0794] The route information obtained as a search result is analyzed by a generation AI on the server. The generation AI evaluates the latest road conditions and traffic information and quantifies the reliability of each route. This generation AI module uses models such as GPT-4.

[0795] Calculating the reliability score

[0796] Based on the data evaluated by the generative AI, a reliability score for each route is calculated using Scikit-learn's MLPClassifier or a custom function, and data from the emotion engine is also taken into account to adjust the score according to specific emotional states.

[0797] Customize route display

[0798] Based on the reliability score, the system displays the optimal route based on the user's emotional state on the device's navigation system screen. For example, it highlights safe and reliable routes for a user who is feeling anxious.

[0799] Specific examples

[0800] For example, if a user inputs the "safest route" into the navigation system of an autonomous vehicle, the system may recognize from the user's facial expressions and voice that they are feeling "anxiety" or "impatience." The emotion data and the searched route information are sent together to the server, and the generation AI analyzes these routes.

[0801] The generation AI calculates a reliability score for each route and uses prompts such as:

[0802] "If you want to avoid routes where there are traffic accidents, please provide a reliable route taking into account the current road conditions."

[0803] Based on this, the generation AI evaluates the most reliable route and displays a customized version that reflects the emotional data. Ultimately, route information that gives a sense of security to a user who is feeling anxious is displayed on the device, and the optimal route is suggested for the user.

[0804] In this way, the present invention provides a personalized navigation experience that responds to the user's emotional state.

[0805] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0806] Step 1:

[0807] The user inputs a destination into the navigation system inside the autonomous vehicle. The user's input is made via keyboard or voice. At this time, the camera and voice recognition system capture the user's facial expressions and voice in real time, and emotional data is acquired. The input destination data and emotional data are then sent to the terminal.

[0808] Step 2:

[0809] The device sends the acquired destination data and emotion data to the server. The server uses the destination data to search for multiple route information using the Google Maps API. The route information includes the distance, travel time, current traffic conditions, etc. for each route.

[0810] Server input: Destination data, emotion data

[0811] Server output: Route information

[0812] Step 3:

[0813] The server sends each route to the AI ​​generator, which analyzes the route and evaluates road conditions (traffic, accidents, construction, etc.) and other important factors. A generative AI model such as GPT-4 is used for the analysis. The evaluation results are quantified as a reliability score.

[0814] Input for the generation AI: Route information, prompt (e.g., "If I want to avoid a route where a traffic accident has occurred, please provide a reliable route taking into account the current road conditions.")

[0815] Generative AI output: Confidence score

[0816] Step 4:

[0817] The server adjusts the reliability score obtained by the AI ​​generator by taking into account emotional data. For example, if the user is in a hurry, it will give a higher score to safer and more reliable routes.

[0818] Server input: confidence score, sentiment data

[0819] Server output: Adjusted reliability score

[0820] Step 5:

[0821] The server reevaluates the reliability of each route based on the adjusted reliability score, selects the route that best suits the user's emotional state, and sends the selection result to the device.

[0822] Server Input: Adjusted Reliability Score

[0823] Server output: Optimal route information

[0824] Step 6:

[0825] The device displays the optimal route information sent from the server on the navigation system screen. When the user is in a hurry, reliable routes are visually highlighted, allowing the user to select them with confidence.

[0826] Input on device: Optimal route information

[0827] Device output: Route display on navigation screen

[0828] In this way, the present invention provides a navigation system that takes into account the emotional state of the user.

[0829] 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.

[0830] 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.

[0831] 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.

[0832] [Third embodiment]

[0833] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0834] 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.

[0835] 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).

[0836] 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.

[0837] 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.

[0838] 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).

[0839] 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.

[0840] 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.

[0841] 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.

[0842] 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.

[0843] 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.

[0844] 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."

[0845] The present invention provides a system for evaluating the reliability of web pages using generative AI, allowing users to easily select reliable information from search results.

[0846] System configuration

[0847] The system consists of the following main components:

[0848] 1. How to enter search keywords

[0849] 2. How to search for related web pages

[0850] 3. Web page analysis using generative AI

[0851] 4. Terminology and Source Credibility Assessment Tools

[0852] 5. How the reliability score is calculated

[0853] 6. Means of integrating trustworthiness scores into search results

[0854] 7. Display of integrated results

[0855] Program processing

[0856] Program Overview

[0857] The system operates through the following processes:

[0858] 1. Enter search keywords

[0859] 2. Obtaining search results

[0860] 3. Web page analysis

[0861] 4. Reliability evaluation

[0862] 5. Integration of evaluation results

[0863] 6. Display

[0864] The system begins with the user entering a search keyword. The device then sends the keyword to the server, which searches for relevant web pages. The resulting search results are analyzed by a generative AI to evaluate the trustworthiness of each web page. A trustworthiness score is then calculated and integrated into the search results. Finally, the search results with the associated trustworthiness score are displayed on the user's device.

[0865] Detailed program description

[0866] Enter search keywords

[0867] The user enters a keyword (e.g., "cold treatment") on the search screen and presses the search button, which starts the search process.

[0868] Getting search results

[0869] The device sends the input keywords to the server, which then searches for related web pages using existing search algorithms and temporarily stores the results.

[0870] Web page analysis

[0871] The server sends the URL of each web page retrieved as a search result to the generation AI module, which downloads the HTML content of each web page and analyzes it, including extracting the key content and metadata within the page.

[0872] Terminology and source credibility assessment

[0873] The generative AI evaluates the use of technical terms within a webpage and the reliability of the source. Specifically, it checks whether technical terms are used accurately and evaluates whether the cited sources are trustworthy. For example, if the source is a public institution or an academic paper, it will assign a high reliability score.

[0874] Calculating the reliability score

[0875] The server calculates a credibility score for each web page based on the results evaluated by the AI, which is expressed as a number or letter so that users can determine the trustworthiness of each search result at a glance.

[0876] Integrating trust scores into search results

[0877] The server integrates the authority score into the search results by adding the authority score below the title of each web page.

[0878] Viewing the integration results

[0879] The terminal displays the integrated search results received from the server to the user, who can then review the displayed search results and select which web page to view based on the reliability score.

[0880] Specific examples

[0881] For example, when searching for the keyword "cold treatment," the user first enters the keyword and presses the search button. The device sends the keyword to the server, which retrieves approximately 10 related web pages. The server then analyzes each web page using generative AI to evaluate the use of technical terms and the reliability of cited sources. As a result of the evaluation, a reliability score is calculated, and the server integrates the reliability score into the search results. Finally, the device displays the search results with the reliability score assigned to them to the user. The user can check the reliability score of each web page and select the most reliable page to view.

[0882] As described above, the system of the present invention helps users to easily and accurately determine the reliability of web pages, and can particularly increase the reliability of pages that contain specialized information.

[0883] The processing flow will be explained below.

[0884] Step 1: Enter search keywords

[0885] 1. A user enters a keyword (e.g., "cold treatment") into the search field of a search engine.

[0886] 2. The user presses the "Search" button.

[0887] Step 2: Getting search results

[0888] 1. The device sends the entered search keywords to the server.

[0889] 2. The server searches for relevant web pages based on the keywords.

[0890] 3. The server generates a list of URLs for the web pages retrieved as search results.

[0891] 4. The server temporarily stores the URL list.

[0892] Step 3: Analyze the web page

[0893] 1. The server retrieves the URL of each web page from the URL list in turn.

[0894] 2. The server downloads the HTML content of each web page.

[0895] 3. The server sends the downloaded HTML content to the generation AI module.

[0896] 4. Generative AI analyzes the page structure and extracts key content and metadata.

[0897] Step 4: Assess terminology and source credibility

[0898] 1. Analyze the key content extracted by the generative AI and evaluate the use of technical terms.

[0899] 2. Generative AI analyzes citations and links within web pages and evaluates the source of the citation or link.

[0900] 3. If the generating AI determines that the source is based on a public institution or academic paper, it will assign a higher credibility score to that source.

[0901] Step 5: Calculate the reliability score

[0902] 1. The server calculates the trustworthiness score for each web page based on the content evaluated by the AI.

[0903] 2. The server temporarily stores the credibility score for each web page.

[0904] Step 6: Integrating Trust Scores into Search Results

[0905] 1. The server integrates the confidence score into the search results.

[0906] 2. The server creates a data structure that adds an authority score to each web page below its title.

[0907] Step 7: Viewing the integration results

[0908] 1. The server sends the consolidated search results to the device.

[0909] 2. The terminal displays the search results received to the user.

[0910] 3. Users can see the credibility score under the title of each search result displayed.

[0911] Example 1

[0912] 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."

[0913] The amount of information on the Internet is enormous, making it difficult for users to quickly find reliable information. This problem is particularly pronounced in fields where reliability is essential, such as specialized or medical information. Furthermore, existing search engines lack the functionality to evaluate the reliability of search results, leaving users with insufficient means to select appropriate information.

[0914] 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.

[0915] In this invention, the server includes means for inputting search keywords, means for searching for related web pages based on the input search keywords, means for activating a generation AI to analyze each web page obtained as a search result, means for evaluating the use of technical terms and the reliability of information sources using the generation AI, means for calculating a reliability score for each web page based on the evaluation results, means for integrating the reliability scores into search results, means for displaying the integrated search results, means for transmitting the search keywords to the server and temporarily saving the obtained search results, means for the generation AI to download the HTML content of the web page and extract key content and metadata within the page, and means for the user's terminal to display the integrated search results received from the server, thereby enabling users to quickly identify reliable information and browse appropriate information sources.

[0916] A "search keyword" is a character string that a user inputs to search for information.

[0917] The "input means" is a device or interface for a user to input search keywords.

[0918] A "search tool" is a function or process that searches the Internet for relevant web pages based on input search keywords.

[0919] "Generative AI" is a system that uses artificial intelligence technology to analyze data and perform evaluation and generation.

[0920] The "analysis means" is a function that analyzes the acquired web page in detail and extracts information.

[0921] "Terminology" is a term that has a specialized meaning in a particular field.

[0922] A "source" is the source or citation of information contained within a web page.

[0923] The "trustworthiness assessment tool" is a function that uses generative AI to evaluate the use of terminology within a web page and the reliability of the source of information.

[0924] The "trustworthiness score" is an evaluation result that indicates the trustworthiness of each web page numerically or alphabetically.

[0925] The "calculation means" is a function that calculates a reliability score based on the evaluation results of the generating AI.

[0926] The "search result integration means" is a function that incorporates reliability scores into search results.

[0927] The "display means" is a function that displays the integrated search results on the user's terminal.

[0928] A "server" is a computer system for processing, storing, and distributing data.

[0929] A "terminal" is an electronic device that a user operates, and typically refers to a PC or smartphone.

[0930] "HTML content" is data written in a markup language that makes up a web page.

[0931] "Metadata" is data that provides additional information about the content of a web page.

[0932] DETAILED DESCRIPTION OF THE INVENTION The present invention is a system that utilizes a generative AI model to evaluate the trustworthiness of web pages, allowing users to easily select reliable information from search results.

[0933] System configuration

[0934] The system consists of the following main components:

[0935] 1. How to enter search keywords

[0936] 2. A way to find related web pages

[0937] 3. A means to trigger a generative AI to analyze each web page retrieved as a search result.

