System

Abstract

A system is provided.SOLUTION: A system comprising: means for allowing a user to input a retrieval keyword; means for allowing a server to acquire relevant documents from an academic database; means for allowing the server to score relevance of the documents and select high-ranking documents; means for allowing the server to summarize abstracts, results, and discussions of the documents by using generated AI; and means for allowing the terminals to display the received summary information.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In today's information society, anyone can disseminate information, making it important for information recipients to be able to discern accurate information. However, academic literature often contains specialized content, which can be difficult for general readers to understand, making it difficult to know where to start. For this reason, there is a need for a system that allows users to easily search for literature on topics they are interested in and quickly understand its content. [Means for solving the problem]

[0005] The present invention provides a system that includes a means for users to input search keywords, a means for a server to retrieve relevant literature from academic databases, a means for the server to score the relevance of the literature and select the top literature, and a means for the server to summarize the literature abstracts, results, and discussions using generative AI. The system also includes a means for the server to send the summarized data to the user's terminal and a means for the terminal to display the received summary data. This system allows users to quickly and easily understand the key points of relevant academic literature.

[0006] A "user" refers to a person who operates this system, inputs search keywords, and obtains information.

[0007] "Server" refers to a computer system that retrieves documents from relevant academic databases based on searched keywords, processes and edits them, and sends them to the user's terminal.

[0008] "Terminal" refers to a device (e.g., a PC, smartphone, tablet, etc.) that a user operates to input information and receives and displays data sent from a server.

[0009] "Search keywords" refer to words or phrases that a user enters into a terminal to identify the information they want to know.

[0010] "Academic database" refers to an online source that aggregates and provides searchable academic articles and research materials.

[0011] "Literature" refers to papers and research materials stored in academic databases.

[0012] The "relevance score" is a numerical representation of the degree of relevance of the retrieved documents to the search keywords entered by the user.

[0013] "Generative AI" refers to an artificial intelligence system that uses natural language processing techniques to create literature summaries.

[0014] An "abstract" is a section that briefly summarizes the outline or gist of a document.

[0015] "Results" refers to the section of the literature that summarizes the results of research or investigations.

[0016] "Discussion" refers to the section of the paper that provides interpretation, consideration, and conclusions of the results. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11]FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0018] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0020] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0021] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0022] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0023] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0025] [First embodiment]

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

[0027] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0028] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0029] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0030] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0034] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0035] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0036] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0037] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0038] This invention is a system in which a user inputs search keywords, a server retrieves relevant literature from an academic database based on those keywords, creates a summary of that literature using a generative AI, and displays it on the user's terminal.

[0039] Specifically, the user operates their device to enter search keywords and click the search button. The server receives the entered search keywords and sends a search request to relevant academic databases. The server analyzes the literature data retrieved from the academic databases and selects the top literature based on relevance scores. This literature data is then used by generative AI to create abstracts, results, and discussion summaries. The summarized data is then sent from the server to the user's device, where it is displayed in an easy-to-read format.

[0040] The following will explain this with some specific examples.

[0041] Specific examples

[0042] 1. User searches for a theme

[0043] The user enters the keyword "COVID-19 vaccine effectiveness" into the device and clicks the search button.

[0044] 2. The server searches the academic database

[0045] The server sends an API request to academic databases using the keyword "COVID-19 vaccine effectiveness," such as PubMed and IEEE Xplore.

[0046] 3. The server evaluates and scores the relevance of the documents

[0047] The server analyzes the metadata of the retrieved documents and scores their relevance. For example, documents that contain many keywords such as "COVID-19," "vaccine," and "effectiveness" in the title or abstract are rated as highly relevant.

[0048] 4. Server generates summary using AI

[0049] The server uses generative AI to summarize the abstract, results, and discussion of the top 10 articles, extracting key information from each section and generating a concise summary.

[0050] 5. The server formats and sends the summary data

[0051] The server formats the summarized data into an easy-to-read format (for example, HTML format) and sends it to the user's terminal via an HTTP request.

[0052] 6. The device displays the summary data.

[0053] The device analyzes the received data and displays it in an easy-to-understand manner. For example, a page designed in card format is displayed, listing the paper title, abstract, results, and discussion in that order.

[0054] Overview of the overall system operation

[0055] This system allows users to initiate a search, and the server quickly collects relevant literature and uses generative AI to quickly create summaries, providing users with fast and accurate information. By using this system, users can easily understand the key points of academic literature, even if they lack specialized knowledge.

[0056] The processing flow will be explained below.

[0057] Step 1:

[0058] The user enters search keywords

[0059] The user operates the device, enters keywords related to the topic they want to research in the search box, and clicks the search button.

[0060] Step 2:

[0061] The server receives the search keywords.

[0062] The server receives the search keywords sent from the device, analyzes them, and normalizes them to the appropriate format (for example, converting them to lowercase and removing unnecessary spaces).

[0063] Step 3:

[0064] The server searches academic databases

[0065] The server uses the normalized keywords to send a search query to the search API of an academic database (e.g., PubMed or IEEE Xplore), which returns a list of relevant literature.

[0066] Step 4:

[0067] The server scores the relevance of the documents

[0068] The server receives the list of documents and analyzes the metadata of each document (title, abstract, author information, publication year, etc.), calculates a relevance score, and ranks documents with high scores higher.

[0069] Step 5:

[0070] The server selects the top documents

[0071] Based on the scoring results, the top few documents (for example, 10 documents) are selected, and the data of these selected documents is sent to the next processing step.

[0072] Step 6:

[0073] The server creates a summary using generative AI

[0074] The server passes the abstract, results, and discussion of each selected paper to the AI ​​generator, instructing it to summarize those parts. The AI ​​generator uses natural language processing techniques to extract the key parts of the text and create a concise summary.

[0075] Step 7:

[0076] The server formats the summary results

[0077] The server formats the summary results received from the generation AI, for example, formatting the paper title, abstract, results, and discussion sections in an easy-to-understand format (HTML or JSON format).

[0078] Step 8:

[0079] The server sends the summary results to the terminal.

[0080] The server sends the formatted summary to the user's device using the HTTP protocol.

[0081] Step 9:

[0082] The terminal displays summary data

[0083] The user's device receives and analyzes the summary data sent from the server. The analysis results are then displayed in an easy-to-understand format for the user. For example, the interface can be arranged so that the title of the paper is a clickable link, and the abstract, results, and discussion are displayed in card format.

[0084] This process allows users to quickly understand the gist of complex academic literature.

[0085] Example 1

[0086] 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."

[0087] Currently, many academic databases contain vast amounts of literature data, but there is a lack of a way to quickly find relevant literature based on specific search keywords and provide appropriate summaries. Furthermore, it is difficult for non-experts to understand the content of academic literature. To solve this problem, a system is needed that can efficiently search for and summarize literature and provide it in a user-friendly format.

[0088] 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.

[0089] In this invention, the server includes a means for a user to input search keywords, a means for retrieving relevant literature from an academic database, a means for analyzing the metadata of the literature to evaluate relevance and select top-ranked literature, a means for summarizing the abstracts, results, and discussions of the literature using a generative AI model, a means for transmitting the summarized data to the user's terminal, and a means for displaying the summarized data received by the terminal, thereby enabling the user to quickly find relevant academic literature based on the search keywords and obtain the summaries in an easy-to-understand format.

[0090] A "user" is a user who uses the system to input search keywords and obtain abstract information of academic literature.

[0091] The "server" is a computer system that processes search requests from users, retrieves literature from academic databases, creates summaries using generative AI models, and transmits the summary data to users' devices.

[0092] An "academic database" is a digital database that collects, stores, and makes searchable scientific papers and research results. Examples include PubMed and IEEE Xplore.

[0093] "Metadata" refers to information about a document, specifically data such as the title, author, abstract, and publication date.

[0094] The "relevance score" is a numerical evaluation scale that quantifies the relevance of a document based on search keywords.

[0095] A "generative AI model" is a type of artificial intelligence that generates literature summaries and other text based on input text data (prompts).

[0096] An "abstract" is a section that briefly summarizes the outline or gist of a document.

[0097] "Results" is the section that shows the results of the research described in the literature.

[0098] The "Discussion" section includes interpretations and considerations of the research results described in the literature.

[0099] "Summary data" refers to the abstracts, results, and discussions of literature summarized by generative AI models.

[0100] A "terminal" is a device such as a computer or smartphone that is used by a user to operate the device.

[0101] "Display" means to visually show information on the screen of a terminal.

[0102] This invention is a system in which a user inputs search keywords, a server retrieves relevant literature from an academic database based on those keywords, creates a summary of that literature using a generative AI, and displays it on the user's terminal.

[0103] Specifically, the system is implemented as follows.

[0104] The user operates the device to enter search keywords and clicks the search button. The device then sends a request containing the search keywords to the server via an Internet connection. The server receives this request and sends a search request to academic databases based on the specified keywords. For example, databases such as PubMed and IEEE Xplore can be used.

[0105] The server retrieves relevant literature data from academic databases, which are then stored and analyzed as metadata, including the title, abstract, author, publication date, and so on.

[0106] The server then evaluates and scores the relevance of the retrieved documents. The relevance score is calculated based on the degree of match between the search keywords and the document metadata. Documents with high relevance are ranked higher.

[0107] The server then uses a generative AI model to create summaries of the top-ranked articles, including the abstract, results, and discussion sections, extracting key information from the article and formatting it into a summary.

[0108] Once the summary is complete, the server formats the generated summary data into an easy-to-read format such as HTML, and sends the formatted data to the user's device via an HTTP request.

[0109] The device analyzes the received data and displays it in a user-friendly format, such as a card-style page displaying the paper title, abstract, results, and discussion in that order.

[0110] Specific examples

[0111] Prompt Sentence Examples

[0112] Below is a specific example of a prompt sentence. In this example, the keyword "COVID-19 vaccine effectiveness" is used.

[0113] 1. User searches for a theme

[0114] The user enters the keyword "COVID-19 vaccine effectiveness" into the device and clicks the search button.

[0115] 2. The server searches the academic database

[0116] The server sends an API request to academic databases using the keyword "COVID-19 vaccine effectiveness," such as PubMed and IEEE Xplore.

[0117] 3. The server evaluates and scores the relevance of the documents

[0118] The server analyzes the metadata of the retrieved documents and scores their relevance. For example, documents that contain many keywords such as "COVID-19," "vaccine," and "effectiveness" in the title or abstract are rated as highly relevant.

[0119] 4. Server generates summary using AI

[0120] The server uses generative AI to summarize the abstract, results, and discussion of the top 10 articles, extracting key information from each section and generating a concise summary.

[0121] 5. The server formats and sends the summary data

[0122] The server formats the summarized data into an easy-to-read format (for example, HTML format) and sends it to the user's terminal via an HTTP request.

[0123] 6. The device displays the summary data.

[0124] The device analyzes the received data and displays it in an easy-to-understand manner. For example, a page designed in card format is displayed, listing the paper title, abstract, results, and discussion in that order.

[0125] In this way, users can easily understand the key points of academic literature even if they lack specialized knowledge.

[0126] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0127] Step 1:

[0128] The user enters search keywords into the device and clicks the search button.

[0129] Specific actions: A user opens a browser or application, enters a keyword (e.g., "COVID-19 vaccine effectiveness") in the search box, and clicks the search button.

[0130] Input: Search keyword (e.g., "COVID-19 vaccine effectiveness")

[0131] Output: HTTP request sent from the user's device

[0132] Step 2:

[0133] The server analyzes the search keywords received from the user and sends a search request to the academic database.

[0134] How it works: The server receives HTTP requests, extracts search keywords, and then sends API requests to academic databases such as PubMed and IEEE Xplore.

[0135] Input: HTTP request from user device (e.g., search keyword "COVID-19 vaccine effectiveness")

[0136] Data processing: Extract search keywords and create API requests

[0137] Output: API request to academic database

[0138] Step 3:

[0139] The server analyzes literature data retrieved from academic databases and evaluates its relevance.

[0140] Specific operation: The server receives literature data (in JSON format) returned from academic databases, extracts literature metadata (title, author, abstract, publication date, etc.), and calculates a relevance score.

[0141] Input: Literature data from academic databases (JSON format)

[0142] Data processing: extracting metadata, calculating relevance scores

[0143] Output: Scored bibliography

[0144] Step 4:

[0145] The server selects the top articles based on their relevance scores and creates summaries using a generative AI model.

[0146] Specific operation: The server selects the documents with the highest relevance scores (for example, the top 10), generates a prompt for each document, and inputs the prompt into the generative AI model to generate a summary (abstract, results, and discussion).

[0147] Input: Scored bibliography

[0148] Data processing: Prompt generation, summarization using generative AI models

[0149] Output: Abstracted bibliographic data

[0150] Step 5:

[0151] The server formats the summarized bibliographic data into HTML format and sends it to the user's terminal.

[0152] What it does: It inserts the summarized data into an HTML template, formats it in an easy-to-read format, and then sends it to the user's device as an HTTP response.

[0153] Input: Abstracted bibliographic data

[0154] Data processing: Insertion into HTML template, formatting

[0155] Output: HTML summary data

[0156] Step 6:

[0157] The user's terminal analyzes the received HTML data and displays it to the user.

[0158] Specific operation: The user device receives the HTTP response from the server, parses the HTML data, and displays the data in the browser, showing the paper title, abstract, results, and discussion in an easy-to-understand format.

[0159] Input: HTTP response (HTML data) from the server

[0160] Data processing: HTML data analysis

[0161] Output: Summary information displayed to the user

[0162] (Application example 1)

[0163] 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."

[0164] In conventional virtual stores, systems that allow users to efficiently understand product reviews and support purchasing decisions have not been adequately developed. In particular, checking a huge number of reviews one by one is inefficient, and an improvement in the user experience is required.

[0165] 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.

[0166] In this invention, the server includes a means for a user to input search keywords, a means for the server to retrieve related information from a database, a means for the server to score the relevance of the information and select the top information, a means for the server to summarize the information using a generation AI, a means for the server to send the summarized data to the user's terminal, and a means for the terminal to display the received summarized data. This allows the user to efficiently understand product reviews in a virtual store and make a quick purchase decision.

[0167] "User" means a subject who uses the system to enter search keywords.

[0168] A "server" is a computer system that receives requests from users, retrieves relevant information from a database, and generates a summary of the information.

[0169] A "database" is a collection of information in which various pieces of information are stored, and is referenced by the server to obtain related information.

[0170] "Information" refers to various data such as data obtained from databases, literature, reviews, etc.

[0171] "Relevance scoring" refers to the process of numerically evaluating how relevant the retrieved information is to the user's search keywords.

[0172] "Generative AI" is a system that uses artificial intelligence technology to automatically generate summaries of information.

[0173] A "summary" is a short summary of the main points of the information obtained.

[0174] A "terminal" is a device (such as a smartphone, computer, or HMD) that allows a user to input information and receive and display summary data from a server.

[0175] This invention relates to a product review summarization system for a virtual store, which allows users to efficiently understand reviews related to a specific product and assists them in making purchasing decisions. Specific embodiments for implementing this system are described below.

[0176] First, the user operates a device (such as a smartphone or head-mounted display) to input keywords related to the product. The keywords entered by the user are sent to the server via the interface.

[0177] Based on the received keywords, the server retrieves relevant information from a database that stores reviews of virtual stores and is accessed via an API.

[0178] The server then analyzes the metadata of the retrieved information and scores its relevance. Specifically, it selects the information that is most relevant to the keywords searched by the user, taking into account the frequency and matching rate of keywords contained in the title and text.

[0179] For information selected as highly relevant, the server uses generative AI (e.g., the OpenAI® GPT-3® model) to generate a summary of the review content, which extracts the key points of the original review and presents them in a concise format.

[0180] Once the summary is complete, the server sends the summary data to the user's device, where it is formatted in an appropriate format (e.g., card or list) and displayed in a user-friendly way.

[0181] This system allows users to quickly grasp the main points without having to check a huge number of reviews one by one, which allows users to efficiently gather information about products and make purchasing decisions more quickly.

[0182] For example, if a user enters the keyword "new 4K TV," the server will search for reviews based on that keyword and display a summary of the key points.

[0183] Example prompt sentence:

[0184] Please summarize the following review:

[0185] "This 4K TV offers excellent picture quality, with excellent color reproduction, making it ideal for watching movies. However, the sound quality is average and additional speakers may be needed. The controls are intuitive and easy to use, although the remote can be a bit slow to respond. Overall, it's a great value at this price point."

[0186] summary:

[0187] In this way, the user can quickly understand the detailed review content of the product.

[0188] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0189] Step 1:

[0190] A user operates a terminal to input keywords related to a specific product. The input keywords are sent to the server through the user interface. The input is in text format, and the text data is passed to the server as output.

[0191] Step 2:

[0192] The server receives the received keywords and sends a search request to a database. The database used is a source of accumulated product review information. The input is a keyword, and the output is a collection of related review information.

[0193] Step 3:

[0194] The server analyzes the review information retrieved from the database and scores its relevance. Scoring is based on the frequency of keywords in the title and text. The input is the retrieved review information, and the output is a list of review information with a relevance score.

[0195] Step 4:

[0196] The server selects reviews with high relevance scores and generates summaries using a generative AI model (e.g., OpenAI GPT-3). The text data of the selected reviews is input into the AI ​​model, and the generated summary text is obtained as the output.

[0197] Step 5:

[0198] The server formats the generated summary data into an appropriate format (e.g., HTML or card format) and sends it to the user's device. The input is the summary data, and the output is the formatted summary data.

[0199] Step 6:

[0200] The terminal displays the summary data received. The display method is designed with the user's convenience in mind, for example, presenting the summary data in a card format or list format. The input is the formatted summary data, and the output is visual information that the user can see.

[0201] This allows users to efficiently understand product review information and make purchasing decisions quickly.

[0202] 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.

[0203] This invention is a system that allows a user to input search keywords, searches for, retrieves, scores, and summarizes related literature, and provides the results to the user.It also has the ability to recognize the user's emotions and adjust the search results and summary display.

[0204] Specifically, when a user enters a search keyword into a device, the server searches academic databases using that keyword to retrieve relevant literature. The retrieved literature is then relevance-scored by the server, and the top-ranking literature is selected. For the selected literature, a generative AI automatically summarizes the literature's abstract, results, and discussion. The summarized data is then formatted appropriately by an emotion engine that recognizes the user's emotions and sent to the device.

[0205] A unique feature of this system is that it recognizes the user's emotions in real time and provides information in a form that is optimal for the user's psychological state. We will explain this with some concrete examples below.

[0206] Specific examples

[0207] 1. User searches for a theme

[0208] A user enters the keyword "COVID-19 vaccine effectiveness" into their device and clicks the search button. At that time, the device's camera and microphone are used to collect emotional data from the user's facial expressions and voice. For example, emotional information such as whether the user is nervous or relaxed is collected.

[0209] 2. The server searches the academic database

[0210] The server sends a search query to academic databases using the keyword "COVID-19 vaccine effectiveness," retrieving a list of relevant literature from databases such as PubMed and IEEE Xplore.

[0211] 3. The server scores the relevance of the documents

[0212] The server analyzes the list of documents it has acquired, calculates a relevance score based on the metadata of each document, and ranks documents with high scores higher.

[0213] 4. The server selects the top documents

[0214] Based on the scoring results, the top 10 documents are selected and the next step is carried out on these documents.

[0215] 5. The server creates a summary using generative AI

[0216] The server passes the abstracts, results, and discussions of the selected papers to the AI ​​generator, which then extracts the key information from each section and generates a concise summary.

[0217] 6. The server adjusts the summary using the emotion engine.

[0218] The server uses an emotion engine to format the summary results to best suit the user's emotional state. For example, if the user is feeling anxious, the summary results will be adjusted to be more concise and easy to understand.

[0219] 7. The server sends the summary results to the terminal.

[0220] The server sends the formatted summary to the user's device using the HTTP protocol. The summary is displayed in an optimal interface based on the analysis results of the emotion engine.

[0221] 8. The terminal displays the summary data.

[0222] The device analyzes the data it receives and displays it in an easy-to-understand manner. For example, if the emotion engine determines that the user is feeling stressed, it adjusts the color, font size, and layout of the summary data to make it easier to read.

[0223] This series of processes allows users to receive appropriate information according to their emotional state, enabling them to quickly understand the main points of complex academic literature. This system is not only effective in providing accurate information, but also in reducing the psychological burden on users.

[0224] The processing flow will be explained below.

[0225] Step 1:

[0226] The user enters search keywords

[0227] The user operates the device, enters keywords related to the topic they want to research (e.g., "COVID-19 vaccine effectiveness") in the search box, and clicks the search button. At this time, the device's camera and microphone collect the user's emotional data (facial expressions and tone of voice).

[0228] Step 2:

[0229] The device sends emotion data

[0230] The device sends the collected emotion data to a server using a secure communication protocol (e.g., HTTPS).

[0231] Step 3:

[0232] The server receives the search keywords.

[0233] The server receives the search keywords sent from the device, parses them, and normalizes them to the appropriate format (e.g., converts them to lowercase, removes unnecessary spaces).

[0234] Step 4:

[0235] The server searches academic databases

[0236] The server uses the normalized keywords to send a search query to the search API of an academic database (e.g., PubMed or IEEE Xplore), which returns a list of relevant literature.

[0237] Step 5:

[0238] The server scores the relevance of the documents

[0239] The server receives the list of documents and analyzes the metadata of each document (title, abstract, author information, publication year, etc.), calculates a relevance score, and ranks documents with high scores higher.

[0240] Step 6:

[0241] The server selects the top documents

[0242] Based on the scoring results, the top few documents (for example, 10 documents) are selected, and the data of these selected documents is sent to the next processing step.

[0243] Step 7:

[0244] The server creates a summary using generative AI

[0245] The server passes the abstract, results, and discussion of each selected paper to the AI ​​generator, instructing it to summarize those parts. The AI ​​generator uses natural language processing techniques to extract the key parts of the text and create a concise summary.

[0246] Step 8:

[0247] The server adjusts the summary using the emotion engine

[0248] The server uses an emotion engine to analyze the user's emotion data sent from the device. Based on the analysis results, the summary results are formatted to best suit the user's emotional state. For example, if the user is feeling anxious, the summary results are adjusted to be more concise and easy to understand.

[0249] Step 9:

[0250] The server formats and sends the summary results

[0251] The server sends the formatted summary to the user's device using the HTTP protocol.

[0252] Step 10:

[0253] The terminal displays summary data

[0254] The device analyzes the data it receives and displays it in an easy-to-understand manner. For example, if the emotion engine determines that the user is feeling stressed, it adjusts the color, font size, and layout of the summary data to make it easier to read.

[0255] This process allows users to receive appropriate information tailored to their emotional state and quickly understand the key points of complex academic literature.

[0256] Example 2

[0257] 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."

[0258] In conventional literature search systems, it is difficult for users to find the most appropriate information from a large number of search results, and users who are easily affected by emotional states (e.g., anxiety or stress) find it particularly difficult to obtain and understand the information. Furthermore, literature summaries are not provided in a format that is optimal for the user's emotional state, and information may not be conveyed appropriately.

[0259] 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 a means for a user to input search keywords, a means for a terminal to collect the user's emotions from facial expressions and voice, a means for the server to acquire relevant literature from an academic database, a means for the server to score the relevance of the literature and select the top literature, a means for the server to summarize the literature abstracts, results, and discussions using a generative AI, a means for the server to format the summarized data based on the user's emotional data, a means for the server to transmit the summarized data to the user's terminal, and a means for the terminal to display the summarized data received. This allows the user to receive appropriate information according to their emotional state and quickly understand the main points of complex academic literature.

[0260] "User" refers to a person who utilizes the system to search for and retrieve information.

[0261] "Terminal" refers to an electronic device used by a user, such as a computer, smartphone, or tablet.

[0262] "Server" refers to a computer system that receives requests from users and searches, retrieves, and processes documents in response.

[0263] "Search keywords" refer to terms or phrases that a user enters when conducting a search.

[0264] "Emotion" refers to the user's psychological state and is analyzed based on data collected from facial expressions and voice.

[0265] An "academic database" refers to a data store that stores academic papers and research materials and makes them searchable and retrievalable.

[0266] "Document" refers to academic papers, research reports, or other related materials.

[0267] "Generative AI" refers to a technology or system that uses artificial intelligence to generate literature summaries.

[0268] "Abstract" refers to the summary section of a document.

[0269] "Results" refers to the research findings section of the document.

[0270] "Discussion" refers to the part of the paper that discusses and interprets the results.

[0271] "Emotional data" refers to information about a user's psychological state collected from their facial expressions and voice.

[0272] "Formatting" refers to the process of converting the summary results into a format that is optimal for the user's emotions and level of understanding.

[0273] "Relevance scoring" refers to a method of numerically evaluating the relevance of documents based on their metadata.

[0274] This invention is a system that allows a user to input search keywords, search, retrieve, score, and summarize related documents, and provides the results to the user. It also has a function that recognizes the user's emotions and adjusts the search results and summary display. Specific embodiments are described below.

[0275] First, the user uses the device to enter a search keyword. For example, they enter the keyword "COVID-19 vaccine effectiveness" into the device's search input field and click the search button. At this time, the device's camera and microphone are activated to collect the user's facial expressions and voice. The collected data is sent to an emotion engine, which analyzes the user's emotional state.

[0276] The server then uses the search keywords received from the user to send a search query to academic databases, such as PubMed or IEEE Xplore, to retrieve a list of relevant articles, using requests via APIs.

[0277] The server then analyzes the metadata for each document in the retrieved document list and calculates a relevance score, including the document's title, author, abstract, publication date, etc. The server uses its own algorithm to calculate the relevance score and stores the result in a list.

[0278] Next, the server selects the top 10 articles based on the scoring results. The IDs, links, and metadata of the selected articles are passed to the generation AI. The generation AI analyzes the abstracts, results, and discussion sections of these articles to generate summaries. The operation of the generation AI is controlled using the following prompts:

[0279] "Please write a summary of the Abstract, Results, and Discussion sections of this article."

[0280] The generated summary is formatted by the emotion engine to best suit the user's emotional state. For example, if the user is feeling anxious, the summary will be formatted to be concise and easy to understand. This formatted summary data is then sent from the server to the user's device. Data transmission is via the HTTP protocol.

[0281] Finally, the device analyzes the received summary data and displays it in an easy-to-understand manner for the user. Based on the analysis results of the emotion engine, the font size, color, layout, etc. are adjusted. This allows the user to receive information optimized for their emotional state. For example, if a user is feeling nervous, the summary data will be displayed in visually friendly colors, font size, and layout.

[0282] This series of processes allows users to quickly understand the gist of complex academic literature. The system not only ensures accuracy of information, but also reduces the psychological burden on users. This invention will be highly acclaimed as a new form of information provision.

[0283] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0284] Step 1:

[0285] A user enters the search keyword "COVID-19 vaccine effectiveness" into their device and clicks the search button. The device collects the user's facial expressions and voice through the camera and microphone and sends them to the emotion engine. Input: Search keyword, user's facial expression data, voice data. Output: Emotion data sent to the emotion engine.

[0286] Step 2:

[0287] The server receives the user's search keywords and sends a search query to academic databases. The server retrieves a list of relevant literature from databases such as PubMed and IEEE Xplore. Input: Search keywords. Output: List of relevant literature.

[0288] Step 3:

[0289] The server processes the retrieved literature list and analyzes the metadata of each literature (title, author, abstract, publication date, etc.). The server calculates a relevance score based on these metadata. Input: literature list. Output: literature list with relevance score.

[0290] Step 4:

[0291] The server selects the top 10 documents based on the scoring results. The IDs, links, and metadata of the selected documents are passed to the next process. Input: Document list with relevance scores. Output: Document list of the top 10 documents.

[0292] Step 5:

[0293] The server passes the abstract, results, and discussion sections of the selected papers to the generation AI, which generates summaries of each section. The generation AI is prompted with the following prompt: "Please create a summary of the abstract, results, and discussion sections of this paper." Input: Top 10 papers, prompt. Output: Summarized paper data.

[0294] Step 6:

[0295] The server uses an emotion engine to format the summary results to best suit the user's emotional state. For example, if the user is feeling anxious, the summary results will be adjusted to be concise and easy to understand. Input: Summarized literature data, emotion data. Output: Formatted summary data.

[0296] Step 7:

[0297] The server sends the formatted summary data to the user's terminal using the HTTP protocol. Input: Formatted summary data. Output: Data sent to the user's terminal.

[0298] Step 8:

[0299] The device analyzes the data received and displays it in an easy-to-understand manner for the user. Based on the analysis results of the emotion engine, it adjusts the font size, color, layout, etc. displayed. Input: Received data, analysis results of the emotion engine. Output: Displayed summary data.

[0300] (Application example 2)

[0301] 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."

[0302] Conventional information retrieval systems can search for and summarize the relevant literature a user desires, but they are unable to provide information that takes into account the user's psychological state and emotions, and therefore lack the functionality to help users understand the information or reduce their psychological burden. As a result, it is difficult to provide appropriate information to users who are feeling stressed or anxious, which can potentially impair their user experience. Furthermore, because information is displayed uniformly and is not adjusted based on the user's emotions, they are unable to flexibly respond to the needs of individual users.

[0303] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0304] In this invention, the server includes: a means for a user to input search keywords; a means for the server to retrieve relevant literature from an academic database; a means for the server to score the relevance of the literature and select the top literature; a means for the server to summarize the literature abstracts, results, and discussions using a generative AI; a means for the server to send the summarized data to the user's terminal; a means for recognizing the user's emotions; a means for adjusting the summary results based on the recognized emotions; and a means for the terminal to display the summary data received. This makes it possible to provide information in an optimal form according to the user's emotions, improving the user experience and reducing the user's psychological burden.

[0305] "User" refers to a human or end-user who searches for and uses information.

[0306] A "search keyword" is a specific word or phrase that a user inputs into a terminal, and refers to input data used to search for related information.

[0307] A "server" refers to a computer system that accepts search queries and performs database searches and data analysis.

[0308] An "academic database" refers to a database that stores academic papers and research materials, including, for example, scientific and technical papers and medical literature.

[0309] "Means for scoring relevance" refers to algorithms or calculation methods used to evaluate the content of retrieved documents and assign a score to their relevance to the search keywords.

[0310] "Generative AI" refers to artificial intelligence that uses natural language processing technology to summarize literature and generate text.

[0311] An "abstract" refers to the summary section of a document, and mainly describes the background and purpose of the research.

[0312] "Results" refers to the results section of a document, which describes the results and findings of research or experiments.

[0313] "Discussion" refers to the section of the literature that considers the findings, including their interpretation and evaluation, and comparison with other literature.

[0314] "Summarization tools" refers to the ability to use generative AI to condense literature abstracts, results, and discussions, extracting key information into a concise format.

[0315] "Means for recognizing emotions" refers to technology for reading emotions from the user's facial expressions, voice, etc., and uses, for example, a camera or microphone.

[0316] "Emotion-based adjustment" refers to techniques for changing the presentation or content of information in consideration of the perceived emotions of a user.