[0938] 4. Using generative AI to assess terminology and source credibility

[0939] 5. A method for calculating the credibility score of each web page based on the evaluation results

[0940] 6. A way to integrate trustworthiness scores into search results

[0941] 7. A way to display consolidated search results

[0942] 8. A means of sending search keywords to a server and temporarily storing the retrieved search results

[0943] 9. How generative AI downloads the HTML content of a webpage and extracts key content and metadata from the page.

[0944] 10. Means for displaying the integrated search results received by the user's device from the server

[0945] Hardware and software used

[0946] Client terminal: PC or smartphone used by the user

[0947] Server: A server (e.g., a cloud computing service) to host the search engine and the generative AI.

[0948] Generative AI models: Natural language processing models such as GPT-4

[0949] Program processing overview

[0950] When a user enters a specific keyword (e.g., "cold treatment") on the search screen and presses the search button, the device sends the keyword to the server. The server uses an existing search engine to search for related web pages and temporarily stores the retrieved results. The server then passes the temporarily stored URL list of search results to the generation AI module, which downloads and analyzes the HTML content of each web page. Based on the analyzed text content, the generation AI evaluates the use of technical terms and the reliability of cited sources. The server then calculates a reliability score for each web page based on the generation AI's evaluation results and integrates the reliability scores into the search results. Finally, the device displays the integrated search results received from the server to the user, who then selects which web page to view based on the reliability score.

[0951] Specific examples

[0952] For example, let's take a specific example of searching for the keyword "cold treatment." When a user enters the keyword and presses the search button, the device sends the keyword to the server, which retrieves approximately 10 related web pages. The server then uses generative AI to analyze each web page and evaluates the use of technical terms and the reliability of cited sources. As a result of the evaluation, a reliability score is calculated, and the server integrates this score into the search results. Finally, the device displays the search results with the assigned reliability score to the user. The user can check the reliability score of each web page and select the most reliable page to view.

[0953] Example prompts for generative AI models

[0954] "Evaluate the titles and content of web pages and calculate a credibility score based on the following criteria: accurate use of technical terminology, credibility of cited sources (e.g., official institutions, academic papers, news outlets, etc.). For example, if you are evaluating web pages resulting from a search for the keyword 'cold treatment methods,' rate each page using the criteria above."

[0955] The above describes the embodiments of the present invention, which enable a user to quickly identify reliable information and browse appropriate information sources.

[0956] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0957] Step 1:

[0958] The user enters a specific keyword (e.g., "cold treatment") on the search screen and presses the search button. The input here is the action of the user entering text into the search box, and this text becomes the input data for the next step. The device receives the user's input and prepares to send that information to the server. In concrete terms, the user opens a search form in a browser, enters "cold treatment" into the search box, and clicks the search button.

[0959] Step 2:

[0960] The device sends the entered keywords to the server, which then uses the entered keywords for the next search process. The server uses a wide range of search engines to search for related web pages based on the received keywords and temporarily stores the results. Specifically, the device sends the user's input information to the server using a communication protocol (e.g., HTTP), and the server uses Google's search API or similar to obtain a list of URLs for related pages. The obtained list of URLs is temporarily stored in memory.

[0961] Step 3:

[0962] The server passes the temporarily saved list of search result URLs to the generation AI module. This becomes the input data. The generation AI downloads the HTML content of each web page and analyzes it as text. Specifically, the server passes the URL list to the generation AI, which then uses an HTTP request to retrieve the HTML of each web page, analyzes it, and extracts the text content and metadata.

[0963] Step 4:

[0964] Based on the text content analyzed by the generative AI, the usage of technical terms and the reliability of cited sources are evaluated. This serves as input data for the next evaluation step. Specifically, technical terms are extracted from the analyzed text and checked to see if they are used accurately. The reliability of cited sources, such as public institutions and academic papers, is also evaluated. Specifically, the generative AI extracts technical terms from the text and checks the cited sources.

[0965] Step 5:

[0966] The server calculates a reliability score for each web page based on the evaluation results of the generation AI. This becomes the input data for the evaluation score used in the next step. The reliability score serves as an indicator when users check the results. Specifically, the server receives the evaluation data from the generation AI, inputs it into the scoring algorithm, and calculates the reliability score. The calculated score is generated in a form corresponding to the URL of each web page.

[0967] Step 6:

[0968] The server integrates the calculated reliability score into the search results, generating integrated data. This integration result becomes the input data for the next display step. The reliability score is integrated by adding it below the title of each web page. Specifically, the server compares the list of search result URLs with the calculated reliability score, adds the reliability score to the metadata of each URL, and generates the final list of search results.

[0969] Step 7:

[0970] The terminal displays the integrated search results received from the server to the user. This is the final output data. The user can check the displayed search results and select which web page to view based on the reliability score. Specifically, the terminal receives the response from the server, analyzes the response data, and converts it into a format to be displayed in the browser. The user then clicks on each web page to view it based on the reliability score.

[0971] (Application example 1)

[0972] 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."

[0973] In today's world, it is extremely important for users to verify the reliability of information they search for on the Internet. Particularly on online shopping sites, the reliability of product descriptions and reviews significantly influences users' purchasing decisions. However, current search result displays do not provide a concrete means for evaluating reliability, leaving users to make their own judgments about the quality of information. To address this issue, the present invention aims to provide a system that automatically assigns reliability scores to search results on online shopping sites, allowing users to easily select highly reliable information.

[0974] 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.

[0975] In this invention, the server includes means for inputting search keywords, means for searching for related web pages based on the input search keywords, means for activating a generation AI for analyzing each web page obtained as a search result, means for evaluating the use of technical terms and the reliability of information sources using the generation AI, means for calculating a reliability score for each web page based on the evaluation results, means for integrating the reliability score into the search results, means for displaying the integrated search results, means for analyzing the content of each product information page and evaluating the quality of product descriptions and reviews, and means for assigning a reliability score to the evaluated product information page. This allows users to easily select reliable product information pages and make purchasing decisions with confidence.

[0976] "Search keywords" are words or phrases that users enter to describe the information they are looking for.

[0977] A "web page" is a document containing information that can be accessed on the Internet and is written in a format such as HTML.

[0978] "Generative AI" is a system that uses artificial intelligence technology to generate text and analyze information.

[0979] "Jargon" is a specialized word or phrase used in a particular field or area.

[0980] A "source" is a source or data source from which particular information is obtained.

[0981] A "credibility score" is a number or indicator used to evaluate the reliability of a particular piece of information or source of information.

[0982] "Content" refers to the information contained in a web page or document, such as text, images, or video.

[0983] A "review" is an evaluation or comment written by a user or purchaser describing their own experience or opinion.

[0984] "Search results" are the list of relevant web pages or information that appears when a user enters keywords into a search engine.

[0985] A "product information page" is a web page that contains detailed descriptions, specifications, and reviews of a particular product.

[0986] This invention relates to a system that evaluates the reliability of each product information page when a user searches for a product on an online shopping site, allowing the user to make a purchasing decision based on reliable information. This system is realized by using generative AI to evaluate the use of technical terms and the reliability of information sources, from the input of search keywords to the display of search results, and calculates a reliability score.

[0987] System configuration

[0988] The system consists of the following main components:

[0989] 1. How to enter search keywords

[0990] 2. How to search for related web pages

[0991] 3. Web page analysis using generative AI

[0992] 4. Terminology and Source Credibility Assessment Tools

[0993] 5. How the reliability score is calculated

[0994] 6. A way to integrate trustworthiness scores into search results

[0995] 7. Display of integrated results

[0996] 8. A way to analyze the content of each product page and evaluate the quality of the product descriptions and reviews.

[0997] 9. A means of assigning a credibility score to rated product information pages

[0998] Program processing

[0999] 1. Enter search keywords

[1000] The user enters the product name into the smartphone app and presses the search button.

[1001] 2. Obtaining search results

[1002] The device sends the entered search keywords to the server, which retrieves related product information pages. The search results are temporarily saved.

[1003] 3. Web page analysis

[1004] The server sends the URL of each product information page retrieved as a search result to the generation AI module, which downloads the HTML content of each web page and analyzes it, including extracting the key content and metadata within the page.

[1005] 4. Terminology and source credibility assessment

[1006] The generative AI evaluates the use of technical terms on product information pages and the reliability of the sources. Specifically, it checks whether technical terms are used accurately and evaluates the reliability of cited sources. For example, if the source is a public institution or academic paper, it will assign a high reliability score.

[1007] 5. Calculating the reliability score

[1008] The server calculates a reliability score for each product information page based on the results of the AI ​​generation. The reliability score is displayed as a number or letter, allowing users to determine the reliability of each product information page at a glance.

[1009] 6. Integrating Trust Scores into Search Results

[1010] The server integrates the reliability score into the search results by adding it below the title of each product information page.

[1011] 7. Displaying the integrated results

[1012] The terminal displays the integrated search results received from the server to the user, who can then review the displayed search results and select which information page to view based on the reliability score.

[1013] Hardware and software used

[1014] Hardware: Servers, smartphones

[1015] Software: Generative AI (GPT-4, etc.), Python, web frameworks (Django, Flask), databases (PostgreSQL)

[1016] Specific examples

[1017] For example, suppose a user enters "vitamin supplements" into a smartphone app and presses the search button. The device sends this search keyword to the server, which retrieves related product information pages. The server then uses generative AI to analyze each product information page and evaluates the use of technical terms and the reliability of the information source. A reliability score is calculated based on the evaluation results, and the server integrates the reliability score into the search results. The device then displays the search results with the reliability score to the user.

[1018] Prompt Sentence Examples

[1019] The prompt for the search "vitamin supplement" is:

[1020] Rate the trustworthiness of this product information page:

[1021] URL: https: / / example.com / vitamin-supplement

[1022] Key points: Accuracy of product description, quality of reviews, citation of reliable sources

[1023] The analysis results obtained from the generative AI model include numerical values ​​such as "content accuracy: 85%" and "source reliability: 90%." A reliability score is calculated based on this and displayed to the user. Users can then select products and make purchasing decisions with confidence based on the reliability score.

[1024] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1025] Step 1:

[1026] The user enters the product name into the smartphone app and presses the search button. The entered keywords are sent to the device as "search keywords." The entered data is the product name specified by the user, and this becomes the base data for subsequent processing.

[1027] Step 2:

[1028] The server receives the "search keywords" sent from the terminal. The server uses existing search engine functions to search for the relevant product information page. At this point, the only input is the "search keywords," and the output is a list of URLs for related product information pages.

[1029] Step 3:

[1030] The server sends a list of URLs for each product information page obtained as a search result to the generation AI module. The generation AI downloads the HTML content of each product and analyzes it. The analysis includes extracting the main content and metadata within the page. The input data is the list of URLs for the product information pages, and the output data is the analyzed HTML content.

[1031] Step 4:

[1032] The generative AI evaluates the use of technical terms on product information pages and the reliability of the sources. Specifically, it checks whether technical terms are used accurately and evaluates whether the cited sources are trustworthy. If highly reliable sources such as public institutions or academic papers are cited, the rating is higher. The input data is HTML content, and the output data is a reliability rating score.

[1033] Step 5:

[1034] The server calculates the reliability score of each product information page based on the results of the evaluation by the generation AI. The reliability score is calculated based on a specific algorithm and is displayed as a number or letter that allows users to determine the reliability of each product information page at a glance. The input data is the reliability evaluation score of each product information page, and the output data is the reliability score.

[1035] Step 6:

[1036] The server integrates the reliability scores into the search results. Specifically, it adds the reliability score below the title of each product information page. This process visually displays the reliability rating for all search results. The input data is the reliability score and a list of search results, and the output data is a list of search results with the reliability scores added.

[1037] Step 7:

[1038] The terminal displays the integrated search results with the reliability scores received from the server to the user. The user checks the displayed search results and selects which product information page to view based on the reliability scores. The input data are the search results with the reliability scores added, and the output data are the product information pages displayed to the user.

[1039] Through the above steps, the user can easily select highly reliable product information and make a purchase decision with confidence.

[1040] 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.

[1041] The present invention provides a system for providing more personalized search results by combining a conventional system that uses generative AI to evaluate the trustworthiness of web pages with an emotion engine that recognizes user emotions.

[1042] System configuration

[1043] The system consists of the following main components:

[1044] 1. How to enter search keywords

[1045] 2. How to search for related web pages

[1046] 3. Web page analysis using generative AI

[1047] 4. Terminology and Source Credibility Assessment Tools

[1048] 5. How the reliability score is calculated

[1049] 6. Means of integrating trustworthiness scores into search results

[1050] 7. Display of integrated results

[1051] 8. Emotion engine that recognizes user emotions

[1052] 9. Sentiment-Based Search Results Customization

[1053] Program processing

[1054] The system proceeds as follows:

[1055] Enter search keywords

[1056] A user enters a keyword (e.g., "cold treatment") into the search field of a search engine and presses the search button. This starts the search process, but at this point the emotion engine is also activated. The emotion engine analyzes the user's input method and speed, facial expressions, tone of voice, etc., to recognize the user's emotional state (e.g., impatience, anxiety, relaxation, etc.).

[1057] Getting search results

[1058] The device sends the emotion data recognized by the emotion engine along with the entered search keywords to the server, where the server searches for related web pages using existing search algorithms and temporarily stores the results.

[1059] Web page analysis

[1060] The server sends the URL of each web page retrieved as a search result to the generation AI module, which downloads the HTML content of each web page and analyzes it, including extracting the key content and metadata within the page.

[1061] Terminology and source credibility assessment

[1062] The generative AI evaluates the use of domain-specific terminology and determines the trustworthiness of cited sources. For example, it assigns a higher credibility score to sources from public institutions or academic papers. Furthermore, it can prioritize sources that inspire a sense of security based on the user's emotional state.

[1063] Calculating the reliability score

[1064] The server calculates a credibility score for each web page based on the evaluation results of the generation AI. In addition, it takes into account data from the emotion engine and adjusts the score according to the user's current emotional state. For example, if a user is feeling anxious, it assigns a higher score to information that is highly reliable and reassuring.

[1065] Integrating trust scores into search results

[1066] The server integrates the reliability score into the search results, and adjusts the display of search results based on the user's emotional state. For example, if a user is in a hurry, the server highlights the results so that the reliability can be seen at a glance.

[1067] Viewing the integration results

[1068] The server sends the integrated search results to the device, which then displays the results in a format that takes the user's emotions into consideration.The user can then review the displayed search results and select the most appropriate web page to view, with the reliability score and emotional engine having been adjusted accordingly.

[1069] Specific examples

[1070] For example, if the emotion engine recognizes that a user searching for the keyword "cold treatment" is feeling anxious or impatient when typing, the emotion engine sends this information to the server. The server retrieves search results based on that information, and the generation AI analyzes the webpage. It then evaluates the reliability of the terminology and information source, and calculates a reliability score taking into account the emotional data. Finally, when the search results incorporating the reliability score are sent to the user's device, reliable information that will reassure users who are feeling anxious is highlighted.

[1071] This allows users to easily access the most appropriate information based on their emotional state at the time, providing a more personalized search experience.

[1072] The processing flow will be explained below.

[1073] Step 1: Enter search keywords

[1074] 1. A user enters a keyword (e.g., "cold treatment") into the search field of a search engine.

[1075] 2. The user presses the "Search" button.

[1076] 3. The device monitors the user's input method (speed, rhythm, etc.), facial recognition camera, and tone of voice during voice input, and activates the emotion engine.

[1077] 4. The emotion engine analyzes the user's emotional state (e.g., impatience, anxiety, relaxation, etc.) and generates emotion data.

[1078] Step 2: Getting search results

[1079] 1. The device sends the entered search keywords and the emotion data recognized by the emotion engine to the server.

[1080] 2. The server searches for relevant web pages based on the keywords.

[1081] 3. The server generates a list of URLs of the web pages obtained as search results and temporarily stores them.

[1082] Step 3: Analyze the web page

[1083] 1. The server retrieves the URL of each web page from the URL list in turn.

[1084] 2. The server downloads the HTML content of each web page.

[1085] 3. The server sends the downloaded HTML content to the generation AI module.

[1086] 4. Generative AI analyzes the page structure and extracts key content and metadata.

[1087] Step 4: Assess terminology and source credibility

[1088] 1. Analyze the key content extracted by the generative AI and evaluate the use of technical terms.

[1089] 2. Generative AI analyzes citations and links within web pages and evaluates the source of the citation or link.

[1090] 3. If the generating AI determines that the source is based on a public institution or academic paper, it will assign a higher credibility score to that source.

[1091] 4. The generative AI uses data from the emotion engine to adjust the importance of information according to the user's emotional state. For example, if a user is feeling anxious, information that gives a sense of security will be given a high score.

[1092] Step 5: Calculate the reliability score

[1093] 1. The server calculates the trustworthiness score for each web page based on the content evaluated by the AI.

[1094] 2. The server also takes into account data from the emotion engine and adjusts the reliability score according to the user's emotional state.

[1095] 3. The server generates and temporarily stores a final credibility score for each web page.

[1096] Step 6: Integrating Trust Scores into Search Results

[1097] 1. The server integrates the confidence score into the search results.

[1098] 2. The server creates a data structure that adds an authority score to each web page below its title.

[1099] 3. The emotion engine determines how search results are displayed (e.g., highlighted, color-coded) depending on the user's emotional state.

[1100] Step 7: Viewing the integration results

[1101] 1. The server sends the consolidated search results to the device.

[1102] 2. The search results received by the device are displayed in a format that takes into account the user's emotional state.

[1103] 3. Users can see the credibility score under the title of each search result displayed and view customized information based on their emotional state.

[1104] Through the above steps, the system of the present invention provides personalized search results that adapt to the user's emotional state to enhance the user's search experience.

[1105] Example 2

[1106] 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."

[1107] Conventional search systems present relevant information based on keywords entered by the user, but because they do not take the user's emotional state into consideration, it is difficult to provide the optimal information the user is looking for. As a result, even if the user is feeling anxious or impatient, information that provides a sense of security is not prioritized, resulting in a poor user experience.

[1108] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for inputting a search keyword; means for searching for related web pages based on the input search keyword and user emotional data; means for activating an emotional engine and analyzing the user's emotional state; means for activating a generation AI for analyzing each web page obtained as a search result; means for evaluating the use of technical terms and the reliability of information sources using the generation AI; means for calculating a reliability score for each web page taking into account the user's emotional data; means for integrating the reliability score into the search result and adjusting the display method based on the user's emotional state; and means for displaying the integrated search result. This makes it possible to provide personalized search results that take into account the user's emotional state.

[1109] The "means for inputting search keywords" refers to a device or function that allows a user to input search keywords through the interface of a search engine.

[1110] "Means for searching for related web pages based on input search keywords and user emotion data" refers to a device or function for finding related web pages using search keywords input by the user and emotion data analyzed by the emotion engine.

[1111] "Means for activating an emotion engine and analyzing the user's emotional state" refers to a device or function for recognizing the user's emotions (e.g., impatience, anxiety, relaxation) by analyzing the user's input method, speed, facial expression, tone of voice, etc.

[1112] "Means for launching a generative AI to analyze each web page obtained as a search result" refers to a device or function that launches a generative AI model to analyze the content of each web page obtained as a search result.

[1113] "Means for using generative AI to evaluate terminology usage and source reliability" refers to a device or function that uses generative AI to analyze and evaluate the use of terminology within a web page and the reliability of cited sources.

[1114] The "means for calculating the reliability score of each web page taking into account the user's emotional data" refers to a device or function for calculating the reliability score of each web page taking into account the evaluation results of the generation AI and the user's emotional state.

[1115] "Means for integrating confidence scores into search results and adjusting the display based on the user's emotional state" refers to a device or function for incorporating calculated confidence scores into search results and changing the display of search results depending on the user's emotional state.

[1116] The "means for displaying integrated search results" is a device or function for displaying search results including reliability scores on a user's terminal so that the user can view them.

[1117] The following describes in detail the mode for carrying out the present invention: The system utilizes generative AI to evaluate the trustworthiness of web pages, and combines it with an emotion engine that recognizes user emotions to provide more personalized search results.

[1118] System configuration

[1119] The system consists of the following main elements:

[1120] 1. How to enter search keywords

[1121] 2. How to search for related web pages

[1122] 3. Web page analysis using generative AI

[1123] 4. Terminology and Source Credibility Assessment Tools

[1124] 5. How the reliability score is calculated

[1125] 6. Means of integrating trustworthiness scores into search results

[1126] 7. Display of integrated results

[1127] 8. Emotion engine that recognizes user emotions

[1128] 9. Sentiment-Based Search Results Customization

[1129] Enter search keywords

[1130] A user enters a keyword (e.g., "cold treatment") into the search field of a search engine and presses the search button. This action starts the search process and simultaneously activates the emotion engine. The emotion engine analyzes the user's input method, speed, facial expression, tone of voice, etc., to recognize the user's emotional state (e.g., impatience, anxiety, relaxation, etc.).

[1131] Getting search results

[1132] The device sends the entered search keywords and the emotion data recognized by the emotion engine to the server, which then uses existing search algorithms to search for related web pages and temporarily stores the results.

[1133] Web page analysis

[1134] The server sends the URL of each web page retrieved as a search result to the generative AI module. The generative AI downloads the HTML content of each web page, extracts key content and metadata, and analyzes it. The generative AI model can be implemented using Python libraries such as "BeautifulSoup" and "Selenium."

[1135] Terminology and source credibility assessment

[1136] Based on the analyzed content, the generative AI evaluates the use of technical terms and the reliability of cited sources. For example, if the source is a public institution or academic paper, it will give a high reliability score. It can also take into account the user's emotional state and prioritize sources that give a sense of security.

[1137] Calculating the reliability score

[1138] The server calculates a reliability score for each web page based on the evaluation results of the generated AI and data from the emotion engine. For example, if a user is feeling anxious, a high score can be assigned to information that is highly reliable and reassuring.

[1139] Integrating trust scores into search results

[1140] The server integrates the reliability score into search results, adjusting the presentation based on the user's emotional state—for example, highlighting results to make their reliability clear at a glance to users in a hurry.