[0317] "Terminal" refers to an interface device used by a user, including, for example, a smartphone, tablet, or computer.

[0318] The system of this invention consists of a series of processes: a user inputs search keywords, retrieves academic literature related to those keywords, uses a generative AI to create a summary, and then adjusts and displays the results based on the user's emotions. The main elements that make up the system are a user terminal, a server, an academic database, a generative AI, and an emotion recognition engine. A specific implementation of this system is described below.

[0319] A user enters search keywords into a device such as a smartphone or tablet. For example, they enter "latest technology news." The device's camera and microphone are used to collect the user's facial expressions and voice data, and emotional data is obtained. This emotional data is then analyzed using biometric authentication technology.

[0320] The server uses the received keywords to search academic databases. Examples of academic databases include databases that collect scientific and technical papers and medical literature. The server analyzes the retrieved documents and calculates a relevance score based on the metadata. As a result, documents with high scores are automatically selected.

[0321] The server then uses a generative AI to generate summaries of the selected literature. The generative AI automatically summarizes the abstracts, results, and discussions of the literature and extracts key information. The summarized data is further refined by an emotion recognition engine. Based on the user's emotional state, the summary results are formatted appropriately. For example, if the user is feeling stressed, the summary results will be displayed in a concise and easy-to-understand format.

[0322] The server then sends the adjusted summary to the user's device, where it analyzes the data and displays it in a user-friendly format, adjusting font size, color, and layout to improve visibility.

[0323] As a concrete example, consider a case where a user searches for "latest technology news" and their smartphone camera captures their stress state. If the emotion recognition engine determines that the user is stressed, the summary results will be displayed in a simpler and more understandable format.

[0324] An example of a prompt might be:

[0325] "Enter your topic of interest, such as 'latest technology news,' and start searching. Also, allow your camera to capture your facial expression data."

[0326] The system of the present invention provides appropriate information according to the user's emotions, thereby improving the user experience, making information comprehension more efficient, and reducing psychological burden.

[0327] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0328] Program processing steps

[0329] Step 1:

[0330] A user inputs search keywords into a device. The device's camera and microphone capture the user's facial expressions and voice. This input data includes, for example, the keyword "latest technology news," an image of the user's face, and a recording of their voice.

[0331] Step 2:

[0332] The device performs image processing and audio analysis to analyze the captured emotional data. Specifically, the device software uses a facial recognition algorithm to analyze facial expression data and an audio analysis algorithm to analyze tone of voice. The output is the user's current emotional state (e.g., "stressed" or "relaxed").

[0333] Step 3:

[0334] The search keywords and emotion data entered from the terminal are sent to the server, which then receives the keyword "latest technology news" and data about the user's emotional state.

[0335] Step 4:

[0336] The server searches academic databases. Using the received search keywords, the server sends a search query to multiple academic databases (e.g., news databases, paper databases, etc.). The output is a list of related literature and news articles.

[0337] Step 5:

[0338] The server scores the relevance of retrieved documents. An algorithm on the server analyzes the metadata and calculates a relevance score for each document. Documents with higher scores are ranked higher. The output of this process is a ranked list of highly relevant documents.

[0339] Step 6:

[0340] The server passes the abstracts, results, and discussions of the top-ranked articles to a generative AI, which then automatically generates summaries. The generative AI model (such as BART or T5) analyzes the articles and extracts the necessary information as a summary. The output is a summary.

[0341] Step 7:

[0342] The server uses an emotion recognition engine to adjust the summary result based on the user's emotional state. For example, if the user is stressed, the server converts the summary result into a more concise and understandable format. The output is an adjusted summary sentence.

[0343] Step 8:

[0344] The server sends the adjusted summary data to the terminal, which receives it using the HTTP protocol.

[0345] Step 9:

[0346] The terminal analyzes the received summary data and displays it in a format that is easy for the user to understand. The terminal arranges the summary data in an easy-to-read font, color, and layout, and displays the optimized summary result as the output for the user.

[0347] These steps allow users to receive information tailored to their emotional state and quickly understand the gist of complex literature or news articles.

[0348] 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.

[0349] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search<url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0350] 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.

[0351] [Second embodiment]

[0352] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0353] 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.

[0354] 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).

[0355] 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.

[0356] 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.

[0357] 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).

[0358] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0359] 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.

[0360] 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.

[0361] 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.

[0362] 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.

[0363] 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."

[0364] This invention is a system in which a user inputs search keywords, a server retrieves relevant literature from an academic database based on those keywords, creates a summary of that literature using a generative AI, and displays it on the user's terminal.

[0365] Specifically, the user operates their device to enter search keywords and click the search button. The server receives the entered search keywords and sends a search request to relevant academic databases. The server analyzes the literature data retrieved from the academic databases and selects the top literature based on relevance scores. This literature data is then used by generative AI to create abstracts, results, and discussion summaries. The summarized data is then sent from the server to the user's device, where it is displayed in an easy-to-read format.

[0366] The following will explain this with some specific examples.

[0367] Specific examples

[0368] 1. User searches for a theme

[0369] The user enters the keyword "COVID-19 vaccine effectiveness" into the device and clicks the search button.

[0370] 2. The server searches the academic database

[0371] The server sends an API request to academic databases using the keyword "COVID-19 vaccine effectiveness," such as PubMed and IEEE Xplore.

[0372] 3. The server evaluates and scores the relevance of the documents

[0373] The server analyzes the metadata of the retrieved documents and scores their relevance. For example, documents that contain many keywords such as "COVID-19," "vaccine," and "effectiveness" in the title or abstract are rated as highly relevant.

[0374] 4. Server generates summary using AI

[0375] The server uses generative AI to summarize the abstract, results, and discussion of the top 10 articles, extracting key information from each section and generating a concise summary.

[0376] 5. The server formats and sends the summary data

[0377] The server formats the summarized data into an easy-to-read format (for example, HTML format) and sends it to the user's terminal via an HTTP request.

[0378] 6. The device displays the summary data.

[0379] The device analyzes the received data and displays it in an easy-to-understand manner. For example, a page designed in card format is displayed, listing the paper title, abstract, results, and discussion in that order.

[0380] Overview of the overall system operation

[0381] This system allows users to initiate a search, and the server quickly collects relevant literature and uses generative AI to quickly create summaries, providing users with fast and accurate information. By using this system, users can easily understand the key points of academic literature, even if they lack specialized knowledge.

[0382] The processing flow will be explained below.

[0383] Step 1:

[0384] The user enters search keywords

[0385] The user operates the device, enters keywords related to the topic they want to research in the search box, and clicks the search button.

[0386] Step 2:

[0387] The server receives the search keywords.

[0388] The server receives the search keywords sent from the device, analyzes them, and normalizes them to the appropriate format (for example, converting them to lowercase and removing unnecessary spaces).

[0389] Step 3:

[0390] The server searches academic databases

[0391] The server uses the normalized keywords to send a search query to the search API of an academic database (e.g., PubMed or IEEE Xplore), which returns a list of relevant literature.

[0392] Step 4:

[0393] The server scores the relevance of the documents

[0394] The server receives the list of documents and analyzes the metadata of each document (title, abstract, author information, publication year, etc.), calculates a relevance score, and ranks documents with high scores higher.

[0395] Step 5:

[0396] The server selects the top documents

[0397] Based on the scoring results, the top few documents (for example, 10 documents) are selected, and the data of these selected documents is sent to the next processing step.

[0398] Step 6:

[0399] The server creates a summary using generative AI

[0400] The server passes the abstract, results, and discussion of each selected paper to the AI ​​generator, instructing it to summarize those parts. The AI ​​generator uses natural language processing techniques to extract the key parts of the text and create a concise summary.

[0401] Step 7:

[0402] The server formats the summary results

[0403] The server formats the summary results received from the generation AI, for example, formatting the paper title, abstract, results, and discussion sections in an easy-to-understand format (HTML or JSON format).

[0404] Step 8:

[0405] The server sends the summary results to the terminal.

[0406] The server sends the formatted summary to the user's device using the HTTP protocol.

[0407] Step 9:

[0408] The terminal displays summary data

[0409] The user's device receives and analyzes the summary data sent from the server. The analysis results are then displayed in an easy-to-understand format for the user. For example, the interface can be arranged so that the title of the paper is a clickable link, and the abstract, results, and discussion are displayed in card format.

[0410] This process allows users to quickly understand the gist of complex academic literature.

[0411] Example 1

[0412] 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."

[0413] Currently, many academic databases contain vast amounts of literature data, but there is a lack of a way to quickly find relevant literature based on specific search keywords and provide appropriate summaries. Furthermore, it is difficult for non-experts to understand the content of academic literature. To solve this problem, a system is needed that can efficiently search for and summarize literature and provide it in a user-friendly format.

[0414] 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.

[0415] In this invention, the server includes a means for a user to input search keywords, a means for retrieving relevant literature from an academic database, a means for analyzing the metadata of the literature to evaluate relevance and select top-ranked literature, a means for summarizing the abstracts, results, and discussions of the literature using a generative AI model, a means for transmitting the summarized data to the user's terminal, and a means for displaying the summarized data received by the terminal, thereby enabling the user to quickly find relevant academic literature based on the search keywords and obtain the summaries in an easy-to-understand format.

[0416] A "user" is a user who uses the system to input search keywords and obtain abstract information of academic literature.

[0417] The "server" is a computer system that processes search requests from users, retrieves literature from academic databases, creates summaries using generative AI models, and transmits the summary data to users' devices.

[0418] An "academic database" is a digital database that collects, stores, and makes searchable scientific papers and research results. Examples include PubMed and IEEE Xplore.

[0419] "Metadata" refers to information about a document, specifically data such as the title, author, abstract, and publication date.

[0420] The "relevance score" is a numerical evaluation scale that quantifies the relevance of a document based on search keywords.

[0421] A "generative AI model" is a type of artificial intelligence that generates literature summaries and other text based on input text data (prompts).

[0422] An "abstract" is a section that briefly summarizes the outline or gist of a document.

[0423] "Results" is the section that shows the results of the research described in the literature.

[0424] The "Discussion" section includes interpretations and considerations of the research results described in the literature.

[0425] "Summary data" refers to the abstracts, results, and discussions of literature summarized by generative AI models.

[0426] A "terminal" is a device such as a computer or smartphone that is used by a user to operate the device.

[0427] "Display" means to visually show information on the screen of a terminal.

[0428] This invention is a system in which a user inputs search keywords, a server retrieves relevant literature from an academic database based on those keywords, creates a summary of that literature using a generative AI, and displays it on the user's terminal.

[0429] Specifically, the system is implemented as follows.

[0430] The user operates the device to enter search keywords and clicks the search button. The device then sends a request containing the search keywords to the server via an Internet connection. The server receives this request and sends a search request to academic databases based on the specified keywords. For example, databases such as PubMed and IEEE Xplore can be used.

[0431] The server retrieves relevant literature data from academic databases, which are then stored and analyzed as metadata, including the title, abstract, author, publication date, and so on.

[0432] The server then evaluates and scores the relevance of the retrieved documents. The relevance score is calculated based on the degree of match between the search keywords and the document metadata. Documents with high relevance are ranked higher.

[0433] The server then uses a generative AI model to create summaries of the top-ranked articles, including the abstract, results, and discussion sections, extracting key information from the article and formatting it into a summary.

[0434] Once the summary is complete, the server formats the generated summary data into an easy-to-read format such as HTML, and sends the formatted data to the user's device via an HTTP request.

[0435] The device analyzes the received data and displays it in a user-friendly format, such as a card-style page displaying the paper title, abstract, results, and discussion in that order.

[0436] Specific examples

[0437] Prompt Sentence Examples

[0438] Below is a specific example of a prompt sentence. In this example, the keyword "COVID-19 vaccine effectiveness" is used.

[0439] 1. User searches for a theme

[0440] The user enters the keyword "COVID-19 vaccine effectiveness" into the device and clicks the search button.

[0441] 2. The server searches the academic database

[0442] The server sends an API request to academic databases using the keyword "COVID-19 vaccine effectiveness," such as PubMed and IEEE Xplore.

[0443] 3. The server evaluates and scores the relevance of the documents

[0444] The server analyzes the metadata of the retrieved documents and scores their relevance. For example, documents that contain many keywords such as "COVID-19," "vaccine," and "effectiveness" in the title or abstract are rated as highly relevant.

[0445] 4. Server generates summary using AI

[0446] The server uses generative AI to summarize the abstract, results, and discussion of the top 10 articles, extracting key information from each section and generating a concise summary.

[0447] 5. The server formats and sends the summary data

[0448] The server formats the summarized data into an easy-to-read format (for example, HTML format) and sends it to the user's terminal via an HTTP request.

[0449] 6. The device displays the summary data.

[0450] The device analyzes the received data and displays it in an easy-to-understand manner. For example, a page designed in card format is displayed, listing the paper title, abstract, results, and discussion in that order.

[0451] In this way, users can easily understand the key points of academic literature even if they lack specialized knowledge.

[0452] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0453] Step 1:

[0454] The user enters search keywords into the device and clicks the search button.

[0455] Specific actions: A user opens a browser or application, enters a keyword (e.g., "COVID-19 vaccine effectiveness") in the search box, and clicks the search button.

[0456] Input: Search keyword (e.g., "COVID-19 vaccine effectiveness")

[0457] Output: HTTP request sent from the user's device

[0458] Step 2:

[0459] The server analyzes the search keywords received from the user and sends a search request to the academic database.

[0460] How it works: The server receives HTTP requests, extracts search keywords, and then sends API requests to academic databases such as PubMed and IEEE Xplore.

[0461] Input: HTTP request from user device (e.g., search keyword "COVID-19 vaccine effectiveness")

[0462] Data processing: Extract search keywords and create API requests

[0463] Output: API request to academic database

[0464] Step 3:

[0465] The server analyzes literature data retrieved from academic databases and evaluates its relevance.

[0466] Specific operation: The server receives literature data (in JSON format) returned from academic databases, extracts literature metadata (title, author, abstract, publication date, etc.), and calculates a relevance score.

[0467] Input: Literature data from academic databases (JSON format)

[0468] Data processing: extracting metadata, calculating relevance scores

[0469] Output: Scored bibliography

[0470] Step 4:

[0471] The server selects the top articles based on their relevance scores and creates summaries using a generative AI model.

[0472] Specific operation: The server selects the documents with the highest relevance scores (for example, the top 10), generates a prompt for each document, and inputs the prompt into the generative AI model to generate a summary (abstract, results, and discussion).

[0473] Input: Scored bibliography

[0474] Data processing: Prompt generation, summarization using generative AI models

[0475] Output: Abstracted bibliographic data

[0476] Step 5:

[0477] The server formats the summarized bibliographic data into HTML format and sends it to the user's terminal.

[0478] What it does: It inserts the summarized data into an HTML template, formats it in an easy-to-read format, and then sends it to the user's device as an HTTP response.

[0479] Input: Abstracted bibliographic data

[0480] Data processing: Insertion into HTML template, formatting

[0481] Output: HTML summary data

[0482] Step 6:

[0483] The user's terminal analyzes the received HTML data and displays it to the user.

[0484] Specific operation: The user device receives the HTTP response from the server, parses the HTML data, and displays the data in the browser, showing the paper title, abstract, results, and discussion in an easy-to-understand format.

[0485] Input: HTTP response (HTML data) from the server

[0486] Data processing: HTML data analysis

[0487] Output: Summary information displayed to the user

[0488] (Application example 1)

[0489] 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."

[0490] In conventional virtual stores, systems that allow users to efficiently understand product reviews and support purchasing decisions have not been adequately developed. In particular, checking a huge number of reviews one by one is inefficient, and an improvement in the user experience is required.

[0491] 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.