[1141] Viewing the integration results

[1142] The server sends the integrated search results to the device, which then displays the search results in a format that takes the user's emotions into consideration.The user can review the displayed search results and select the most appropriate web page to view, with the results adjusted by the reliability score and emotion engine.

[1143] Specific examples

[1144] For example, if the emotion engine recognizes that a user searching for the keyword "cold treatment" is feeling anxious or impatient when typing, the emotion engine sends this emotional data to the server. The server retrieves search results based on that information, and the generation AI analyzes the webpage. The analysis results then evaluate the reliability of the terminology and information sources, and calculate a reliability score taking the emotional data into account. Finally, the search results, incorporating the reliability score, are sent to the user's device, and reliable information that will reassure anxious users is highlighted and displayed.

[1145] Example prompt

[1146] If the user enters the keyword "cold treatment" and the emotion engine recognizes that the user is feeling anxious, prioritize displaying reliable information that provides a sense of security.

[1147] This system allows users to easily obtain the most appropriate information based on their emotional state at the time, providing a more personalized search experience.

[1148] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1149] Step 1:

[1150] The user enters a search keyword. Specifically, the user enters "cold treatment" in the search field of the search engine and presses the search button. This input triggers the search process, obtaining "cold treatment" as the input and proceeding to the next step.

[1151] Step 2:

[1152] The emotion engine is activated. It collects and analyzes information such as the user's input method and speed, facial expressions, and tone of voice. For example, if the user's input speed is fast and they are putting a lot of force into the keys, it is likely that they are impatient. The emotion engine uses this data to recognize the user's emotional state (e.g., impatience, anxiety, relaxation). The user's behavioral data is taken as input, and the emotion analysis results are output.

[1153] Step 3:

[1154] The device sends the search keywords and emotion data to the server. Here, the device sends the keyword "cold treatment" and the emotion data (e.g., "impatience") analyzed by the emotion engine to the server. The search keywords and emotion data are sent to the server as input, and the next processing step is executed based on this.

[1155] Step 4:

[1156] The server searches for relevant web pages. Based on the received search keyword "cold treatment," the server uses an existing search algorithm to search for relevant web pages. At this time, emotion data is temporarily stored. The input is the search keyword, and the output is a list of relevant web pages.

[1157] Step 5:

[1158] The server sends the URLs of the web pages to the generation AI. The server then sends the URLs of each web page obtained as a search result to the generation AI module. The list of URLs obtained as input is sent, and analysis begins in the next step.

[1159] Step 6:

[1160] The generative AI analyzes the content of web pages. It downloads the HTML content of each web page and extracts key content and metadata. This analysis is performed using tools such as "BeautifulSoup" and "Selenium." The input is the HTML content, and the output is the parsed data.

[1161] Step 7:

[1162] Generative AI evaluates the reliability of terminology and sources. Based on the analyzed data, generative AI determines the use of terminology and the reliability of cited sources. For example, if the source is a public institution or academic paper, it will be assigned a high reliability score. The analyzed data is input, and a reliability evaluation score is generated as output.

[1163] Step 8:

[1164] The server calculates the reliability score. The server calculates the reliability score for each web page based on the evaluation results of the generation AI and data from the emotion engine. For example, if a user is feeling anxious, a high score is assigned to information that is highly reliable and gives a sense of security. The inputs are the evaluation results and emotion data, and the output is a reliability score.

[1165] Step 9:

[1166] The server integrates the reliability score into the search results. The calculated reliability score is incorporated into the search results and the display method is adjusted based on the user's emotional state. For example, for a user in a hurry, results are highlighted so that the reliability can be seen at a glance. The input is the reliability score and the search results, and the output is the adjusted search results.

[1167] Step 10:

[1168] The server sends the integrated results to the terminal. The server sends the integrated search results to the terminal, where they can be viewed by the user. The integrated search results are taken as input and sent to the terminal as output.

[1169] Step 11:

[1170] The device displays the search results. The device displays the search results in a format that takes the user's emotions into consideration. The user can review the displayed search results and select and view the optimal web page, adjusted by the reliability score and emotion engine. The input is the integrated results, and the output is the displayed search results.

[1171] (Application example 2)

[1172] 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."

[1173] Conventional navigation systems suggest routes without taking the user's emotional state into consideration, and therefore are unable to provide the optimal route, especially when the user is feeling anxious or impatient. Furthermore, they lack the reliability of search results and the ability to evaluate safety in real time, creating a need for a navigation system that users can use with peace of mind.

[1174] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for inputting search keywords, means for searching for related information based on the input search keywords, means for activating a generation AI for analyzing each piece of information obtained as a search result, means for evaluating the use of technical terms and the reliability of information sources using the generation AI, means for calculating a reliability score for each piece of information based on the evaluation results, means for integrating the reliability score into the search results, means for displaying the integrated search results, means for detecting the user's emotions using an emotion recognition device, and means for customizing the search results based on the user's emotion data. This enables personalized route selection according to the user's emotional state, thereby providing safe and reliable navigation.

[1175] A "search keyword" is a word or phrase that a user enters when searching for information.

[1176] "Search results" are a collection of related information obtained based on search keywords.

[1177] "Generative AI" is a system or program that uses artificial intelligence technology to analyze and evaluate information.

[1178] "Jargon" is specialized words and phrases used in a particular field.

[1179] "Source credibility" is a criterion for assessing the reliability of the source or source from which information is provided.

[1180] The "trustworthiness score" is a numerical representation of the reliability of information or sources evaluated by the generating AI.

[1181] An "emotion recognition device" is a device that uses a camera, a voice recognition system, etc. to detect a user's emotions in real time.

[1182] "Emotion data" is information that represents the emotional state of a user detected by an emotion recognition device.

[1183] A "means for customizing search results" is a method or apparatus that tailors and personalizes the search results obtained based on the user's emotional data.

[1184] The embodiment of the present invention will be described as a navigation system that operates within an autonomous vehicle. This system recognizes the user's emotions and uses generative AI to evaluate and provide road information and route reliability in real time. The following elements are required to implement this system:

[1185] System configuration

[1186] Enter search keywords

[1187] When a user inputs a destination into a navigation system, the system also recognizes the user's emotions.

[1188] User Emotion Recognition

[1189] The device is equipped with a camera and a voice recognition system that analyzes the user's facial expressions and voice in real time. This function uses a camera (e.g., Logitech C920) and a voice recognition system (e.g., Google Cloud Speech-to-Text). The recognized emotion data is sent to a server.

[1190] Find a route

[1191] The server searches for available routes based on the input destination, using traffic information services such as Google Maps API.

[1192] Analysis by generative AI

[1193] The route information obtained as a search result is analyzed by a generation AI on the server. The generation AI evaluates the latest road conditions and traffic information and quantifies the reliability of each route. This generation AI module uses models such as GPT-4.

[1194] Calculating the reliability score

[1195] Based on the data evaluated by the generative AI, a reliability score for each route is calculated using Scikit-learn's MLPClassifier or a custom function, and data from the emotion engine is also taken into account to adjust the score according to specific emotional states.

[1196] Customize route display

[1197] Based on the reliability score, the system displays the optimal route based on the user's emotional state on the device's navigation system screen. For example, it highlights safe and reliable routes for a user who is feeling anxious.

[1198] Specific examples

[1199] For example, if a user inputs the "safest route" into the navigation system of an autonomous vehicle, the system may recognize from the user's facial expressions and voice that they are feeling "anxiety" or "impatience." The emotion data and the searched route information are sent together to the server, and the generation AI analyzes these routes.

[1200] The generation AI calculates a reliability score for each route and uses prompts such as:

[1201] "If you want to avoid routes where there are traffic accidents, please provide a reliable route taking into account the current road conditions."

[1202] Based on this, the generation AI evaluates the most reliable route and displays a customized version that reflects the emotional data. Ultimately, route information that gives a sense of security to a user who is feeling anxious is displayed on the device, and the optimal route is suggested for the user.

[1203] In this way, the present invention provides a personalized navigation experience that responds to the user's emotional state.

[1204] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1205] Step 1:

[1206] The user inputs a destination into the navigation system inside the autonomous vehicle. The user's input is made via keyboard or voice. At this time, the camera and voice recognition system capture the user's facial expressions and voice in real time, and emotional data is acquired. The input destination data and emotional data are then sent to the terminal.

[1207] Step 2:

[1208] The device sends the acquired destination data and emotion data to the server. The server uses the destination data to search for multiple route information using the Google Maps API. The route information includes the distance, travel time, current traffic conditions, etc. for each route.

[1209] Server input: Destination data, emotion data

[1210] Server output: Route information

[1211] Step 3:

[1212] The server sends each route to the AI ​​generator, which analyzes the route and evaluates road conditions (traffic, accidents, construction, etc.) and other important factors. A generative AI model such as GPT-4 is used for the analysis. The evaluation results are quantified as a reliability score.

[1213] Input for the generation AI: Route information, prompt (e.g., "If I want to avoid a route where a traffic accident has occurred, please provide a reliable route taking into account the current road conditions.")

[1214] Generative AI output: Confidence score

[1215] Step 4:

[1216] The server adjusts the reliability score obtained by the AI ​​generator by taking into account emotional data. For example, if the user is in a hurry, it will give a higher score to safer and more reliable routes.

[1217] Server input: confidence score, sentiment data

[1218] Server output: Adjusted reliability score

[1219] Step 5:

[1220] The server reevaluates the reliability of each route based on the adjusted reliability score, selects the route that best suits the user's emotional state, and sends the selection result to the device.

[1221] Server Input: Adjusted Reliability Score

[1222] Server output: Optimal route information

[1223] Step 6:

[1224] The device displays the optimal route information sent from the server on the navigation system screen. When the user is in a hurry, reliable routes are visually highlighted, allowing the user to select them with confidence.

[1225] Input on device: Optimal route information

[1226] Device output: Route display on navigation screen

[1227] In this way, the present invention provides a navigation system that takes into account the emotional state of the user.

[1228] 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.

[1229] 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.

[1230] 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.

[1231] [Fourth embodiment]

[1232] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1233] 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.

[1234] 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).

[1235] 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.

[1236] 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.

[1237] 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).

[1238] 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.

[1239] 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.

[1240] 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.

[1241] 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.

[1242] 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.

[1243] 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.

[1244] 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."

[1245] The present invention provides a system for evaluating the reliability of web pages using generative AI, allowing users to easily select reliable information from search results.

[1246] System configuration

[1247] The system consists of the following main components:

[1248] 1. How to enter search keywords

[1249] 2. How to search for related web pages

[1250] 3. Web page analysis using generative AI

[1251] 4. Terminology and Source Credibility Assessment Tools

[1252] 5. How the reliability score is calculated

[1253] 6. Means of integrating trustworthiness scores into search results

[1254] 7. Display of integrated results

[1255] Program processing

[1256] Program Overview

[1257] The system operates through the following processes:

[1258] 1. Enter search keywords

[1259] 2. Obtaining search results

[1260] 3. Web page analysis

[1261] 4. Reliability evaluation

[1262] 5. Integration of evaluation results

[1263] 6. Display

[1264] The system begins with the user entering a search keyword. The device then sends the keyword to the server, which searches for relevant web pages. The resulting search results are analyzed by a generative AI to evaluate the trustworthiness of each web page. A trustworthiness score is then calculated and integrated into the search results. Finally, the search results with the associated trustworthiness score are displayed on the user's device.