[0492] In this invention, the server includes a means for a user to input search keywords, a means for the server to retrieve related information from a database, a means for the server to score the relevance of the information and select the top information, a means for the server to summarize the information using a generation AI, a means for the server to send the summarized data to the user's terminal, and a means for the terminal to display the received summarized data. This allows the user to efficiently understand product reviews in a virtual store and make a quick purchase decision.

[0493] "User" means a subject who uses the system to enter search keywords.

[0494] A "server" is a computer system that receives requests from users, retrieves relevant information from a database, and generates a summary of the information.

[0495] A "database" is a collection of information in which various pieces of information are stored, and is referenced by the server to obtain related information.

[0496] "Information" refers to various data such as data obtained from databases, literature, reviews, etc.

[0497] "Relevance scoring" refers to the process of numerically evaluating how relevant the retrieved information is to the user's search keywords.

[0498] "Generative AI" is a system that uses artificial intelligence technology to automatically generate summaries of information.

[0499] A "summary" is a short summary of the main points of the information obtained.

[0500] A "terminal" is a device (such as a smartphone, computer, or HMD) that allows a user to input information and receive and display summary data from a server.

[0501] This invention relates to a product review summarization system for a virtual store, which allows users to efficiently understand reviews related to a specific product and assists them in making purchasing decisions. Specific embodiments for implementing this system are described below.

[0502] First, the user operates a device (such as a smartphone or head-mounted display) to input keywords related to the product. The keywords entered by the user are sent to the server via the interface.

[0503] Based on the received keywords, the server retrieves relevant information from a database that stores reviews of virtual stores and is accessed via an API.

[0504] The server then analyzes the metadata of the retrieved information and scores its relevance. Specifically, it selects the information that is most relevant to the keywords searched by the user, taking into account the frequency and matching rate of keywords contained in the title and text.

[0505] For information selected as highly relevant, the server uses generative AI (e.g., OpenAI GPT-3 model) to generate a summary of the review content, which extracts the key points of the original review and presents them in a concise format.

[0506] Once the summary is complete, the server sends the summary data to the user's device, where it is formatted in an appropriate format (e.g., card or list) and displayed in a user-friendly way.

[0507] This system allows users to quickly grasp the main points without having to check a huge number of reviews one by one, which allows users to efficiently gather information about products and make purchasing decisions more quickly.

[0508] For example, if a user enters the keyword "new 4K TV," the server will search for reviews based on that keyword and display a summary of the key points.

[0509] Example prompt sentence:

[0510] Please summarize the following review:

[0511] "This 4K TV offers excellent picture quality, with excellent color reproduction, making it ideal for watching movies. However, the sound quality is average and additional speakers may be needed. The controls are intuitive and easy to use, although the remote can be a bit slow to respond. Overall, it's a great value at this price point."

[0512] summary:

[0513] In this way, the user can quickly understand the detailed review content of the product.

[0514] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0515] Step 1:

[0516] A user operates a terminal to input keywords related to a specific product. The input keywords are sent to the server through the user interface. The input is in text format, and the text data is passed to the server as output.

[0517] Step 2:

[0518] The server receives the received keywords and sends a search request to a database. The database used is a source of accumulated product review information. The input is a keyword, and the output is a collection of related review information.

[0519] Step 3:

[0520] The server analyzes the review information retrieved from the database and scores its relevance. Scoring is based on the frequency of keywords in the title and text. The input is the retrieved review information, and the output is a list of review information with a relevance score.

[0521] Step 4:

[0522] The server selects reviews with high relevance scores and generates summaries using a generative AI model (e.g., OpenAI GPT-3). The text data of the selected reviews is input into the AI ​​model, and the generated summary text is obtained as the output.

[0523] Step 5:

[0524] The server formats the generated summary data into an appropriate format (e.g., HTML or card format) and sends it to the user's device. The input is the summary data, and the output is the formatted summary data.

[0525] Step 6:

[0526] The terminal displays the summary data received. The display method is designed with the user's convenience in mind, for example, presenting the summary data in a card format or list format. The input is the formatted summary data, and the output is visual information that the user can see.

[0527] This allows users to efficiently understand product review information and make purchasing decisions quickly.

[0528] 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.

[0529] This invention is a system that allows a user to input search keywords, searches for, retrieves, scores, and summarizes related literature, and provides the results to the user.It also has the ability to recognize the user's emotions and adjust the search results and summary display.

[0530] Specifically, when a user enters a search keyword into a device, the server searches academic databases using that keyword to retrieve relevant literature. The retrieved literature is then relevance-scored by the server, and the top-ranking literature is selected. For the selected literature, a generative AI automatically summarizes the literature's abstract, results, and discussion. The summarized data is then formatted appropriately by an emotion engine that recognizes the user's emotions and sent to the device.

[0531] A unique feature of this system is that it recognizes the user's emotions in real time and provides information in a form that is optimal for the user's psychological state. We will explain this with some concrete examples below.

[0532] Specific examples

[0533] 1. User searches for a theme

[0534] A user enters the keyword "COVID-19 vaccine effectiveness" into their device and clicks the search button. At that time, the device's camera and microphone are used to collect emotional data from the user's facial expressions and voice. For example, emotional information such as whether the user is nervous or relaxed is collected.

[0535] 2. The server searches the academic database

[0536] The server sends a search query to academic databases using the keyword "COVID-19 vaccine effectiveness," retrieving a list of relevant literature from databases such as PubMed and IEEE Xplore.

[0537] 3. The server scores the relevance of the documents

[0538] The server analyzes the list of documents it has acquired, calculates a relevance score based on the metadata of each document, and ranks documents with high scores higher.

[0539] 4. The server selects the top documents

[0540] Based on the scoring results, the top 10 documents are selected and the next step is carried out on these documents.

[0541] 5. The server creates a summary using generative AI

[0542] The server passes the abstracts, results, and discussions of the selected papers to the AI ​​generator, which then extracts the key information from each section and generates a concise summary.

[0543] 6. The server adjusts the summary using the emotion engine.

[0544] The server uses an emotion engine to format the summary results to best suit the user's emotional state. For example, if the user is feeling anxious, the summary results will be adjusted to be more concise and easy to understand.

[0545] 7. The server sends the summary results to the terminal.

[0546] The server sends the formatted summary to the user's device using the HTTP protocol. The summary is displayed in an optimal interface based on the analysis results of the emotion engine.

[0547] 8. The terminal displays the summary data.

[0548] The device analyzes the data it receives and displays it in an easy-to-understand manner. For example, if the emotion engine determines that the user is feeling stressed, it adjusts the color, font size, and layout of the summary data to make it easier to read.

[0549] This series of processes allows users to receive appropriate information according to their emotional state, enabling them to quickly understand the main points of complex academic literature. This system is not only effective in providing accurate information, but also in reducing the psychological burden on users.

[0550] The processing flow will be explained below.

[0551] Step 1:

[0552] The user enters search keywords

[0553] The user operates the device, enters keywords related to the topic they want to research (e.g., "COVID-19 vaccine effectiveness") in the search box, and clicks the search button. At this time, the device's camera and microphone collect the user's emotional data (facial expressions and tone of voice).

[0554] Step 2:

[0555] The device sends emotion data

[0556] The device sends the collected emotion data to a server using a secure communication protocol (e.g., HTTPS).

[0557] Step 3:

[0558] The server receives the search keywords.

[0559] The server receives the search keywords sent from the device, parses them, and normalizes them to the appropriate format (e.g., converts them to lowercase, removes unnecessary spaces).

[0560] Step 4:

[0561] The server searches academic databases

[0562] The server uses the normalized keywords to send a search query to the search API of an academic database (e.g., PubMed or IEEE Xplore), which returns a list of relevant literature.

[0563] Step 5:

[0564] The server scores the relevance of the documents

[0565] The server receives the list of documents and analyzes the metadata of each document (title, abstract, author information, publication year, etc.), calculates a relevance score, and ranks documents with high scores higher.

[0566] Step 6:

[0567] The server selects the top documents

[0568] Based on the scoring results, the top few documents (for example, 10 documents) are selected, and the data of these selected documents is sent to the next processing step.

[0569] Step 7:

[0570] The server creates a summary using generative AI

[0571] The server passes the abstract, results, and discussion of each selected paper to the AI ​​generator, instructing it to summarize those parts. The AI ​​generator uses natural language processing techniques to extract the key parts of the text and create a concise summary.

[0572] Step 8:

[0573] The server adjusts the summary using the emotion engine

[0574] The server uses an emotion engine to analyze the user's emotion data sent from the device. Based on the analysis results, the summary results are formatted to best suit the user's emotional state. For example, if the user is feeling anxious, the summary results are adjusted to be more concise and easy to understand.

[0575] Step 9:

[0576] The server formats and sends the summary results

[0577] The server sends the formatted summary to the user's device using the HTTP protocol.

[0578] Step 10:

[0579] The terminal displays summary data

[0580] The device analyzes the data it receives and displays it in an easy-to-understand manner. For example, if the emotion engine determines that the user is feeling stressed, it adjusts the color, font size, and layout of the summary data to make it easier to read.

[0581] This process allows users to receive appropriate information tailored to their emotional state and quickly understand the key points of complex academic literature.

[0582] Example 2

[0583] 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."

[0584] In conventional literature search systems, it is difficult for users to find the most appropriate information from a large number of search results, and users who are easily affected by emotional states (e.g., anxiety or stress) find it particularly difficult to obtain and understand the information. Furthermore, literature summaries are not provided in a format that is optimal for the user's emotional state, and information may not be conveyed appropriately.

[0585] 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 a means for a user to input search keywords, a means for a terminal to collect the user's emotions from facial expressions and voice, a means for the server to acquire relevant literature from an academic database, a means for the server to score the relevance of the literature and select the top literature, a means for the server to summarize the literature abstracts, results, and discussions using a generative AI, a means for the server to format the summarized data based on the user's emotional data, a means for the server to transmit the summarized data to the user's terminal, and a means for the terminal to display the summarized data received. This allows the user to receive appropriate information according to their emotional state and quickly understand the main points of complex academic literature.

[0586] "User" refers to a person who utilizes the system to search for and retrieve information.

[0587] "Terminal" refers to an electronic device used by a user, such as a computer, smartphone, or tablet.

[0588] "Server" refers to a computer system that receives requests from users and searches, retrieves, and processes documents in response.

[0589] "Search keywords" refer to terms or phrases that a user enters when conducting a search.

[0590] "Emotion" refers to the user's psychological state and is analyzed based on data collected from facial expressions and voice.

[0591] An "academic database" refers to a data store that stores academic papers and research materials and makes them searchable and retrievalable.

[0592] "Document" refers to academic papers, research reports, or other related materials.

[0593] "Generative AI" refers to a technology or system that uses artificial intelligence to generate literature summaries.

[0594] "Abstract" refers to the summary section of a document.

[0595] "Results" refers to the research findings section of the document.

[0596] "Discussion" refers to the part of the paper that discusses and interprets the results.

[0597] "Emotional data" refers to information about a user's psychological state collected from their facial expressions and voice.

[0598] "Formatting" refers to the process of converting the summary results into a format that is optimal for the user's emotions and level of understanding.

[0599] "Relevance scoring" refers to a method of numerically evaluating the relevance of documents based on their metadata.

[0600] This invention is a system that allows a user to input search keywords, search, retrieve, score, and summarize related documents, and provides the results to the user. It also has a function that recognizes the user's emotions and adjusts the search results and summary display. Specific embodiments are described below.

[0601] First, the user uses the device to enter a search keyword. For example, they enter the keyword "COVID-19 vaccine effectiveness" into the device's search input field and click the search button. At this time, the device's camera and microphone are activated to collect the user's facial expressions and voice. The collected data is sent to an emotion engine, which analyzes the user's emotional state.

[0602] The server then uses the search keywords received from the user to send a search query to academic databases, such as PubMed or IEEE Xplore, to retrieve a list of relevant articles, using requests via APIs.

[0603] The server then analyzes the metadata for each document in the retrieved document list and calculates a relevance score, including the document's title, author, abstract, publication date, etc. The server uses its own algorithm to calculate the relevance score and stores the result in a list.

[0604] Next, the server selects the top 10 articles based on the scoring results. The IDs, links, and metadata of the selected articles are passed to the generation AI. The generation AI analyzes the abstracts, results, and discussion sections of these articles to generate summaries. The operation of the generation AI is controlled using the following prompts:

[0605] "Please write a summary of the Abstract, Results, and Discussion sections of this article."

[0606] The generated summary is formatted by the emotion engine to best suit the user's emotional state. For example, if the user is feeling anxious, the summary will be formatted to be concise and easy to understand. This formatted summary data is then sent from the server to the user's device. Data transmission is via the HTTP protocol.

[0607] Finally, the device analyzes the received summary data and displays it in an easy-to-understand manner for the user. Based on the analysis results of the emotion engine, the font size, color, layout, etc. are adjusted. This allows the user to receive information optimized for their emotional state. For example, if a user is feeling nervous, the summary data will be displayed in visually friendly colors, font size, and layout.

[0608] This series of processes allows users to quickly understand the gist of complex academic literature. The system not only ensures accuracy of information, but also reduces the psychological burden on users. This invention will be highly acclaimed as a new form of information provision.

[0609] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0610] Step 1:

[0611] A user enters the search keyword "COVID-19 vaccine effectiveness" into their device and clicks the search button. The device collects the user's facial expressions and voice through the camera and microphone and sends them to the emotion engine. Input: Search keyword, user's facial expression data, voice data. Output: Emotion data sent to the emotion engine.

[0612] Step 2:

[0613] The server receives the user's search keywords and sends a search query to academic databases. The server retrieves a list of relevant literature from databases such as PubMed and IEEE Xplore. Input: Search keywords. Output: List of relevant literature.

[0614] Step 3:

[0615] The server processes the retrieved literature list and analyzes the metadata of each literature (title, author, abstract, publication date, etc.). The server calculates a relevance score based on these metadata. Input: literature list. Output: literature list with relevance score.

[0616] Step 4:

[0617] The server selects the top 10 documents based on the scoring results. The IDs, links, and metadata of the selected documents are passed to the next process. Input: Document list with relevance scores. Output: Document list of the top 10 documents.

[0618] Step 5:

[0619] The server passes the abstract, results, and discussion sections of the selected papers to the generation AI, which generates summaries of each section. The generation AI is prompted with the following prompt: "Please create a summary of the abstract, results, and discussion sections of this paper." Input: Top 10 papers, prompt. Output: Summarized paper data.

[0620] Step 6:

[0621] The server uses an emotion engine to format the summary results to best suit the user's emotional state. For example, if the user is feeling anxious, the summary results will be adjusted to be concise and easy to understand. Input: Summarized literature data, emotion data. Output: Formatted summary data.

[0622] Step 7:

[0623] The server sends the formatted summary data to the user's terminal using the HTTP protocol. Input: Formatted summary data. Output: Data sent to the user's terminal.

[0624] Step 8:

[0625] The device analyzes the data received and displays it in an easy-to-understand manner for the user. Based on the analysis results of the emotion engine, it adjusts the font size, color, layout, etc. displayed. Input: Received data, analysis results of the emotion engine. Output: Displayed summary data.

[0626] (Application example 2)

[0627] 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."

[0628] Conventional information retrieval systems can search for and summarize the relevant literature a user desires, but they are unable to provide information that takes into account the user's psychological state and emotions, and therefore lack the functionality to help users understand the information or reduce their psychological burden. As a result, it is difficult to provide appropriate information to users who are feeling stressed or anxious, which can potentially impair their user experience. Furthermore, because information is displayed uniformly and is not adjusted based on the user's emotions, they are unable to flexibly respond to the needs of individual users.