[1265] Detailed program description

[1266] Enter search keywords

[1267] The user enters a keyword (e.g., "cold treatment") on the search screen and presses the search button, which starts the search process.

[1268] Getting search results

[1269] The device sends the input keywords to the server, which then searches for related web pages using existing search algorithms and temporarily stores the results.

[1270] Web page analysis

[1271] The server sends the URL of each web page retrieved as a search result to the generation AI module, which downloads the HTML content of each web page and analyzes it, including extracting the key content and metadata within the page.

[1272] Terminology and source credibility assessment

[1273] The generative AI evaluates the use of technical terms within a webpage and the reliability of the source. Specifically, it checks whether technical terms are used accurately and evaluates whether the cited sources are trustworthy. For example, if the source is a public institution or an academic paper, it will assign a high reliability score.

[1274] Calculating the reliability score

[1275] The server calculates a credibility score for each web page based on the results evaluated by the AI, which is expressed as a number or letter so that users can determine the trustworthiness of each search result at a glance.

[1276] Integrating trust scores into search results

[1277] The server integrates the authority score into the search results by adding the authority score below the title of each web page.

[1278] Viewing the integration results

[1279] The terminal displays the integrated search results received from the server to the user, who can then review the displayed search results and select which web page to view based on the reliability score.

[1280] Specific examples

[1281] For example, when searching for the keyword "cold treatment," the user first enters the keyword and presses the search button. The device sends the keyword to the server, which retrieves approximately 10 related web pages. The server then analyzes each web page using generative AI to evaluate the use of technical terms and the reliability of cited sources. As a result of the evaluation, a reliability score is calculated, and the server integrates the reliability score into the search results. Finally, the device displays the search results with the reliability score assigned to them to the user. The user can check the reliability score of each web page and select the most reliable page to view.

[1282] As described above, the system of the present invention helps users to easily and accurately determine the reliability of web pages, and can particularly increase the reliability of pages that contain specialized information.

[1283] The processing flow will be explained below.

[1284] Step 1: Enter search keywords

[1285] 1. A user enters a keyword (e.g., "cold treatment") into the search field of a search engine.

[1286] 2. The user presses the "Search" button.

[1287] Step 2: Getting search results

[1288] 1. The device sends the entered search keywords to the server.

[1289] 2. The server searches for relevant web pages based on the keywords.

[1290] 3. The server generates a list of URLs for the web pages retrieved as search results.

[1291] 4. The server temporarily stores the URL list.

[1292] Step 3: Analyze the web page

[1293] 1. The server retrieves the URL of each web page from the URL list in turn.

[1294] 2. The server downloads the HTML content of each web page.

[1295] 3. The server sends the downloaded HTML content to the generation AI module.

[1296] 4. Generative AI analyzes the page structure and extracts key content and metadata.

[1297] Step 4: Assess terminology and source credibility

[1298] 1. Analyze the key content extracted by the generative AI and evaluate the use of technical terms.

[1299] 2. Generative AI analyzes citations and links within web pages and evaluates the source of the citation or link.

[1300] 3. If the generating AI determines that the source is based on a public institution or academic paper, it will assign a higher credibility score to that source.

[1301] Step 5: Calculate the reliability score

[1302] 1. The server calculates the trustworthiness score for each web page based on the content evaluated by the AI.

[1303] 2. The server temporarily stores the credibility score for each web page.

[1304] Step 6: Integrating Trust Scores into Search Results

[1305] 1. The server integrates the confidence score into the search results.

[1306] 2. The server creates a data structure that adds an authority score to each web page below its title.

[1307] Step 7: Viewing the integration results

[1308] 1. The server sends the consolidated search results to the device.

[1309] 2. The terminal displays the search results received to the user.

[1310] 3. Users can see the credibility score under the title of each search result displayed.

[1311] Example 1

[1312] 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."

[1313] The amount of information on the Internet is enormous, making it difficult for users to quickly find reliable information. This problem is particularly pronounced in fields where reliability is essential, such as specialized or medical information. Furthermore, existing search engines lack the functionality to evaluate the reliability of search results, leaving users with insufficient means to select appropriate information.

[1314] 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.

[1315] In this invention, the server includes means for inputting search keywords, means for searching for related web pages based on the input search keywords, means for activating a generation AI to analyze each web page obtained as a search result, means for evaluating the use of technical terms and the reliability of information sources using the generation AI, means for calculating a reliability score for each web page based on the evaluation results, means for integrating the reliability scores into search results, means for displaying the integrated search results, means for transmitting the search keywords to the server and temporarily saving the obtained search results, means for the generation AI to download the HTML content of the web page and extract key content and metadata within the page, and means for the user's terminal to display the integrated search results received from the server, thereby enabling users to quickly identify reliable information and browse appropriate information sources.

[1316] A "search keyword" is a character string that a user inputs to search for information.

[1317] The "input means" is a device or interface for a user to input search keywords.

[1318] A "search tool" is a function or process that searches the Internet for relevant web pages based on input search keywords.

[1319] "Generative AI" is a system that uses artificial intelligence technology to analyze data and perform evaluation and generation.

[1320] The "analysis means" is a function that analyzes the acquired web page in detail and extracts information.

[1321] "Terminology" is a term that has a specialized meaning in a particular field.

[1322] A "source" is the source or citation of information contained within a web page.

[1323] The "trustworthiness assessment tool" is a function that uses generative AI to evaluate the use of terminology within a web page and the reliability of the source of information.

[1324] The "trustworthiness score" is an evaluation result that indicates the trustworthiness of each web page numerically or alphabetically.

[1325] The "calculation means" is a function that calculates a reliability score based on the evaluation results of the generating AI.

[1326] The "search result integration means" is a function that incorporates reliability scores into search results.

[1327] The "display means" is a function that displays the integrated search results on the user's terminal.

[1328] A "server" is a computer system for processing, storing, and distributing data.

[1329] A "terminal" is an electronic device that a user operates, and typically refers to a PC or smartphone.

[1330] "HTML content" is data written in a markup language that makes up a web page.

[1331] "Metadata" is data that provides additional information about the content of a web page.

[1332] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention is a system that utilizes a generative AI model to evaluate the trustworthiness of web pages, allowing users to easily select reliable information from search results.

[1333] System configuration

[1334] The system consists of the following main components:

[1335] 1. How to enter search keywords

[1336] 2. A way to find related web pages

[1337] 3. A means to trigger a generative AI to analyze each web page retrieved as a search result.

[1338] 4. Using generative AI to assess terminology and source credibility

[1339] 5. A method for calculating the credibility score of each web page based on the evaluation results

[1340] 6. A way to integrate trustworthiness scores into search results

[1341] 7. A way to display consolidated search results

[1342] 8. A means of sending search keywords to a server and temporarily storing the retrieved search results

[1343] 9. How generative AI downloads the HTML content of a webpage and extracts key content and metadata from the page.

[1344] 10. Means for displaying the integrated search results received by the user's device from the server

[1345] Hardware and software used

[1346] Client terminal: PC or smartphone used by the user

[1347] Server: A server (e.g., a cloud computing service) to host the search engine and the generative AI.

[1348] Generative AI models: Natural language processing models such as GPT-4

[1349] Program processing overview

[1350] When a user enters a specific keyword (e.g., "cold treatment") on the search screen and presses the search button, the device sends the keyword to the server. The server uses an existing search engine to search for related web pages and temporarily stores the retrieved results. The server then passes the temporarily stored URL list of search results to the generation AI module, which downloads and analyzes the HTML content of each web page. Based on the analyzed text content, the generation AI evaluates the use of technical terms and the reliability of cited sources. The server then calculates a reliability score for each web page based on the generation AI's evaluation results and integrates the reliability scores into the search results. Finally, the device displays the integrated search results received from the server to the user, who then selects which web page to view based on the reliability score.

[1351] Specific examples

[1352] For example, let's take a specific example of searching for the keyword "cold treatment." When a user enters the keyword and presses the search button, the device sends the keyword to the server, which retrieves approximately 10 related web pages. The server then uses generative AI to analyze each web page and evaluates the use of technical terms and the reliability of cited sources. As a result of the evaluation, a reliability score is calculated, and the server integrates this score into the search results. Finally, the device displays the search results with the assigned reliability score to the user. The user can check the reliability score of each web page and select the most reliable page to view.

[1353] Example prompts for generative AI models

[1354] "Evaluate the titles and content of web pages and calculate a credibility score based on the following criteria: accurate use of technical terminology, credibility of cited sources (e.g., official institutions, academic papers, news outlets, etc.). For example, if you are evaluating web pages resulting from a search for the keyword 'cold treatment methods,' rate each page using the criteria above."

[1355] The above describes the embodiments of the present invention, which enable a user to quickly identify reliable information and browse appropriate information sources.

[1356] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1357] Step 1:

[1358] The user enters a specific keyword (e.g., "cold treatment") on the search screen and presses the search button. The input here is the action of the user entering text into the search box, and this text becomes the input data for the next step. The device receives the user's input and prepares to send that information to the server. In concrete terms, the user opens a search form in a browser, enters "cold treatment" into the search box, and clicks the search button.

[1359] Step 2:

[1360] The device sends the entered keywords to the server, which then uses the entered keywords for the next search process. The server uses a wide range of search engines to search for related web pages based on the received keywords and temporarily stores the results. Specifically, the device sends the user's input information to the server using a communication protocol (e.g., HTTP), and the server uses Google's search API or similar to obtain a list of URLs for related pages. The obtained list of URLs is temporarily stored in memory.

[1361] Step 3:

[1362] The server passes the temporarily saved list of search result URLs to the generation AI module. This becomes the input data. The generation AI downloads the HTML content of each web page and analyzes it as text. Specifically, the server passes the URL list to the generation AI, which then uses an HTTP request to retrieve the HTML of each web page, analyzes it, and extracts the text content and metadata.

[1363] Step 4:

[1364] Based on the text content analyzed by the generative AI, the usage of technical terms and the reliability of cited sources are evaluated. This serves as input data for the next evaluation step. Specifically, technical terms are extracted from the analyzed text and checked to see if they are used accurately. The reliability of cited sources, such as public institutions and academic papers, is also evaluated. Specifically, the generative AI extracts technical terms from the text and checks the cited sources.

[1365] Step 5:

[1366] The server calculates a reliability score for each web page based on the evaluation results of the generation AI. This becomes the input data for the evaluation score used in the next step. The reliability score serves as an indicator when users check the results. Specifically, the server receives the evaluation data from the generation AI, inputs it into the scoring algorithm, and calculates the reliability score. The calculated score is generated in a form corresponding to the URL of each web page.

[1367] Step 6:

[1368] The server integrates the calculated reliability score into the search results, generating integrated data. This integration result becomes the input data for the next display step. The reliability score is integrated by adding it below the title of each web page. Specifically, the server compares the list of search result URLs with the calculated reliability score, adds the reliability score to the metadata of each URL, and generates the final list of search results.

[1369] Step 7:

[1370] The terminal displays the integrated search results received from the server to the user. This is the final output data. The user can check the displayed search results and select which web page to view based on the reliability score. Specifically, the terminal receives the response from the server, analyzes the response data, and converts it into a format to be displayed in the browser. The user then clicks on each web page to view it based on the reliability score.

[1371] (Application example 1)

[1372] 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."

[1373] In today's world, it is extremely important for users to verify the reliability of information they search for on the Internet. Particularly on online shopping sites, the reliability of product descriptions and reviews significantly influences users' purchasing decisions. However, current search result displays do not provide a concrete means for evaluating reliability, leaving users to make their own judgments about the quality of information. To address this issue, the present invention aims to provide a system that automatically assigns reliability scores to search results on online shopping sites, allowing users to easily select highly reliable information.

[1374] 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.

[1375] In this invention, the server includes means for inputting search keywords, means for searching for related web pages based on the input search keywords, means for activating a generation AI for analyzing each web page obtained as a search result, means for evaluating the use of technical terms and the reliability of information sources using the generation AI, means for calculating a reliability score for each web page based on the evaluation results, means for integrating the reliability score into the search results, means for displaying the integrated search results, means for analyzing the content of each product information page and evaluating the quality of product descriptions and reviews, and means for assigning a reliability score to the evaluated product information page. This allows users to easily select reliable product information pages and make purchasing decisions with confidence.

[1376] "Search keywords" are words or phrases that users enter to describe the information they are looking for.

[1377] A "web page" is a document containing information that can be accessed on the Internet and is written in a format such as HTML.

[1378] "Generative AI" is a system that uses artificial intelligence technology to generate text and analyze information.

[1379] "Jargon" is a specialized word or phrase used in a particular field or area.

[1380] A "source" is a source or data source from which particular information is obtained.

[1381] A "credibility score" is a number or indicator used to evaluate the reliability of a particular piece of information or source of information.

[1382] "Content" refers to the information contained in a web page or document, such as text, images, or video.

[1383] A "review" is an evaluation or comment written by a user or purchaser describing their own experience or opinion.

[1384] "Search results" are the list of relevant web pages or information that appears when a user enters keywords into a search engine.

[1385] A "product information page" is a web page that contains detailed descriptions, specifications, and reviews of a particular product.

[1386] This invention relates to a system that evaluates the reliability of each product information page when a user searches for a product on an online shopping site, allowing the user to make a purchasing decision based on reliable information. This system is realized by using generative AI to evaluate the use of technical terms and the reliability of information sources, from the input of search keywords to the display of search results, and calculates a reliability score.

[1387] System configuration

[1388] The system consists of the following main components:

[1389] 1. How to enter search keywords

[1390] 2. How to search for related web pages

[1391] 3. Web page analysis using generative AI

[1392] 4. Terminology and Source Credibility Assessment Tools

[1393] 5. How the reliability score is calculated

[1394] 6. A way to integrate trustworthiness scores into search results

[1395] 7. Display of integrated results

[1396] 8. A way to analyze the content of each product page and evaluate the quality of the product descriptions and reviews.

[1397] 9. A means of assigning a credibility score to rated product information pages

[1398] Program processing

[1399] 1. Enter search keywords

[1400] The user enters the product name into the smartphone app and presses the search button.

[1401] 2. Obtaining search results

[1402] The device sends the entered search keywords to the server, which retrieves related product information pages. The search results are temporarily saved.

[1403] 3. Web page analysis

[1404] The server sends the URL of each product information page retrieved as a search result to the generation AI module, which downloads the HTML content of each web page and analyzes it, including extracting the key content and metadata within the page.

[1405] 4. Terminology and source credibility assessment

[1406] The generative AI evaluates the use of technical terms on product information pages and the reliability of the sources. Specifically, it checks whether technical terms are used accurately and evaluates whether the cited sources are trustworthy. For example, if the source is a public institution or an academic paper, it will assign a high reliability score.

[1407] 5. Calculating the reliability score

[1408] The server calculates a reliability score for each product information page based on the results of the AI ​​generation. The reliability score is displayed as a number or letter, allowing users to determine the reliability of each product information page at a glance.

[1409] 6. Integrating Trust Scores into Search Results

[1410] The server integrates the reliability score into the search results by adding it below the title of each product information page.

[1411] 7. Displaying the integrated results

[1412] The terminal displays the integrated search results received from the server to the user, who can then review the displayed search results and select which information page to view based on the reliability score.

[1413] Hardware and software used

[1414] Hardware: Servers, smartphones

[1415] Software: Generative AI (GPT-4, etc.), Python, web frameworks (Django, Flask), databases (PostgreSQL)

[1416] Specific examples

[1417] For example, suppose a user enters "vitamin supplements" into a smartphone app and presses the search button. The device sends this search keyword to the server, which retrieves related product information pages. The server then uses generative AI to analyze each product information page and evaluates the use of technical terms and the reliability of the information source. A reliability score is calculated based on the evaluation results, and the server integrates the reliability score into the search results. The device then displays the search results with the reliability score to the user.

[1418] Prompt Sentence Examples

[1419] The prompt for the search "vitamin supplement" is:

[1420] Rate the trustworthiness of this product information page:

[1421] URL: https: / / example.com / vitamin-supplement

[1422] Key points: Accuracy of product description, quality of reviews, citation of reliable sources

[1423] The analysis results obtained from the generative AI model include numerical values ​​such as "content accuracy: 85%" and "source reliability: 90%." A reliability score is calculated based on this and displayed to the user. Users can then select products and make purchasing decisions with confidence based on the reliability score.

[1424] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1425] Step 1:

[1426] The user enters the product name into the smartphone app and presses the search button. The entered keywords are sent to the device as "search keywords." The entered data is the product name specified by the user, and this becomes the base data for subsequent processing.

[1427] Step 2:

[1428] The server receives the "search keywords" sent from the terminal. The server uses existing search engine functions to search for the relevant product information page. At this point, the only input is the "search keywords," and the output is a list of URLs for related product information pages.

[1429] Step 3:

[1430] The server sends a list of URLs for each product information page obtained as a search result to the generation AI module. The generation AI downloads the HTML content of each product and analyzes it. The analysis includes extracting the main content and metadata within the page. The input data is the list of URLs for the product information pages, and the output data is the analyzed HTML content.

[1431] Step 4:

[1432] The generative AI evaluates the use of technical terms on product information pages and the reliability of the sources. Specifically, it checks whether technical terms are used accurately and evaluates whether the cited sources are trustworthy. If highly reliable sources such as public institutions or academic papers are cited, the rating is higher. The input data is HTML content, and the output data is a reliability rating score.

[1433] Step 5:

[1434] The server calculates the reliability score of each product information page based on the results of the evaluation by the generation AI. The reliability score is calculated based on a specific algorithm and is displayed as a number or letter that allows users to determine the reliability of each product information page at a glance. The input data is the reliability evaluation score of each product information page, and the output data is the reliability score.

[1435] Step 6:

[1436] The server integrates the reliability scores into the search results. Specifically, it adds the reliability score below the title of each product information page. This process visually displays the reliability rating for all search results. The input data is the reliability score and a list of search results, and the output data is a list of search results with the reliability scores added.

[1437] Step 7:

[1438] The terminal displays the integrated search results with the reliability scores received from the server to the user. The user checks the displayed search results and selects which product information page to view based on the reliability scores. The input data are the search results with the reliability scores added, and the output data are the product information pages displayed to the user.

[1439] Through the above steps, the user can easily select highly reliable product information and make a purchase decision with confidence.

[1440] 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.

[1441] The present invention provides a system for providing more personalized search results by combining a conventional system that uses generative AI to evaluate the trustworthiness of web pages with an emotion engine that recognizes user emotions.

[1442] System configuration

[1443] The system consists of the following main components:

[1444] 1. How to enter search keywords

[1445] 2. How to search for related web pages

[1446] 3. Web page analysis using generative AI

[1447] 4. Terminology and Source Credibility Assessment Tools

[1448] 5. How the reliability score is calculated

[1449] 6. Means of integrating trustworthiness scores into search results

[1450] 7. Display of integrated results

[1451] 8. Emotion engine that recognizes user emotions

[1452] 9. Sentiment-Based Search Results Customization

[1453] Program processing

[1454] The system proceeds as follows:

[1455] Enter search keywords

[1456] A user enters a keyword (e.g., "cold treatment") into the search field of a search engine and presses the search button. This starts the search process, but at this point the emotion engine is also activated. The emotion engine analyzes the user's input method and speed, facial expressions, tone of voice, etc., to recognize the user's emotional state (e.g., impatience, anxiety, relaxation, etc.).

[1457] Getting search results

[1458] The device sends the emotion data recognized by the emotion engine along with the entered search keywords to the server, where the server searches for related web pages using existing search algorithms and temporarily stores the results.

[1459] Web page analysis

[1460] The server sends the URL of each web page retrieved as a search result to the generation AI module, which downloads the HTML content of each web page and analyzes it, including extracting the key content and metadata within the page.

[1461] Terminology and source credibility assessment

[1462] The generative AI evaluates the use of domain-specific terminology and determines the trustworthiness of cited sources. For example, it assigns a higher credibility score to sources from public institutions or academic papers. Furthermore, it can prioritize sources that inspire a sense of security based on the user's emotional state.

[1463] Calculating the reliability score

[1464] The server calculates a credibility score for each web page based on the evaluation results of the generation AI. In addition, it takes into account data from the emotion engine and adjusts the score according to the user's current emotional state. For example, if a user is feeling anxious, it assigns a higher score to information that is highly reliable and reassuring.

[1465] Integrating trust scores into search results

[1466] The server integrates the reliability score into the search results, and adjusts the display of search results based on the user's emotional state. For example, if a user is in a hurry, the server highlights the results so that the reliability can be seen at a glance.

[1467] Viewing the integration results

[1468] The server sends the integrated search results to the device, which then displays the results in a format that takes the user's emotions into consideration.The user can then review the displayed search results and select the most appropriate web page to view, with the reliability score and emotional engine having been adjusted accordingly.

[1469] Specific examples

[1470] For example, if the emotion engine recognizes that a user searching for the keyword "cold treatment" is feeling anxious or impatient when typing, the emotion engine sends this information to the server. The server retrieves search results based on that information, and the generation AI analyzes the webpage. It then evaluates the reliability of the terminology and information source, and calculates a reliability score taking into account the emotional data. Finally, when the search results incorporating the reliability score are sent to the user's device, reliable information that will reassure users who are feeling anxious is highlighted.

[1471] This allows users to easily access the most appropriate information based on their emotional state at the time, providing a more personalized search experience.

[1472] The processing flow will be explained below.

[1473] Step 1: Enter search keywords

[1474] 1. A user enters a keyword (e.g., "cold treatment") into the search field of a search engine.

[1475] 2. The user presses the "Search" button.

[1476] 3. The device monitors the user's input method (speed, rhythm, etc.), facial recognition camera, and tone of voice during voice input, and activates the emotion engine.

[1477] 4. The emotion engine analyzes the user's emotional state (e.g., impatience, anxiety, relaxation, etc.) and generates emotion data.

[1478] Step 2: Getting search results

[1479] 1. The device sends the entered search keywords and the emotion data recognized by the emotion engine to the server.