[0629] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0630] In this invention, the server includes: a means for a user to input search keywords; a means for the server to retrieve relevant literature from an academic database; a means for the server to score the relevance of the literature and select the top literature; a means for the server to summarize the literature abstracts, results, and discussions using a generative AI; a means for the server to send the summarized data to the user's terminal; a means for recognizing the user's emotions; a means for adjusting the summary results based on the recognized emotions; and a means for the terminal to display the summary data received. This makes it possible to provide information in an optimal form according to the user's emotions, improving the user experience and reducing the user's psychological burden.

[0631] "User" refers to a human or end-user who searches for and uses information.

[0632] A "search keyword" is a specific word or phrase that a user inputs into a terminal, and refers to input data used to search for related information.

[0633] A "server" refers to a computer system that accepts search queries and performs database searches and data analysis.

[0634] An "academic database" refers to a database that stores academic papers and research materials, including, for example, scientific and technical papers and medical literature.

[0635] "Means for scoring relevance" refers to algorithms or calculation methods used to evaluate the content of retrieved documents and assign a score to their relevance to the search keywords.

[0636] "Generative AI" refers to artificial intelligence that uses natural language processing technology to summarize literature and generate text.

[0637] An "abstract" refers to the summary section of a document, and mainly describes the background and purpose of the research.

[0638] "Results" refers to the results section of a document, which describes the results and findings of research or experiments.

[0639] "Discussion" refers to the section of the literature that considers the findings, including their interpretation and evaluation, and comparison with other literature.

[0640] "Summarization tools" refers to the ability to use generative AI to condense literature abstracts, results, and discussions, extracting key information into a concise format.

[0641] "Means for recognizing emotions" refers to technology for reading emotions from the user's facial expressions, voice, etc., and uses, for example, a camera or microphone.

[0642] "Emotion-based adjustment" refers to techniques for changing the presentation or content of information in consideration of the perceived emotions of a user.

[0643] "Terminal" refers to an interface device used by a user, including, for example, a smartphone, tablet, or computer.

[0644] The system of this invention consists of a series of processes: a user inputs search keywords, retrieves academic literature related to those keywords, uses a generative AI to create a summary, and then adjusts and displays the results based on the user's emotions. The main elements that make up the system are a user terminal, a server, an academic database, a generative AI, and an emotion recognition engine. A specific implementation of this system is described below.

[0645] A user enters search keywords into a device such as a smartphone or tablet. For example, they enter "latest technology news." The device's camera and microphone are used to collect the user's facial expressions and voice data, and emotional data is obtained. This emotional data is then analyzed using biometric authentication technology.

[0646] The server uses the received keywords to search academic databases. Examples of academic databases include databases that collect scientific and technical papers and medical literature. The server analyzes the retrieved documents and calculates a relevance score based on the metadata. As a result, documents with high scores are automatically selected.

[0647] The server then uses a generative AI to generate summaries of the selected literature. The generative AI automatically summarizes the abstracts, results, and discussions of the literature and extracts key information. The summarized data is further refined by an emotion recognition engine. Based on the user's emotional state, the summary results are formatted appropriately. For example, if the user is feeling stressed, the summary results will be displayed in a concise and easy-to-understand format.

[0648] The server then sends the adjusted summary to the user's device, where it analyzes the data and displays it in a user-friendly format, adjusting font size, color, and layout to improve visibility.

[0649] As a concrete example, consider a case where a user searches for "latest technology news" and their smartphone camera captures their stress state. If the emotion recognition engine determines that the user is stressed, the summary results will be displayed in a simpler and more understandable format.

[0650] An example of a prompt might be:

[0651] "Enter your topic of interest, such as 'latest technology news,' and start searching. Also, allow your camera to capture your facial expression data."

[0652] The system of the present invention provides appropriate information according to the user's emotions, thereby improving the user experience, making information comprehension more efficient, and reducing psychological burden.

[0653] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0654] Program processing steps

[0655] Step 1:

[0656] A user inputs search keywords into a device. The device's camera and microphone capture the user's facial expressions and voice. This input data includes, for example, the keyword "latest technology news," an image of the user's face, and a recording of their voice.

[0657] Step 2:

[0658] The device performs image processing and audio analysis to analyze the captured emotional data. Specifically, the device software uses a facial recognition algorithm to analyze facial expression data and an audio analysis algorithm to analyze tone of voice. The output is the user's current emotional state (e.g., "stressed" or "relaxed").

[0659] Step 3:

[0660] The search keywords and emotion data entered from the terminal are sent to the server, which then receives the keyword "latest technology news" and data about the user's emotional state.

[0661] Step 4:

[0662] The server searches academic databases. Using the received search keywords, the server sends a search query to multiple academic databases (e.g., news databases, paper databases, etc.). The output is a list of related literature and news articles.

[0663] Step 5:

[0664] The server scores the relevance of retrieved documents. An algorithm on the server analyzes the metadata and calculates a relevance score for each document. Documents with higher scores are ranked higher. The output of this process is a ranked list of highly relevant documents.

[0665] Step 6:

[0666] The server passes the abstracts, results, and discussions of the top-ranked articles to a generative AI, which then automatically generates summaries. The generative AI model (such as BART or T5) analyzes the articles and extracts the necessary information as a summary. The output is a summary.

[0667] Step 7:

[0668] The server uses an emotion recognition engine to adjust the summary result based on the user's emotional state. For example, if the user is stressed, the server converts the summary result into a more concise and understandable format. The output is an adjusted summary sentence.

[0669] Step 8:

[0670] The server sends the adjusted summary data to the terminal, which receives it using the HTTP protocol.

[0671] Step 9:

[0672] The terminal analyzes the received summary data and displays it in a format that is easy for the user to understand. The terminal arranges the summary data in an easy-to-read font, color, and layout, and displays the optimized summary result as the output for the user.

[0673] These steps allow users to receive information tailored to their emotional state and quickly understand the gist of complex literature or news articles.

[0674] 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.

[0675] 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.

[0676] 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.

[0677] [Third embodiment]

[0678] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0679] 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.

[0680] 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).

[0681] 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.

[0682] 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.

[0683] 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).

[0684] 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.

[0685] 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.

[0686] 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.

[0687] 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.

[0688] 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.

[0689] 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."

[0690] This invention is a system in which a user inputs search keywords, a server retrieves relevant literature from an academic database based on those keywords, creates a summary of that literature using a generative AI, and displays it on the user's terminal.

[0691] Specifically, the user operates their device to enter search keywords and click the search button. The server receives the entered search keywords and sends a search request to relevant academic databases. The server analyzes the literature data retrieved from the academic databases and selects the top literature based on relevance scores. This literature data is then used by generative AI to create abstracts, results, and discussion summaries. The summarized data is then sent from the server to the user's device, where it is displayed in an easy-to-read format.

[0692] The following will explain this with some specific examples.

[0693] Specific examples

[0694] 1. User searches for a theme

[0695] The user enters the keyword "COVID-19 vaccine effectiveness" into the device and clicks the search button.

[0696] 2. The server searches the academic database

[0697] The server sends an API request to academic databases using the keyword "COVID-19 vaccine effectiveness," such as PubMed and IEEE Xplore.

[0698] 3. The server evaluates and scores the relevance of the documents

[0699] The server analyzes the metadata of the retrieved documents and scores their relevance. For example, documents that contain many keywords such as "COVID-19," "vaccine," and "effectiveness" in the title or abstract are rated as highly relevant.

[0700] 4. Server generates summary using AI

[0701] The server uses generative AI to summarize the abstract, results, and discussion of the top 10 articles, extracting key information from each section and generating a concise summary.

[0702] 5. The server formats and sends the summary data

[0703] The server formats the summarized data into an easy-to-read format (for example, HTML format) and sends it to the user's terminal via an HTTP request.

[0704] 6. The device displays the summary data.

[0705] The device analyzes the received data and displays it in an easy-to-understand manner. For example, a page designed in card format is displayed, listing the paper title, abstract, results, and discussion in that order.

[0706] Overview of the overall system operation

[0707] This system allows users to initiate a search, and the server quickly collects relevant literature and uses generative AI to quickly create summaries, providing users with fast and accurate information. By using this system, users can easily understand the key points of academic literature, even if they lack specialized knowledge.

[0708] The processing flow will be explained below.

[0709] Step 1:

[0710] The user enters search keywords

[0711] The user operates the device, enters keywords related to the topic they want to research in the search box, and clicks the search button.

[0712] Step 2:

[0713] The server receives the search keywords.

[0714] The server receives the search keywords sent from the device, analyzes them, and normalizes them to the appropriate format (for example, converting them to lowercase and removing unnecessary spaces).

[0715] Step 3:

[0716] The server searches academic databases

[0717] The server uses the normalized keywords to send a search query to the search API of an academic database (e.g., PubMed or IEEE Xplore), which returns a list of relevant literature.

[0718] Step 4:

[0719] The server scores the relevance of the documents

[0720] The server receives the list of documents and analyzes the metadata of each document (title, abstract, author information, publication year, etc.), calculates a relevance score, and ranks documents with high scores higher.

[0721] Step 5:

[0722] The server selects the top documents

[0723] Based on the scoring results, the top few documents (for example, 10 documents) are selected, and the data of these selected documents is sent to the next processing step.

[0724] Step 6:

[0725] The server creates a summary using generative AI

[0726] The server passes the abstract, results, and discussion of each selected paper to the AI ​​generator, instructing it to summarize those parts. The AI ​​generator uses natural language processing techniques to extract the key parts of the text and create a concise summary.

[0727] Step 7:

[0728] The server formats the summary results

[0729] The server formats the summary results received from the generation AI, for example, formatting the paper title, abstract, results, and discussion sections in an easy-to-understand format (HTML or JSON format).

[0730] Step 8:

[0731] The server sends the summary results to the terminal.

[0732] The server sends the formatted summary to the user's device using the HTTP protocol.

[0733] Step 9:

[0734] The terminal displays summary data

[0735] The user's device receives and analyzes the summary data sent from the server. The analysis results are then displayed in an easy-to-understand format for the user. For example, the interface can be arranged so that the title of the paper is a clickable link, and the abstract, results, and discussion are displayed in card format.

[0736] This process allows users to quickly understand the gist of complex academic literature.

[0737] Example 1

[0738] 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."

[0739] Currently, many academic databases contain vast amounts of literature data, but there is a lack of a way to quickly find relevant literature based on specific search keywords and provide appropriate summaries. Furthermore, it is difficult for non-experts to understand the content of academic literature. To solve this problem, a system is needed that can efficiently search for and summarize literature and provide it in a user-friendly format.

[0740] 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.

[0741] In this invention, the server includes a means for a user to input search keywords, a means for retrieving relevant literature from an academic database, a means for analyzing the metadata of the literature to evaluate relevance and select top-ranked literature, a means for summarizing the abstracts, results, and discussions of the literature using a generative AI model, a means for transmitting the summarized data to the user's terminal, and a means for displaying the summarized data received by the terminal, thereby enabling the user to quickly find relevant academic literature based on the search keywords and obtain the summaries in an easy-to-understand format.

[0742] A "user" is a user who uses the system to input search keywords and obtain abstract information of academic literature.

[0743] The "server" is a computer system that processes search requests from users, retrieves literature from academic databases, creates summaries using generative AI models, and transmits the summary data to users' devices.

[0744] An "academic database" is a digital database that collects, stores, and makes searchable scientific papers and research results. Examples include PubMed and IEEE Xplore.

[0745] "Metadata" refers to information about a document, specifically data such as the title, author, abstract, and publication date.

[0746] The "relevance score" is a numerical evaluation scale that quantifies the relevance of a document based on search keywords.

[0747] A "generative AI model" is a type of artificial intelligence that generates literature summaries and other text based on input text data (prompts).

[0748] An "abstract" is a section that briefly summarizes the outline or gist of a document.

[0749] "Results" is the section that shows the results of the research described in the literature.

[0750] The "Discussion" section includes interpretations and considerations of the research results described in the literature.

[0751] "Summary data" refers to the abstracts, results, and discussions of literature summarized by generative AI models.

[0752] A "terminal" is a device such as a computer or smartphone that is used by a user to operate the device.

[0753] "Display" means to visually show information on the screen of a terminal.

[0754] This invention is a system in which a user inputs search keywords, a server retrieves relevant literature from an academic database based on those keywords, creates a summary of that literature using a generative AI, and displays it on the user's terminal.

[0755] Specifically, the system is implemented as follows.

[0756] The user operates the device to enter search keywords and clicks the search button. The device then sends a request containing the search keywords to the server via an Internet connection. The server receives this request and sends a search request to academic databases based on the specified keywords. For example, databases such as PubMed and IEEE Xplore can be used.

[0757] The server retrieves relevant literature data from academic databases, which are then stored and analyzed as metadata, including the title, abstract, author, publication date, and so on.

[0758] The server then evaluates and scores the relevance of the retrieved documents. The relevance score is calculated based on the degree of match between the search keywords and the document metadata. Documents with high relevance are ranked higher.

[0759] The server then uses a generative AI model to create summaries of the top-ranked articles, including the abstract, results, and discussion sections, extracting key information from the article and formatting it into a summary.

[0760] Once the summary is complete, the server formats the generated summary data into an easy-to-read format such as HTML, and sends the formatted data to the user's device via an HTTP request.

[0761] The device analyzes the received data and displays it in a user-friendly format, such as a card-style page displaying the paper title, abstract, results, and discussion in that order.

[0762] Specific examples

[0763] Prompt Sentence Examples

[0764] Below is a specific example of a prompt sentence. In this example, the keyword "COVID-19 vaccine effectiveness" is used.

[0765] 1. User searches for a theme

[0766] The user enters the keyword "COVID-19 vaccine effectiveness" into the device and clicks the search button.

[0767] 2. The server searches the academic database

[0768] The server sends an API request to academic databases using the keyword "COVID-19 vaccine effectiveness," such as PubMed and IEEE Xplore.

[0769] 3. The server evaluates and scores the relevance of the documents

[0770] The server analyzes the metadata of the retrieved documents and scores their relevance. For example, documents that contain many keywords such as "COVID-19," "vaccine," and "effectiveness" in the title or abstract are rated as highly relevant.

[0771] 4. Server generates summary using AI

[0772] The server uses generative AI to summarize the abstract, results, and discussion of the top 10 articles, extracting key information from each section and generating a concise summary.

[0773] 5. The server formats and sends the summary data

[0774] The server formats the summarized data into an easy-to-read format (for example, HTML format) and sends it to the user's terminal via an HTTP request.

[0775] 6. The device displays the summary data.

[0776] The device analyzes the received data and displays it in an easy-to-understand manner. For example, a page designed in card format is displayed, listing the paper title, abstract, results, and discussion in that order.

[0777] In this way, users can easily understand the key points of academic literature even if they lack specialized knowledge.

[0778] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0779] Step 1:

[0780] The user enters search keywords into the device and clicks the search button.

[0781] Specific actions: A user opens a browser or application, enters a keyword (e.g., "COVID-19 vaccine effectiveness") in the search box, and clicks the search button.

[0782] Input: Search keyword (e.g., "COVID-19 vaccine effectiveness")

[0783] Output: HTTP request sent from the user's device

[0784] Step 2:

[0785] The server analyzes the search keywords received from the user and sends a search request to the academic database.

[0786] How it works: The server receives HTTP requests, extracts search keywords, and then sends API requests to academic databases such as PubMed and IEEE Xplore.

[0787] Input: HTTP request from user device (e.g., search keyword "COVID-19 vaccine effectiveness")

[0788] Data processing: Extract search keywords and create API requests

[0789] Output: API request to academic database

[0790] Step 3:

[0791] The server analyzes literature data retrieved from academic databases and evaluates its relevance.