[1480] 2. The server searches for relevant web pages based on the keywords.

[1481] 3. The server generates a list of URLs of the web pages obtained as search results and temporarily stores them.

[1482] Step 3: Analyze the web page

[1483] 1. The server retrieves the URL of each web page from the URL list in turn.

[1484] 2. The server downloads the HTML content of each web page.

[1485] 3. The server sends the downloaded HTML content to the generation AI module.

[1486] 4. Generative AI analyzes the page structure and extracts key content and metadata.

[1487] Step 4: Assess terminology and source credibility

[1488] 1. Analyze the key content extracted by the generative AI and evaluate the use of technical terms.

[1489] 2. Generative AI analyzes citations and links within web pages and evaluates the source of the citation or link.

[1490] 3. If the generating AI determines that the source is based on a public institution or academic paper, it will assign a higher credibility score to that source.

[1491] 4. The generative AI uses data from the emotion engine to adjust the importance of information according to the user's emotional state. For example, if a user is feeling anxious, information that gives a sense of security will be given a high score.

[1492] Step 5: Calculate the reliability score

[1493] 1. The server calculates the trustworthiness score for each web page based on the content evaluated by the AI.

[1494] 2. The server also takes into account data from the emotion engine and adjusts the reliability score according to the user's emotional state.

[1495] 3. The server generates and temporarily stores a final credibility score for each web page.

[1496] Step 6: Integrating Trust Scores into Search Results

[1497] 1. The server integrates the confidence score into the search results.

[1498] 2. The server creates a data structure that adds an authority score to each web page below its title.

[1499] 3. The emotion engine determines how search results are displayed (e.g., highlighted, color-coded) depending on the user's emotional state.

[1500] Step 7: Viewing the integration results

[1501] 1. The server sends the consolidated search results to the device.

[1502] 2. The search results received by the device are displayed in a format that takes into account the user's emotional state.

[1503] 3. Users can see the credibility score under the title of each search result displayed and view customized information based on their emotional state.

[1504] Through the above steps, the system of the present invention provides personalized search results that adapt to the user's emotional state to enhance the user's search experience.

[1505] Example 2

[1506] 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."

[1507] Conventional search systems present relevant information based on keywords entered by the user, but because they do not take the user's emotional state into consideration, it is difficult to provide the optimal information the user is looking for. As a result, even if the user is feeling anxious or impatient, information that provides a sense of security is not prioritized, resulting in a poor user experience.

[1508] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for inputting a search keyword; means for searching for related web pages based on the input search keyword and user emotional data; means for activating an emotional engine and analyzing the user's emotional state; means for activating a generation AI for analyzing each web page obtained as a search result; means for evaluating the use of technical terms and the reliability of information sources using the generation AI; means for calculating a reliability score for each web page taking into account the user's emotional data; means for integrating the reliability score into the search result and adjusting the display method based on the user's emotional state; and means for displaying the integrated search result. This makes it possible to provide personalized search results that take into account the user's emotional state.

[1509] The "means for inputting search keywords" refers to a device or function that allows a user to input search keywords through the interface of a search engine.

[1510] "Means for searching for related web pages based on input search keywords and user emotion data" refers to a device or function for finding related web pages using search keywords input by the user and emotion data analyzed by the emotion engine.

[1511] "Means for activating an emotion engine and analyzing the user's emotional state" refers to a device or function for recognizing the user's emotions (e.g., impatience, anxiety, relaxation) by analyzing the user's input method, speed, facial expression, tone of voice, etc.

[1512] "Means for launching a generative AI to analyze each web page obtained as a search result" refers to a device or function that launches a generative AI model to analyze the content of each web page obtained as a search result.

[1513] "Means for using generative AI to evaluate terminology usage and source reliability" refers to a device or function that uses generative AI to analyze and evaluate the use of terminology within a web page and the reliability of cited sources.

[1514] The "means for calculating the reliability score of each web page taking into account the user's emotional data" refers to a device or function for calculating the reliability score of each web page taking into account the evaluation results of the generation AI and the user's emotional state.

[1515] "Means for integrating confidence scores into search results and adjusting the display based on the user's emotional state" refers to a device or function for incorporating calculated confidence scores into search results and changing the display of search results depending on the user's emotional state.

[1516] The "means for displaying integrated search results" is a device or function for displaying search results including reliability scores on a user's terminal so that the user can view them.

[1517] The following describes in detail the mode for carrying out the present invention: The system utilizes generative AI to evaluate the trustworthiness of web pages, and combines it with an emotion engine that recognizes user emotions to provide more personalized search results.

[1518] System configuration

[1519] The system consists of the following main elements:

[1520] 1. How to enter search keywords

[1521] 2. How to search for related web pages

[1522] 3. Web page analysis using generative AI

[1523] 4. Terminology and Source Credibility Assessment Tools

[1524] 5. How the reliability score is calculated

[1525] 6. Means of integrating trustworthiness scores into search results

[1526] 7. Display of integrated results

[1527] 8. Emotion engine that recognizes user emotions

[1528] 9. Sentiment-Based Search Results Customization

[1529] Enter search keywords

[1530] A user enters a keyword (e.g., "cold treatment") into the search field of a search engine and presses the search button. This action starts the search process and simultaneously activates the emotion engine. The emotion engine analyzes the user's input method, speed, facial expression, tone of voice, etc., to recognize the user's emotional state (e.g., impatience, anxiety, relaxation, etc.).

[1531] Getting search results

[1532] The device sends the entered search keywords and the emotion data recognized by the emotion engine to the server, which then uses existing search algorithms to search for related web pages and temporarily stores the results.

[1533] Web page analysis

[1534] The server sends the URL of each web page retrieved as a search result to the generative AI module. The generative AI downloads the HTML content of each web page, extracts key content and metadata, and analyzes it. The generative AI model can be implemented using Python libraries such as "BeautifulSoup" and "Selenium."

[1535] Terminology and source credibility assessment

[1536] Based on the analyzed content, the generative AI evaluates the use of technical terms and the reliability of cited sources. For example, if the source is a public institution or academic paper, it will give a high reliability score. It can also take into account the user's emotional state and prioritize sources that give a sense of security.

[1537] Calculating the reliability score

[1538] The server calculates a reliability score for each web page based on the evaluation results of the generated AI and data from the emotion engine. For example, if a user is feeling anxious, a high score can be assigned to information that is highly reliable and reassuring.

[1539] Integrating trust scores into search results

[1540] The server integrates the reliability score into search results, adjusting the presentation based on the user's emotional state—for example, highlighting results to make their reliability clear at a glance to users in a hurry.

[1541] Viewing the integration results

[1542] The server sends the integrated search results to the device, which then displays the search results in a format that takes the user's emotions into consideration.The user can review the displayed search results and select the most appropriate web page to view, with the results adjusted by the reliability score and emotion engine.

[1543] Specific examples

[1544] For example, if the emotion engine recognizes that a user searching for the keyword "cold treatment" is feeling anxious or impatient when typing, the emotion engine sends this emotional data to the server. The server retrieves search results based on that information, and the generation AI analyzes the webpage. The analysis results then evaluate the reliability of the terminology and information sources, and calculate a reliability score taking the emotional data into account. Finally, the search results, incorporating the reliability score, are sent to the user's device, and reliable information that will reassure anxious users is highlighted and displayed.

[1545] Example prompt

[1546] If the user enters the keyword "cold treatment" and the emotion engine recognizes that the user is feeling anxious, prioritize displaying reliable information that provides a sense of security.

[1547] This system allows users to easily obtain the most appropriate information based on their emotional state at the time, providing a more personalized search experience.

[1548] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1549] Step 1:

[1550] The user enters a search keyword. Specifically, the user enters "cold treatment" in the search field of the search engine and presses the search button. This input triggers the search process, obtaining "cold treatment" as the input and proceeding to the next step.

[1551] Step 2:

[1552] The emotion engine is activated. It collects and analyzes information such as the user's input method and speed, facial expressions, and tone of voice. For example, if the user's input speed is fast and they are putting a lot of force into the keys, it is likely that they are impatient. The emotion engine uses this data to recognize the user's emotional state (e.g., impatience, anxiety, relaxation). The user's behavioral data is taken as input, and the emotion analysis results are output.

[1553] Step 3:

[1554] The device sends the search keywords and emotion data to the server. Here, the device sends the keyword "cold treatment" and the emotion data (e.g., "impatience") analyzed by the emotion engine to the server. The search keywords and emotion data are sent to the server as input, and the next processing step is executed based on this.

[1555] Step 4:

[1556] The server searches for relevant web pages. Based on the received search keyword "cold treatment," the server uses an existing search algorithm to search for relevant web pages. At this time, emotion data is temporarily stored. The input is the search keyword, and the output is a list of relevant web pages.

[1557] Step 5:

[1558] The server sends the URLs of the web pages to the generation AI. The server then sends the URLs of each web page obtained as a search result to the generation AI module. The list of URLs obtained as input is sent, and analysis begins in the next step.

[1559] Step 6:

[1560] The generative AI analyzes the content of web pages. It downloads the HTML content of each web page and extracts key content and metadata. This analysis is performed using tools such as "BeautifulSoup" and "Selenium." The input is the HTML content, and the output is the parsed data.

[1561] Step 7:

[1562] Generative AI evaluates the reliability of terminology and sources. Based on the analyzed data, generative AI determines the use of terminology and the reliability of cited sources. For example, if the source is a public institution or academic paper, it will be assigned a high reliability score. The analyzed data is input, and a reliability evaluation score is generated as output.

[1563] Step 8:

[1564] The server calculates the reliability score. The server calculates the reliability score for each web page based on the evaluation results of the generation AI and data from the emotion engine. For example, if a user is feeling anxious, a high score is assigned to information that is highly reliable and gives a sense of security. The inputs are the evaluation results and emotion data, and the output is a reliability score.

[1565] Step 9:

[1566] The server integrates the reliability score into the search results. The calculated reliability score is incorporated into the search results and the display method is adjusted based on the user's emotional state. For example, for a user in a hurry, results are highlighted so that the reliability can be seen at a glance. The input is the reliability score and the search results, and the output is the adjusted search results.

[1567] Step 10:

[1568] The server sends the integrated results to the terminal. The server sends the integrated search results to the terminal, where they can be viewed by the user. The integrated search results are taken as input and sent to the terminal as output.

[1569] Step 11:

[1570] The device displays the search results. The device displays the search results in a format that takes the user's emotions into consideration. The user can review the displayed search results and select and view the optimal web page, adjusted by the reliability score and emotion engine. The input is the integrated results, and the output is the displayed search results.

[1571] (Application example 2)

[1572] 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."

[1573] Conventional navigation systems suggest routes without taking the user's emotional state into consideration, and therefore are unable to provide the optimal route, especially when the user is feeling anxious or impatient. Furthermore, they lack the reliability of search results and the ability to evaluate safety in real time, creating a need for a navigation system that users can use with peace of mind.

[1574] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for inputting search keywords, means for searching for related information based on the input search keywords, means for activating a generation AI for analyzing each piece of information obtained as a search result, means for evaluating the use of technical terms and the reliability of information sources using the generation AI, means for calculating a reliability score for each piece of information based on the evaluation results, means for integrating the reliability score into the search results, means for displaying the integrated search results, means for detecting the user's emotions using an emotion recognition device, and means for customizing the search results based on the user's emotion data. This enables personalized route selection according to the user's emotional state, thereby providing safe and reliable navigation.

[1575] A "search keyword" is a word or phrase that a user enters when searching for information.

[1576] "Search results" are a collection of related information obtained based on search keywords.

[1577] "Generative AI" is a system or program that uses artificial intelligence technology to analyze and evaluate information.

[1578] "Jargon" is specialized words and phrases used in a particular field.

[1579] "Source credibility" is a criterion for assessing the reliability of the source or source from which information is provided.

[1580] The "trustworthiness score" is a numerical representation of the reliability of information or sources evaluated by the generating AI.

[1581] An "emotion recognition device" is a device that uses a camera, a voice recognition system, etc. to detect a user's emotions in real time.

[1582] "Emotion data" is information that represents the emotional state of a user detected by an emotion recognition device.

[1583] A "means for customizing search results" is a method or apparatus that tailors and personalizes the search results obtained based on the user's emotional data.

[1584] The embodiment of the present invention will be described as a navigation system that operates within an autonomous vehicle. This system recognizes the user's emotions and uses generative AI to evaluate and provide road information and route reliability in real time. The following elements are required to implement this system:

[1585] System configuration

[1586] Enter search keywords

[1587] When a user inputs a destination into a navigation system, the system also recognizes the user's emotions.

[1588] User Emotion Recognition

[1589] The device is equipped with a camera and a voice recognition system that analyzes the user's facial expressions and voice in real time. This function uses a camera (e.g., Logitech C920) and a voice recognition system (e.g., Google Cloud Speech-to-Text). The recognized emotion data is sent to a server.

[1590] Find a route

[1591] The server searches for available routes based on the input destination, using traffic information services such as Google Maps API.

[1592] Analysis by generative AI

[1593] The route information obtained as a search result is analyzed by a generation AI on the server. The generation AI evaluates the latest road conditions and traffic information and quantifies the reliability of each route. This generation AI module uses models such as GPT-4.

[1594] Calculating the reliability score

[1595] Based on the data evaluated by the generative AI, a reliability score is calculated for each route using Scikit-learn's MLPClassifier or a custom function, and data from the emotion engine is also taken into account to adjust the score according to specific emotional states.

[1596] Customize route display

[1597] Based on the reliability score, the system displays the optimal route based on the user's emotional state on the device's navigation system screen. For example, it highlights safe and reliable routes for a user who is feeling anxious.

[1598] Specific examples

[1599] For example, if a user inputs the "safest route" into the navigation system of an autonomous vehicle, the system may recognize from the user's facial expressions and voice that they are feeling "anxiety" or "impatience." The emotion data and the searched route information are sent together to the server, and the generation AI analyzes these routes.

[1600] The generation AI calculates a reliability score for each route and uses prompts such as:

[1601] "If you want to avoid routes where there are traffic accidents, please provide a reliable route taking into account the current road conditions."

[1602] Based on this, the generation AI evaluates the most reliable route and displays a customized version that reflects the emotional data. Ultimately, route information that gives a sense of security to a user who is feeling anxious is displayed on the device, and the optimal route is suggested for the user.

[1603] In this way, the present invention provides a personalized navigation experience that responds to the user's emotional state.

[1604] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1605] Step 1:

[1606] The user inputs a destination into the navigation system inside the autonomous vehicle. The user's input is made via keyboard or voice. At this time, the camera and voice recognition system capture the user's facial expressions and voice in real time, and emotional data is acquired. The input destination data and emotional data are then sent to the terminal.

[1607] Step 2:

[1608] The device sends the acquired destination data and emotion data to the server. The server uses the destination data to search for multiple route information using the Google Maps API. The route information includes the distance, travel time, current traffic conditions, etc. for each route.

[1609] Server input: Destination data, emotion data

[1610] Server output: Route information

[1611] Step 3:

[1612] The server sends each route to the AI ​​generator, which analyzes the route and evaluates road conditions (traffic, accidents, construction, etc.) and other important factors. A generative AI model such as GPT-4 is used for the analysis. The evaluation results are quantified as a reliability score.

[1613] Input for the generation AI: Route information, prompt (e.g., "If I want to avoid a route where a traffic accident has occurred, please provide a reliable route taking into account the current road conditions.")

[1614] Generative AI output: Confidence score

[1615] Step 4:

[1616] The server adjusts the reliability score obtained by the AI ​​generator by taking into account emotional data. For example, if the user is in a hurry, it will give a higher score to safer and more reliable routes.

[1617] Server input: confidence score, sentiment data

[1618] Server output: Adjusted reliability score

[1619] Step 5:

[1620] The server reevaluates the reliability of each route based on the adjusted reliability score, selects the route that best suits the user's emotional state, and sends the selection result to the device.

[1621] Server Input: Adjusted Reliability Score

[1622] Server output: Optimal route information

[1623] Step 6:

[1624] The device displays the optimal route information sent from the server on the navigation system screen. When the user is in a hurry, reliable routes are visually highlighted, allowing the user to select them with confidence.

[1625] Input on device: Optimal route information

[1626] Device output: Route display on navigation screen

[1627] In this way, the present invention provides a navigation system that takes into account the emotional state of the user.

[1628] 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.

[1629] 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.

[1630] 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.

[1631] 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.

[1632] 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.

[1633] 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.

[1634] 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).

[1635] 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.

[1636] 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."

[1637] 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.

[1638] 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).

[1639] 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.

[1640] 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.

[1641] 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.

[1642] 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.

[1643] 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.

[1644] 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.

[1645] 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.

[1646] 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.

[1647] 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.

[1648] 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.

[1649] The following is further disclosed regarding the above embodiment.

[1650] (Claim 1)

[1651] A means for inputting search keywords;

[1652] A means for searching for related web pages based on input search keywords;

[1653] means for invoking a generative AI for analyzing each web page retrieved as a search result;

[1654] A means of assessing terminology usage and source credibility using generative AI; and

[1655] means for calculating an authority score for each web page based on the evaluation results;

[1656] a means of integrating trustworthiness scores into search results;

[1657] A system including means for displaying integrated search results.

[1658] (Claim 2)

[1659] The system of claim 1, further comprising means for performing a structural analysis of the web page and extracting key content and metadata when the generating AI analyzes the web page.

[1660] (Claim 3)

[1661] The system of claim 1, further comprising a means for setting a high reliability score when the reliability of the information source evaluated by the generating AI is based on a public institution or an academic paper.

[1662] "Example 1"

[1663] (Claim 1)

[1664] A means for inputting search keywords;

[1665] A means for searching for related web pages based on input search keywords;

[1666] means for invoking a generative AI for analyzing each web page retrieved as a search result;

[1667] A means of assessing terminology usage and source credibility using generative AI; and

[1668] means for calculating an authority score for each web page based on the evaluation results;

[1669] a means of integrating trustworthiness scores into search results;

[1670] a means for displaying the consolidated search results;

[1671] means for transmitting search keywords to a server and temporarily storing the retrieved search results;

[1672] A means for the generative AI to download the HTML content of a webpage and extract key content and metadata within the page;

[1673] The system includes means for displaying the integrated search results received by the user's terminal from the server.

[1674] (Claim 2)

[1675] The system of claim 1, further comprising means for performing a structural analysis of the web page and extracting key content and metadata when the generating AI analyzes the web page.

[1676] (Claim 3)

[1677] The system of claim 1, further comprising a means for setting a high reliability score when the reliability of the information source evaluated by the generating AI is based on a public institution or an academic paper.

[1678] "Application Example 1"

[1679] (Claim 1)

[1680] A means for inputting search keywords;

[1681] A means for searching for related web pages based on input search keywords;

[1682] means for invoking a generative AI for analyzing each web page retrieved as a search result;

[1683] A means of assessing terminology usage and source credibility using generative AI; and

[1684] means for calculating an authority score for each web page based on the evaluation results;

[1685] a means of integrating trustworthiness scores into search results;

[1686] a means for displaying the consolidated search results;

[1687] A means to analyze the content of each product information page and evaluate the quality of the product descriptions and reviews,

[1688] The system includes a means for assigning an authority score to the rated product information page.

[1689] (Claim 2)

[1690] The system of claim 1, further comprising means for performing a structural analysis of the web page and extracting key content and metadata when the generating AI analyzes the web page.

[1691] (Claim 3)

[1692] The system of claim 1, further comprising a means for setting a high reliability score when the reliability of the information source evaluated by the generating AI is based on a public institution or an academic paper.

[1693] "Example 2: Combining Emotion Engines"

[1694] (Claim 1)

[1695] A means for inputting search keywords;

[1696] A means for searching for related web pages based on the input search keyword and user emotion data;

[1697] means for activating an emotion engine and analyzing the user's emotional state;

[1698] means for invoking a generative AI for analyzing each web page retrieved as a search result;

[1699] A means of assessing terminology usage and source credibility using generative AI; and

[1700] means for calculating a credibility score for each web page taking into account user sentiment data;

[1701] a means for integrating confidence scores into search results and adjusting their presentation based on the user's emotional state;

[1702] A system including means for displaying integrated search results.

[1703] (Claim 2)

[1704] The system of claim 1, further comprising means for performing a structural analysis of the web page and extracting key content and metadata when the generating AI analyzes the web page.

[1705] (Claim 3)

[1706] The system of claim 1, further comprising a means for setting a high reliability score when the reliability of the information source evaluated by the generating AI is based on a public institution or an academic paper.

[1707] "Application example 2 when combining emotion engines"

[1708] (Claim 1)

[1709] A means for inputting search keywords;

[1710] A means for searching for related information based on the input search keywords;

[1711] A means for launching a generating AI for analyzing each piece of information obtained as a search result;

[1712] A means of assessing terminology usage and source credibility using generative AI; and

[1713] means for calculating a reliability score for each piece of information based on the evaluation results;

[1714] a means of integrating trustworthiness scores into search results;

[1715] a means for displaying the consolidated search results;

[1716] means for detecting an emotion of a user using an emotion recognition device;

[1717] A system including means for customizing search results based on user sentiment data.

[1718] (Claim 2)

[1719] The system of claim 1, further comprising means for performing structural analysis and extracting key elements and metadata when the generative AI analyzes the information.

[1720] (Claim 3)

[1721] The system of claim 1, further comprising a means for setting a high reliability score when the reliability of the information source evaluated by the generating AI is based on a public institution or academic research. [Explanation of symbols]

[1722] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for inputting search keywords; A means for searching for related web pages based on input search keywords; means for invoking a generative AI for analyzing each web page retrieved as a search result; A means of assessing terminology usage and source credibility using generative AI; and means for calculating an authority score for each web page based on the evaluation results; a means of integrating trustworthiness scores into search results; A system including means for displaying integrated search results.

2. The system of claim 1, further comprising means for performing a structural analysis of the web page and extracting key content and metadata when the generation AI analyzes the web page.

3. The system of claim 1, further comprising a means for setting a high reliability score when the reliability of the information source evaluated by the generation AI is based on a public institution or an academic paper.

Images