[0792] Specific operation: The server receives literature data (in JSON format) returned from academic databases, extracts literature metadata (title, author, abstract, publication date, etc.), and calculates a relevance score.

[0793] Input: Literature data from academic databases (JSON format)

[0794] Data processing: extracting metadata, calculating relevance scores

[0795] Output: Scored bibliography

[0796] Step 4:

[0797] The server selects the top articles based on their relevance scores and creates summaries using a generative AI model.

[0798] Specific operation: The server selects the documents with the highest relevance scores (for example, the top 10), generates a prompt for each document, and inputs the prompt into the generative AI model to generate a summary (abstract, results, and discussion).

[0799] Input: Scored bibliography

[0800] Data processing: Prompt generation, summarization using generative AI models

[0801] Output: Abstracted bibliographic data

[0802] Step 5:

[0803] The server formats the summarized bibliographic data into HTML format and sends it to the user's terminal.

[0804] What it does: It inserts the summarized data into an HTML template, formats it in an easy-to-read format, and then sends it to the user's device as an HTTP response.

[0805] Input: Abstracted bibliographic data

[0806] Data processing: Insertion into HTML template, formatting

[0807] Output: HTML summary data

[0808] Step 6:

[0809] The user's terminal analyzes the received HTML data and displays it to the user.

[0810] Specific operation: The user device receives the HTTP response from the server, parses the HTML data, and displays the data in the browser, showing the paper title, abstract, results, and discussion in an easy-to-understand format.

[0811] Input: HTTP response (HTML data) from the server

[0812] Data processing: HTML data analysis

[0813] Output: Summary information displayed to the user

[0814] (Application example 1)

[0815] 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."

[0816] In conventional virtual stores, systems that allow users to efficiently understand product reviews and support purchasing decisions have not been adequately developed. In particular, checking a huge number of reviews one by one is inefficient, and an improvement in the user experience is required.

[0817] 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.

[0818] In this invention, the server includes a means for a user to input search keywords, a means for the server to retrieve related information from a database, a means for the server to score the relevance of the information and select the top information, a means for the server to summarize the information using a generation AI, a means for the server to send the summarized data to the user's terminal, and a means for the terminal to display the received summarized data. This allows the user to efficiently understand product reviews in a virtual store and make a quick purchase decision.

[0819] "User" means a subject who uses the system to enter search keywords.

[0820] A "server" is a computer system that receives requests from users, retrieves relevant information from a database, and generates a summary of the information.

[0821] A "database" is a collection of information in which various pieces of information are stored, and is referenced by the server to obtain related information.

[0822] "Information" refers to various data such as data obtained from databases, literature, reviews, etc.

[0823] "Relevance scoring" refers to the process of numerically evaluating how relevant the retrieved information is to the user's search keywords.

[0824] "Generative AI" is a system that uses artificial intelligence technology to automatically generate summaries of information.

[0825] A "summary" is a short summary of the main points of the information obtained.

[0826] A "terminal" is a device (such as a smartphone, computer, or HMD) that allows a user to input information and receive and display summary data from a server.

[0827] This invention relates to a product review summarization system for a virtual store, which allows users to efficiently understand reviews related to a specific product and assists them in making purchasing decisions. Specific embodiments for implementing this system are described below.

[0828] First, the user operates a device (such as a smartphone or head-mounted display) to input keywords related to the product. The keywords entered by the user are sent to the server via the interface.

[0829] Based on the received keywords, the server retrieves relevant information from a database that stores reviews of virtual stores and is accessed via an API.

[0830] The server then analyzes the metadata of the retrieved information and scores its relevance. Specifically, it selects the information that is most relevant to the keywords searched by the user, taking into account the frequency and matching rate of keywords contained in the title and text.

[0831] For information selected as highly relevant, the server uses generative AI (e.g., OpenAI GPT-3 model) to generate a summary of the review content, which extracts the key points of the original review and presents them in a concise format.

[0832] Once the summary is complete, the server sends the summary data to the user's device, where it is formatted in an appropriate format (e.g., card or list) and displayed in a user-friendly way.

[0833] This system allows users to quickly grasp the main points without having to check a huge number of reviews one by one, which allows users to efficiently gather information about products and make purchasing decisions more quickly.

[0834] For example, if a user enters the keyword "new 4K TV," the server will search for reviews based on that keyword and display a summary of the key points.

[0835] Example prompt sentence:

[0836] Please summarize the following review:

[0837] "This 4K TV offers excellent picture quality, with excellent color reproduction, making it ideal for watching movies. However, the sound quality is average and additional speakers may be needed. The controls are intuitive and easy to use, although the remote can be a bit slow to respond. Overall, it's a great value at this price point."

[0838] summary:

[0839] In this way, the user can quickly understand the detailed review content of the product.

[0840] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0841] Step 1:

[0842] A user operates a terminal to input keywords related to a specific product. The input keywords are sent to the server through the user interface. The input is in text format, and the text data is passed to the server as output.

[0843] Step 2:

[0844] The server receives the received keywords and sends a search request to a database. The database used is a source of accumulated product review information. The input is a keyword, and the output is a collection of related review information.

[0845] Step 3:

[0846] The server analyzes the review information retrieved from the database and scores its relevance. Scoring is based on the frequency of keywords in the title and text. The input is the retrieved review information, and the output is a list of review information with a relevance score.

[0847] Step 4:

[0848] The server selects reviews with high relevance scores and generates summaries using a generative AI model (e.g., OpenAI GPT-3). The text data of the selected reviews is input into the AI ​​model, and the generated summary text is obtained as the output.

[0849] Step 5:

[0850] The server formats the generated summary data into an appropriate format (e.g., HTML or card format) and sends it to the user's device. The input is the summary data, and the output is the formatted summary data.

[0851] Step 6:

[0852] The terminal displays the summary data received. The display method is designed with the user's convenience in mind, for example, presenting the summary data in a card format or list format. The input is the formatted summary data, and the output is visual information that the user can see.

[0853] This allows users to efficiently understand product review information and make purchasing decisions quickly.

[0854] 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.

[0855] This invention is a system that allows a user to input search keywords, searches for, retrieves, scores, and summarizes related literature, and provides the results to the user.It also has the ability to recognize the user's emotions and adjust the search results and summary display.

[0856] Specifically, when a user enters a search keyword into a device, the server searches academic databases using that keyword to retrieve relevant literature. The retrieved literature is then relevance-scored by the server, and the top-ranking literature is selected. For the selected literature, a generative AI automatically summarizes the literature's abstract, results, and discussion. The summarized data is then formatted appropriately by an emotion engine that recognizes the user's emotions and sent to the device.

[0857] A unique feature of this system is that it recognizes the user's emotions in real time and provides information in a form that is optimal for the user's psychological state. We will explain this with some concrete examples below.

[0858] Specific examples

[0859] 1. User searches for a theme

[0860] A user enters the keyword "COVID-19 vaccine effectiveness" into their device and clicks the search button. At that time, the device's camera and microphone are used to collect emotional data from the user's facial expressions and voice. For example, emotional information such as whether the user is nervous or relaxed is collected.

[0861] 2. The server searches the academic database

[0862] The server sends a search query to academic databases using the keyword "COVID-19 vaccine effectiveness," retrieving a list of relevant literature from databases such as PubMed and IEEE Xplore.

[0863] 3. The server scores the relevance of the documents

[0864] The server analyzes the list of documents it has acquired, calculates a relevance score based on the metadata of each document, and ranks documents with high scores higher.

[0865] 4. The server selects the top documents

[0866] Based on the scoring results, the top 10 documents are selected and the next step is carried out on these documents.

[0867] 5. The server creates a summary using generative AI

[0868] The server passes the abstracts, results, and discussions of the selected papers to the AI ​​generator, which then extracts the key information from each section and generates a concise summary.

[0869] 6. The server adjusts the summary using the emotion engine.

[0870] The server uses an emotion engine to format the summary results to best suit the user's emotional state. For example, if the user is feeling anxious, the summary results will be adjusted to be more concise and easy to understand.

[0871] 7. The server sends the summary results to the terminal.

[0872] The server sends the formatted summary to the user's device using the HTTP protocol. The summary is displayed in an optimal interface based on the analysis results of the emotion engine.

[0873] 8. The terminal displays the summary data.

[0874] The device analyzes the data it receives and displays it in an easy-to-understand manner. For example, if the emotion engine determines that the user is feeling stressed, it adjusts the color, font size, and layout of the summary data to make it easier to read.

[0875] This series of processes allows users to receive appropriate information according to their emotional state, enabling them to quickly understand the main points of complex academic literature. This system is not only effective in providing accurate information, but also in reducing the psychological burden on users.

[0876] The processing flow will be explained below.

[0877] Step 1:

[0878] The user enters search keywords

[0879] The user operates the device, enters keywords related to the topic they want to research (e.g., "COVID-19 vaccine effectiveness") in the search box, and clicks the search button. At this time, the device's camera and microphone collect the user's emotional data (facial expressions and tone of voice).

[0880] Step 2:

[0881] The device sends emotion data

[0882] The device sends the collected emotion data to a server using a secure communication protocol (e.g., HTTPS).

[0883] Step 3:

[0884] The server receives the search keywords.

[0885] The server receives the search keywords sent from the device, parses them, and normalizes them to the appropriate format (e.g., converts them to lowercase, removes unnecessary spaces).

[0886] Step 4:

[0887] The server searches academic databases

[0888] The server uses the normalized keywords to send a search query to the search API of an academic database (e.g., PubMed or IEEE Xplore), which returns a list of relevant literature.

[0889] Step 5:

[0890] The server scores the relevance of the documents

[0891] The server receives the list of documents and analyzes the metadata of each document (title, abstract, author information, publication year, etc.), calculates a relevance score, and ranks documents with high scores higher.

[0892] Step 6:

[0893] The server selects the top documents

[0894] Based on the scoring results, the top few documents (for example, 10 documents) are selected, and the data of these selected documents is sent to the next processing step.

[0895] Step 7:

[0896] The server creates a summary using generative AI

[0897] The server passes the abstract, results, and discussion of each selected paper to the AI ​​generator, instructing it to summarize those parts. The AI ​​generator uses natural language processing techniques to extract the key parts of the text and create a concise summary.

[0898] Step 8:

[0899] The server adjusts the summary using the emotion engine

[0900] The server uses an emotion engine to analyze the user's emotion data sent from the device. Based on the analysis results, the summary results are formatted to best suit the user's emotional state. For example, if the user is feeling anxious, the summary results are adjusted to be more concise and easy to understand.

[0901] Step 9:

[0902] The server formats and sends the summary results

[0903] The server sends the formatted summary to the user's device using the HTTP protocol.

[0904] Step 10:

[0905] The terminal displays summary data

[0906] The device analyzes the data it receives and displays it in an easy-to-understand manner. For example, if the emotion engine determines that the user is feeling stressed, it adjusts the color, font size, and layout of the summary data to make it easier to read.

[0907] This process allows users to receive appropriate information tailored to their emotional state and quickly understand the key points of complex academic literature.

[0908] Example 2

[0909] 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."

[0910] In conventional literature search systems, it is difficult for users to find the most appropriate information from a large number of search results, and users who are easily affected by emotional states (e.g., anxiety or stress) find it particularly difficult to obtain and understand the information. Furthermore, literature summaries are not provided in a format that is optimal for the user's emotional state, and information may not be conveyed appropriately.

[0911] 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 a means for a user to input search keywords, a means for a terminal to collect the user's emotions from facial expressions and voice, a means for the server to acquire relevant literature from an academic database, a means for the server to score the relevance of the literature and select the top literature, a means for the server to summarize the literature abstracts, results, and discussions using a generative AI, a means for the server to format the summarized data based on the user's emotional data, a means for the server to transmit the summarized data to the user's terminal, and a means for the terminal to display the summarized data received. This allows the user to receive appropriate information according to their emotional state and quickly understand the main points of complex academic literature.

[0912] "User" refers to a person who utilizes the system to search for and retrieve information.

[0913] "Terminal" refers to an electronic device used by a user, such as a computer, smartphone, or tablet.

[0914] "Server" refers to a computer system that receives requests from users and searches, retrieves, and processes documents in response.

[0915] "Search keywords" refer to terms or phrases that a user enters when conducting a search.

[0916] "Emotion" refers to the user's psychological state and is analyzed based on data collected from facial expressions and voice.

[0917] An "academic database" refers to a data store that stores academic papers and research materials and makes them searchable and retrievalable.

[0918] "Document" refers to academic papers, research reports, or other related materials.

[0919] "Generative AI" refers to a technology or system that uses artificial intelligence to generate literature summaries.

[0920] "Abstract" refers to the summary section of a document.

[0921] "Results" refers to the research findings section of the document.

[0922] "Discussion" refers to the part of the paper that discusses and interprets the results.

[0923] "Emotional data" refers to information about a user's psychological state collected from their facial expressions and voice.

[0924] "Formatting" refers to the process of converting the summary results into a format that is optimal for the user's emotions and level of understanding.

[0925] "Relevance scoring" refers to a method of numerically evaluating the relevance of documents based on their metadata.

[0926] This invention is a system that allows a user to input search keywords, search, retrieve, score, and summarize related documents, and provides the results to the user. It also has a function that recognizes the user's emotions and adjusts the search results and summary display. Specific embodiments are described below.

[0927] First, the user uses the device to enter a search keyword. For example, they enter the keyword "COVID-19 vaccine effectiveness" into the device's search input field and click the search button. At this time, the device's camera and microphone are activated to collect the user's facial expressions and voice. The collected data is sent to an emotion engine, which analyzes the user's emotional state.

[0928] The server then uses the search keywords received from the user to send a search query to academic databases, such as PubMed or IEEE Xplore, to retrieve a list of relevant articles, using requests via APIs.

[0929] The server then analyzes the metadata for each document in the retrieved document list and calculates a relevance score, including the document's title, author, abstract, publication date, etc. The server uses its own algorithm to calculate the relevance score and stores the result in a list.

[0930] Next, the server selects the top 10 articles based on the scoring results. The IDs, links, and metadata of the selected articles are passed to the generation AI. The generation AI analyzes the abstracts, results, and discussion sections of these articles to generate summaries. The operation of the generation AI is controlled using the following prompts:

[0931] "Please write a summary of the Abstract, Results, and Discussion sections of this article."

[0932] The generated summary is formatted by the emotion engine to best suit the user's emotional state. For example, if the user is feeling anxious, the summary will be formatted to be concise and easy to understand. This formatted summary data is then sent from the server to the user's device. Data transmission is via the HTTP protocol.

[0933] Finally, the device analyzes the received summary data and displays it in an easy-to-understand manner for the user. Based on the analysis results of the emotion engine, the font size, color, layout, etc. are adjusted. This allows the user to receive information optimized for their emotional state. For example, if a user is feeling nervous, the summary data will be displayed in visually friendly colors, font size, and layout.

[0934] This series of processes allows users to quickly understand the gist of complex academic literature. The system not only ensures accuracy of information, but also reduces the psychological burden on users. This invention will be highly acclaimed as a new form of information provision.

[0935] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0936] Step 1:

[0937] A user enters the search keyword "COVID-19 vaccine effectiveness" into their device and clicks the search button. The device collects the user's facial expressions and voice through the camera and microphone and sends them to the emotion engine. Input: Search keyword, user's facial expression data, voice data. Output: Emotion data sent to the emotion engine.

[0938] Step 2:

[0939] The server receives the user's search keywords and sends a search query to academic databases. The server retrieves a list of relevant literature from databases such as PubMed and IEEE Xplore. Input: Search keywords. Output: List of relevant literature.

[0940] Step 3:

[0941] The server processes the retrieved literature list and analyzes the metadata of each literature (title, author, abstract, publication date, etc.). The server calculates a relevance score based on these metadata. Input: literature list. Output: literature list with relevance score.

[0942] Step 4:

[0943] The server selects the top 10 documents based on the scoring results. The IDs, links, and metadata of the selected documents are passed to the next process. Input: Document list with relevance scores. Output: Document list of the top 10 documents.

[0944] Step 5:

[0945] The server passes the abstract, results, and discussion sections of the selected papers to the generation AI, which generates summaries of each section. The generation AI is prompted with the following prompt: "Please create a summary of the abstract, results, and discussion sections of this paper." Input: Top 10 papers, prompt. Output: Summarized paper data.

[0946] Step 6:

[0947] The server uses an emotion engine to format the summary results to best suit the user's emotional state. For example, if the user is feeling anxious, the summary results will be adjusted to be concise and easy to understand. Input: Summarized literature data, emotion data. Output: Formatted summary data.

[0948] Step 7:

[0949] The server sends the formatted summary data to the user's terminal using the HTTP protocol. Input: Formatted summary data. Output: Data sent to the user's terminal.

[0950] Step 8:

[0951] The device analyzes the data received and displays it in an easy-to-understand manner for the user. Based on the analysis results of the emotion engine, it adjusts the font size, color, layout, etc. displayed. Input: Received data, analysis results of the emotion engine. Output: Displayed summary data.

[0952] (Application example 2)

[0953] 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."

[0954] Conventional information retrieval systems can search for and summarize the relevant literature a user desires, but they are unable to provide information that takes into account the user's psychological state and emotions, and therefore lack the functionality to help users understand the information or reduce their psychological burden. As a result, it is difficult to provide appropriate information to users who are feeling stressed or anxious, which can potentially impair their user experience. Furthermore, because information is displayed uniformly and is not adjusted based on the user's emotions, they are unable to flexibly respond to the needs of individual users.

[0955] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0956] In this invention, the server includes: a means for a user to input search keywords; a means for the server to retrieve relevant literature from an academic database; a means for the server to score the relevance of the literature and select the top literature; a means for the server to summarize the literature abstracts, results, and discussions using a generative AI; a means for the server to send the summarized data to the user's terminal; a means for recognizing the user's emotions; a means for adjusting the summary results based on the recognized emotions; and a means for the terminal to display the summary data received. This makes it possible to provide information in an optimal form according to the user's emotions, improving the user experience and reducing the user's psychological burden.

[0957] "User" refers to a human or end-user who searches for and uses information.

[0958] A "search keyword" is a specific word or phrase that a user inputs into a terminal, and refers to input data used to search for related information.

[0959] A "server" refers to a computer system that accepts search queries and performs database searches and data analysis.

[0960] An "academic database" refers to a database that stores academic papers and research materials, including, for example, scientific and technical papers and medical literature.

[0961] "Means for scoring relevance" refers to algorithms or calculation methods used to evaluate the content of retrieved documents and assign a score to their relevance to the search keywords.

[0962] "Generative AI" refers to artificial intelligence that uses natural language processing technology to summarize literature and generate text.

[0963] An "abstract" refers to the summary section of a document, and mainly describes the background and purpose of the research.

[0964] "Results" refers to the results section of a document, which describes the results and findings of research or experiments.

[0965] "Discussion" refers to the section of the literature that considers the findings, including their interpretation and evaluation, and comparison with other literature.

[0966] "Summarization tools" refers to the ability to use generative AI to condense literature abstracts, results, and discussions, extracting key information into a concise format.

[0967] "Means for recognizing emotions" refers to technology for reading emotions from the user's facial expressions, voice, etc., and uses, for example, a camera or microphone.

[0968] "Emotion-based adjustment" refers to techniques for changing the presentation or content of information in consideration of the perceived emotions of a user.

[0969] "Terminal" refers to an interface device used by a user, including, for example, a smartphone, tablet, or computer.

[0970] The system of this invention consists of a series of processes: a user inputs search keywords, retrieves academic literature related to those keywords, uses a generative AI to create a summary, and then adjusts and displays the results based on the user's emotions. The main elements that make up the system are a user terminal, a server, an academic database, a generative AI, and an emotion recognition engine. A specific implementation of this system is described below.

[0971] A user enters search keywords into a device such as a smartphone or tablet. For example, they enter "latest technology news." The device's camera and microphone are used to collect the user's facial expressions and voice data, and emotional data is obtained. This emotional data is then analyzed using biometric authentication technology.

[0972] The server uses the received keywords to search academic databases. Examples of academic databases include databases that collect scientific and technical papers and medical literature. The server analyzes the retrieved documents and calculates a relevance score based on the metadata. As a result, documents with high scores are automatically selected.

[0973] The server then uses a generative AI to generate summaries of the selected literature. The generative AI automatically summarizes the abstracts, results, and discussions of the literature and extracts key information. The summarized data is further refined by an emotion recognition engine. Based on the user's emotional state, the summary results are formatted appropriately. For example, if the user is feeling stressed, the summary results will be displayed in a concise and easy-to-understand format.

[0974] The server then sends the adjusted summary to the user's device, where it analyzes the data and displays it in a user-friendly format, adjusting font size, color, and layout to improve visibility.

[0975] As a concrete example, consider a case where a user searches for "latest technology news" and their smartphone camera captures their stress state. If the emotion recognition engine determines that the user is stressed, the summary results will be displayed in a simpler and more understandable format.

[0976] An example of a prompt might be:

[0977] "Enter your topic of interest, such as 'latest technology news,' and start searching. Also, allow your camera to capture your facial expression data."

[0978] The system of the present invention provides appropriate information according to the user's emotions, thereby improving the user experience, making information comprehension more efficient, and reducing psychological burden.

[0979] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0980] Program processing steps

[0981] Step 1:

[0982] A user inputs search keywords into a device. The device's camera and microphone capture the user's facial expressions and voice. This input data includes, for example, the keyword "latest technology news," an image of the user's face, and a recording of their voice.

[0983] Step 2:

[0984] The device performs image processing and audio analysis to analyze the captured emotional data. Specifically, the device software uses a facial recognition algorithm to analyze facial expression data and an audio analysis algorithm to analyze tone of voice. The output is the user's current emotional state (e.g., "stressed" or "relaxed").

[0985] Step 3:

[0986] The search keywords and emotion data entered from the terminal are sent to the server, which then receives the keyword "latest technology news" and data about the user's emotional state.

[0987] Step 4:

[0988] The server searches academic databases. Using the received search keywords, the server sends a search query to multiple academic databases (e.g., news databases, paper databases, etc.). The output is a list of related literature and news articles.

[0989] Step 5:

[0990] The server scores the relevance of retrieved documents. An algorithm on the server analyzes the metadata and calculates a relevance score for each document. Documents with higher scores are ranked higher. The output of this process is a ranked list of highly relevant documents.

[0991] Step 6:

[0992] The server passes the abstracts, results, and discussions of the top-ranked articles to a generative AI, which then automatically generates summaries. The generative AI model (such as BART or T5) analyzes the articles and extracts the necessary information as a summary. The output is a summary.

[0993] Step 7:

[0994] The server uses an emotion recognition engine to adjust the summary result based on the user's emotional state. For example, if the user is stressed, the server converts the summary result into a more concise and understandable format. The output is an adjusted summary sentence.

[0995] Step 8:

[0996] The server sends the adjusted summary data to the terminal, which receives it using the HTTP protocol.

[0997] Step 9:

[0998] The terminal analyzes the received summary data and displays it in a format that is easy for the user to understand. The terminal arranges the summary data in an easy-to-read font, color, and layout, and displays the optimized summary result as the output for the user.

[0999] These steps allow users to receive information tailored to their emotional state and quickly understand the gist of complex literature or news articles.

[1000] 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.

[1001] 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.

[1002] 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.

[1003] [Fourth embodiment]

[1004] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1005] 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.

[1006] 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).

[1007] 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.

[1008] 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.

[1009] 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).

[1010] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1011] 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.

[1012] 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.

[1013] 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.

[1014] 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.

[1015] 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.

[1016] 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."

[1017] This invention is a system in which a user inputs search keywords, a server retrieves relevant literature from an academic database based on those keywords, creates a summary of that literature using a generative AI, and displays it on the user's terminal.

[1018] Specifically, the user operates their device to enter search keywords and click the search button. The server receives the entered search keywords and sends a search request to relevant academic databases. The server analyzes the literature data retrieved from the academic databases and selects the top literature based on relevance scores. This literature data is then used by generative AI to create abstracts, results, and discussion summaries. The summarized data is then sent from the server to the user's device, where it is displayed in an easy-to-read format.

[1019] The following will explain this with some specific examples.

[1020] Specific examples

[1021] 1. User searches for a theme

[1022] The user enters the keyword "COVID-19 vaccine effectiveness" into the device and clicks the search button.

[1023] 2. The server searches the academic database

[1024] The server sends an API request to academic databases using the keyword "COVID-19 vaccine effectiveness," such as PubMed and IEEE Xplore.

[1025] 3. The server evaluates and scores the relevance of the documents

[1026] The server analyzes the metadata of the retrieved documents and scores their relevance. For example, documents that contain many keywords such as "COVID-19," "vaccine," and "effectiveness" in the title or abstract are rated as highly relevant.

[1027] 4. Server generates summary using AI

[1028] The server uses generative AI to summarize the abstract, results, and discussion of the top 10 articles, extracting key information from each section and generating a concise summary.

[1029] 5. The server formats and sends the summary data

[1030] The server formats the summarized data into an easy-to-read format (for example, HTML format) and sends it to the user's terminal via an HTTP request.

[1031] 6. The device displays the summary data.

[1032] The device analyzes the received data and displays it in an easy-to-understand manner. For example, a page designed in card format is displayed, listing the paper title, abstract, results, and discussion in that order.

[1033] Overview of the overall system operation

[1034] This system allows users to initiate a search, and the server quickly collects relevant literature and uses generative AI to quickly create summaries, providing users with fast and accurate information. By using this system, users can easily understand the key points of academic literature, even if they lack specialized knowledge.

[1035] The processing flow will be explained below.

[1036] Step 1:

[1037] The user enters search keywords

[1038] The user operates the device, enters keywords related to the topic they want to research in the search box, and clicks the search button.

[1039] Step 2:

[1040] The server receives the search keywords.

[1041] The server receives the search keywords sent from the device, analyzes them, and normalizes them to the appropriate format (for example, converting them to lowercase and removing unnecessary spaces).

[1042] Step 3:

[1043] The server searches academic databases

[1044] The server uses the normalized keywords to send a search query to the search API of an academic database (e.g., PubMed or IEEE Xplore), which returns a list of relevant literature.

[1045] Step 4:

[1046] The server scores the relevance of the documents

[1047] The server receives the list of documents and analyzes the metadata of each document (title, abstract, author information, publication year, etc.), calculates a relevance score, and ranks documents with high scores higher.

[1048] Step 5:

[1049] The server selects the top documents

[1050] Based on the scoring results, the top few documents (for example, 10 documents) are selected, and the data of these selected documents is sent to the next processing step.

[1051] Step 6:

[1052] The server creates a summary using generative AI

[1053] The server passes the abstract, results, and discussion of each selected paper to the AI ​​generator, instructing it to summarize those parts. The AI ​​generator uses natural language processing techniques to extract the key parts of the text and create a concise summary.

[1054] Step 7:

[1055] The server formats the summary results

[1056] The server formats the summary results received from the generation AI, for example, formatting the paper title, abstract, results, and discussion sections in an easy-to-understand format (HTML or JSON format).

[1057] Step 8:

[1058] The server sends the summary results to the terminal.

[1059] The server sends the formatted summary to the user's device using the HTTP protocol.

[1060] Step 9:

[1061] The terminal displays summary data

[1062] The user's device receives and analyzes the summary data sent from the server. The analysis results are then displayed in an easy-to-understand format for the user. For example, the interface can be arranged so that the title of the paper is a clickable link, and the abstract, results, and discussion are displayed in card format.

[1063] This process allows users to quickly understand the gist of complex academic literature.

[1064] Example 1

[1065] 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."

[1066] Currently, many academic databases contain vast amounts of literature data, but there is a lack of a way to quickly find relevant literature based on specific search keywords and provide appropriate summaries. Furthermore, it is difficult for non-experts to understand the content of academic literature. To solve this problem, a system is needed that can efficiently search for and summarize literature and provide it in a user-friendly format.

[1067] 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.

[1068] In this invention, the server includes a means for a user to input search keywords, a means for retrieving relevant literature from an academic database, a means for analyzing the metadata of the literature to evaluate relevance and select top-ranked literature, a means for summarizing the abstracts, results, and discussions of the literature using a generative AI model, a means for transmitting the summarized data to the user's terminal, and a means for displaying the summarized data received by the terminal, thereby enabling the user to quickly find relevant academic literature based on the search keywords and obtain the summaries in an easy-to-understand format.

[1069] A "user" is a user who uses the system to input search keywords and obtain abstract information of academic literature.

[1070] The "server" is a computer system that processes search requests from users, retrieves literature from academic databases, creates summaries using generative AI models, and transmits the summary data to users' devices.

[1071] An "academic database" is a digital database that collects, stores, and makes searchable scientific papers and research results. Examples include PubMed and IEEE Xplore.

[1072] "Metadata" refers to information about a document, specifically data such as the title, author, abstract, and publication date.

[1073] The "relevance score" is a numerical evaluation scale that quantifies the relevance of a document based on search keywords.

[1074] A "generative AI model" is a type of artificial intelligence that generates literature summaries and other text based on input text data (prompts).

[1075] An "abstract" is a section that briefly summarizes the outline or gist of a document.

[1076] "Results" is the section that shows the results of the research described in the literature.

[1077] The "Discussion" section includes interpretations and considerations of the research results described in the literature.

[1078] "Summary data" refers to the abstracts, results, and discussions of literature summarized by generative AI models.

[1079] A "terminal" is a device such as a computer or smartphone that is used by a user to operate the device.

[1080] "Display" means to visually show information on the screen of a terminal.

[1081] This invention is a system in which a user inputs search keywords, a server retrieves relevant literature from an academic database based on those keywords, creates a summary of that literature using a generative AI, and displays it on the user's terminal.

[1082] Specifically, the system is implemented as follows.

[1083] The user operates the device to enter search keywords and clicks the search button. The device then sends a request containing the search keywords to the server via an Internet connection. The server receives this request and sends a search request to academic databases based on the specified keywords. For example, databases such as PubMed and IEEE Xplore can be used.

[1084] The server retrieves relevant literature data from academic databases, which are then stored and analyzed as metadata, including the title, abstract, author, publication date, and so on.

[1085] The server then evaluates and scores the relevance of the retrieved documents. The relevance score is calculated based on the degree of match between the search keywords and the document metadata. Documents with high relevance are ranked higher.

[1086] The server then uses a generative AI model to create summaries of the top-ranked articles, including the abstract, results, and discussion sections, extracting key information from the article and formatting it into a summary.

[1087] Once the summary is complete, the server formats the generated summary data into an easy-to-read format such as HTML, and sends the formatted data to the user's device via an HTTP request.

[1088] The device analyzes the received data and displays it in a user-friendly format, such as a card-style page displaying the paper title, abstract, results, and discussion in that order.

[1089] Specific examples

[1090] Prompt Sentence Examples

[1091] Below is a specific example of a prompt sentence. In this example, the keyword "COVID-19 vaccine effectiveness" is used.

[1092] 1. User searches for a theme

[1093] The user enters the keyword "COVID-19 vaccine effectiveness" into the device and clicks the search button.

[1094] 2. The server searches the academic database

[1095] The server sends an API request to academic databases using the keyword "COVID-19 vaccine effectiveness," such as PubMed and IEEE Xplore.

[1096] 3. The server evaluates and scores the relevance of the documents

[1097] The server analyzes the metadata of the retrieved documents and scores their relevance. For example, documents that contain many keywords such as "COVID-19," "vaccine," and "effectiveness" in the title or abstract are rated as highly relevant.

[1098] 4. Server generates summary using AI

[1099] The server uses generative AI to summarize the abstract, results, and discussion of the top 10 articles, extracting key information from each section and generating a concise summary.

[1100] 5. The server formats and sends the summary data

[1101] The server formats the summarized data into an easy-to-read format (for example, HTML format) and sends it to the user's terminal via an HTTP request.

[1102] 6. The device displays the summary data.

[1103] The device analyzes the received data and displays it in an easy-to-understand manner. For example, a page designed in card format is displayed, listing the paper title, abstract, results, and discussion in that order.

[1104] In this way, users can easily understand the key points of academic literature even if they lack specialized knowledge.

[1105] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1106] Step 1:

[1107] The user enters search keywords into the device and clicks the search button.

[1108] Specific actions: A user opens a browser or application, enters a keyword (e.g., "COVID-19 vaccine effectiveness") in the search box, and clicks the search button.

[1109] Input: Search keyword (e.g., "COVID-19 vaccine effectiveness")

[1110] Output: HTTP request sent from the user's device

[1111] Step 2:

[1112] The server analyzes the search keywords received from the user and sends a search request to the academic database.

[1113] How it works: The server receives HTTP requests, extracts search keywords, and then sends API requests to academic databases such as PubMed and IEEE Xplore.

[1114] Input: HTTP request from user device (e.g., search keyword "COVID-19 vaccine effectiveness")

[1115] Data processing: Extract search keywords and create API requests

[1116] Output: API request to academic database

[1117] Step 3:

[1118] The server analyzes literature data retrieved from academic databases and evaluates its relevance.

[1119] Specific operation: The server receives literature data (in JSON format) returned from academic databases, extracts literature metadata (title, author, abstract, publication date, etc.), and calculates a relevance score.

[1120] Input: Literature data from academic databases (JSON format)

[1121] Data processing: extracting metadata, calculating relevance scores

[1122] Output: Scored bibliography

[1123] Step 4:

[1124] The server selects the top articles based on their relevance scores and creates summaries using a generative AI model.

[1125] Specific operation: The server selects the documents with the highest relevance scores (for example, the top 10), generates a prompt for each document, and inputs the prompt into the generative AI model to generate a summary (abstract, results, and discussion).

[1126] Input: Scored bibliography

[1127] Data processing: Prompt generation, summarization using generative AI models

[1128] Output: Abstracted bibliographic data

[1129] Step 5:

[1130] The server formats the summarized bibliographic data into HTML format and sends it to the user's terminal.

[1131] What it does: It inserts the summarized data into an HTML template, formats it in an easy-to-read format, and then sends it to the user's device as an HTTP response.

[1132] Input: Abstracted bibliographic data

[1133] Data processing: Insertion into HTML template, formatting

[1134] Output: HTML summary data

[1135] Step 6:

[1136] The user's terminal analyzes the received HTML data and displays it to the user.

[1137] Specific operation: The user device receives the HTTP response from the server, parses the HTML data, and displays the data in the browser, showing the paper title, abstract, results, and discussion in an easy-to-understand format.

[1138] Input: HTTP response (HTML data) from the server

[1139] Data processing: HTML data analysis

[1140] Output: Summary information displayed to the user

[1141] (Application example 1)

[1142] 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."

[1143] In conventional virtual stores, systems that allow users to efficiently understand product reviews and support purchasing decisions have not been adequately developed. In particular, checking a huge number of reviews one by one is inefficient, and an improvement in the user experience is required.

[1144] 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.

[1145] In this invention, the server includes a means for a user to input search keywords, a means for the server to retrieve related information from a database, a means for the server to score the relevance of the information and select the top information, a means for the server to summarize the information using a generation AI, a means for the server to send the summarized data to the user's terminal, and a means for the terminal to display the received summarized data. This allows the user to efficiently understand product reviews in a virtual store and make a quick purchase decision.

[1146] "User" means a subject who uses the system to enter search keywords.

[1147] A "server" is a computer system that receives requests from users, retrieves relevant information from a database, and generates a summary of the information.

[1148] A "database" is a collection of information in which various pieces of information are stored, and is referenced by the server to obtain related information.

[1149] "Information" refers to various data such as data obtained from databases, literature, reviews, etc.

[1150] "Relevance scoring" refers to the process of numerically evaluating how relevant the retrieved information is to the user's search keywords.

[1151] "Generative AI" is a system that uses artificial intelligence technology to automatically generate summaries of information.

[1152] A "summary" is a short summary of the main points of the information obtained.

[1153] A "terminal" is a device (such as a smartphone, computer, or HMD) that allows a user to input information and receive and display summary data from a server.

[1154] This invention relates to a product review summarization system for a virtual store, which allows users to efficiently understand reviews related to a specific product and assists them in making purchasing decisions. Specific embodiments for implementing this system are described below.

[1155] First, the user operates a device (such as a smartphone or head-mounted display) to input keywords related to the product. The keywords entered by the user are sent to the server via the interface.

[1156] Based on the received keywords, the server retrieves relevant information from a database that stores reviews of virtual stores and is accessed via an API.

[1157] The server then analyzes the metadata of the retrieved information and scores its relevance. Specifically, it selects the information that is most relevant to the keywords searched by the user, taking into account the frequency and matching rate of keywords contained in the title and text.

[1158] For information selected as highly relevant, the server uses generative AI (e.g., OpenAI GPT-3 model) to generate a summary of the review content, which extracts the key points of the original review and presents them in a concise format.

[1159] Once the summary is complete, the server sends the summary data to the user's device, where it is formatted in an appropriate format (e.g., card or list) and displayed in a user-friendly way.

[1160] This system allows users to quickly grasp the main points without having to check a huge number of reviews one by one, which allows users to efficiently gather information about products and make purchasing decisions more quickly.

[1161] For example, if a user enters the keyword "new 4K TV," the server will search for reviews based on that keyword and display a summary of the key points.

[1162] Example prompt sentence:

[1163] Please summarize the following review:

[1164] "This 4K TV offers excellent picture quality, with excellent color reproduction, making it ideal for watching movies. However, the sound quality is average and additional speakers may be needed. The controls are intuitive and easy to use, although the remote can be a bit slow to respond. Overall, it's a great value at this price point."

[1165] summary:

[1166] In this way, the user can quickly understand the detailed review content of the product.

[1167] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1168] Step 1:

[1169] A user operates a terminal to input keywords related to a specific product. The input keywords are sent to the server through the user interface. The input is in text format, and the text data is passed to the server as output.

[1170] Step 2:

[1171] The server receives the received keywords and sends a search request to a database. The database used is a source of accumulated product review information. The input is a keyword, and the output is a collection of related review information.

[1172] Step 3:

[1173] The server analyzes the review information retrieved from the database and scores its relevance. Scoring is based on the frequency of keywords in the title and text. The input is the retrieved review information, and the output is a list of review information with a relevance score.

[1174] Step 4:

[1175] The server selects reviews with high relevance scores and generates summaries using a generative AI model (e.g., OpenAI GPT-3). The text data of the selected reviews is input into the AI ​​model, and the generated summary text is obtained as the output.

[1176] Step 5:

[1177] The server formats the generated summary data into an appropriate format (e.g., HTML or card format) and sends it to the user's device. The input is the summary data, and the output is the formatted summary data.

[1178] Step 6:

[1179] The terminal displays the summary data received. The display method is designed with the user's convenience in mind, for example, presenting the summary data in a card format or list format. The input is the formatted summary data, and the output is visual information that the user can see.

[1180] This allows users to efficiently understand product review information and make purchasing decisions quickly.

[1181] 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.

[1182] This invention is a system that allows a user to input search keywords, searches for, retrieves, scores, and summarizes related literature, and provides the results to the user.It also has the ability to recognize the user's emotions and adjust the search results and summary display.

[1183] Specifically, when a user enters a search keyword into a device, the server searches academic databases using that keyword to retrieve relevant literature. The retrieved literature is then relevance-scored by the server, and the top-ranking literature is selected. For the selected literature, a generative AI automatically summarizes the literature's abstract, results, and discussion. The summarized data is then formatted appropriately by an emotion engine that recognizes the user's emotions and sent to the device.

[1184] A unique feature of this system is that it recognizes the user's emotions in real time and provides information in a form that is optimal for the user's psychological state. We will explain this with some concrete examples below.

[1185] Specific examples

[1186] 1. User searches for a theme

[1187] A user enters the keyword "COVID-19 vaccine effectiveness" into their device and clicks the search button. At that time, the device's camera and microphone are used to collect emotional data from the user's facial expressions and voice. For example, emotional information such as whether the user is nervous or relaxed is collected.

[1188] 2. The server searches the academic database

[1189] The server sends a search query to academic databases using the keyword "COVID-19 vaccine effectiveness," retrieving a list of relevant literature from databases such as PubMed and IEEE Xplore.

[1190] 3. The server scores the relevance of the documents

[1191] The server analyzes the list of documents it has acquired, calculates a relevance score based on the metadata of each document, and ranks documents with high scores higher.

[1192] 4. The server selects the top documents

[1193] Based on the scoring results, the top 10 documents are selected and the next step is carried out on these documents.

[1194] 5. The server creates a summary using generative AI

[1195] The server passes the abstracts, results, and discussions of the selected papers to the AI ​​generator, which then extracts the key information from each section and generates a concise summary.

[1196] 6. The server adjusts the summary using the emotion engine.

[1197] The server uses an emotion engine to format the summary results to best suit the user's emotional state. For example, if the user is feeling anxious, the summary results will be adjusted to be more concise and easy to understand.

[1198] 7. The server sends the summary results to the terminal.

[1199] The server sends the formatted summary to the user's device using the HTTP protocol. The summary is displayed in an optimal interface based on the analysis results of the emotion engine.

[1200] 8. The terminal displays the summary data.

[1201] The device analyzes the data it receives and displays it in an easy-to-understand manner. For example, if the emotion engine determines that the user is feeling stressed, it adjusts the color, font size, and layout of the summary data to make it easier to read.

[1202] This series of processes allows users to receive appropriate information according to their emotional state, enabling them to quickly understand the main points of complex academic literature. This system is not only effective in providing accurate information, but also in reducing the psychological burden on users.

[1203] The processing flow will be explained below.

[1204] Step 1:

[1205] The user enters search keywords

[1206] The user operates the device, enters keywords related to the topic they want to research (e.g., "COVID-19 vaccine effectiveness") in the search box, and clicks the search button. At this time, the device's camera and microphone collect the user's emotional data (facial expressions and tone of voice).

[1207] Step 2:

[1208] The device sends emotion data

[1209] The device sends the collected emotion data to a server using a secure communication protocol (e.g., HTTPS).

[1210] Step 3:

[1211] The server receives the search keywords.

[1212] The server receives the search keywords sent from the device, parses them, and normalizes them to the appropriate format (e.g., converts them to lowercase, removes unnecessary spaces).

[1213] Step 4:

[1214] The server searches academic databases

[1215] The server uses the normalized keywords to send a search query to the search API of an academic database (e.g., PubMed or IEEE Xplore), which returns a list of relevant literature.

[1216] Step 5:

[1217] The server scores the relevance of the documents

[1218] The server receives the list of documents and analyzes the metadata of each document (title, abstract, author information, publication year, etc.), calculates a relevance score, and ranks documents with high scores higher.

[1219] Step 6:

[1220] The server selects the top documents

[1221] Based on the scoring results, the top few documents (for example, 10 documents) are selected, and the data of these selected documents is sent to the next processing step.

[1222] Step 7:

[1223] The server creates a summary using generative AI

[1224] The server passes the abstract, results, and discussion of each selected paper to the AI ​​generator, instructing it to summarize those parts. The AI ​​generator uses natural language processing techniques to extract the key parts of the text and create a concise summary.

[1225] Step 8:

[1226] The server adjusts the summary using the emotion engine

[1227] The server uses an emotion engine to analyze the user's emotion data sent from the device. Based on the analysis results, the summary results are formatted to best suit the user's emotional state. For example, if the user is feeling anxious, the summary results are adjusted to be more concise and easy to understand.

[1228] Step 9:

[1229] The server formats and sends the summary results

[1230] The server sends the formatted summary to the user's device using the HTTP protocol.

[1231] Step 10:

[1232] The terminal displays summary data

[1233] The device analyzes the data it receives and displays it in an easy-to-understand manner. For example, if the emotion engine determines that the user is feeling stressed, it adjusts the color, font size, and layout of the summary data to make it easier to read.

[1234] This process allows users to receive appropriate information tailored to their emotional state and quickly understand the key points of complex academic literature.

[1235] Example 2

[1236] 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."

[1237] In conventional literature search systems, it is difficult for users to find the most appropriate information from a large number of search results, and users who are easily affected by emotional states (e.g., anxiety or stress) find it particularly difficult to obtain and understand the information. Furthermore, literature summaries are not provided in a format that is optimal for the user's emotional state, and information may not be conveyed appropriately.

[1238] 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 a means for a user to input search keywords, a means for a terminal to collect the user's emotions from facial expressions and voice, a means for the server to acquire relevant literature from an academic database, a means for the server to score the relevance of the literature and select the top literature, a means for the server to summarize the literature abstracts, results, and discussions using a generative AI, a means for the server to format the summarized data based on the user's emotional data, a means for the server to transmit the summarized data to the user's terminal, and a means for the terminal to display the summarized data received. This allows the user to receive appropriate information according to their emotional state and quickly understand the main points of complex academic literature.

[1239] "User" refers to a person who utilizes the system to search for and retrieve information.

[1240] "Terminal" refers to an electronic device used by a user, such as a computer, smartphone, or tablet.

[1241] "Server" refers to a computer system that receives requests from users and searches, retrieves, and processes documents in response.

[1242] "Search keywords" refer to terms or phrases that a user enters when conducting a search.

[1243] "Emotion" refers to the user's psychological state and is analyzed based on data collected from facial expressions and voice.

[1244] An "academic database" refers to a data store that stores academic papers and research materials and makes them searchable and retrievalable.

[1245] "Document" refers to academic papers, research reports, or other related materials.

[1246] "Generative AI" refers to a technology or system that uses artificial intelligence to generate literature summaries.

[1247] "Abstract" refers to the summary section of a document.

[1248] "Results" refers to the research findings section of the document.

[1249] "Discussion" refers to the part of the paper that discusses and interprets the results.

[1250] "Emotional data" refers to information about a user's psychological state collected from their facial expressions and voice.

[1251] "Formatting" refers to the process of converting the summary results into a format that is optimal for the user's emotions and level of understanding.

[1252] "Relevance scoring" refers to a method of numerically evaluating the relevance of documents based on their metadata.

[1253] This invention is a system that allows a user to input search keywords, search, retrieve, score, and summarize related documents, and provides the results to the user. It also has a function that recognizes the user's emotions and adjusts the search results and summary display. Specific embodiments are described below.

[1254] First, the user uses the device to enter a search keyword. For example, they enter the keyword "COVID-19 vaccine effectiveness" into the device's search input field and click the search button. At this time, the device's camera and microphone are activated to collect the user's facial expressions and voice. The collected data is sent to an emotion engine, which analyzes the user's emotional state.

[1255] The server then uses the search keywords received from the user to send a search query to academic databases, such as PubMed or IEEE Xplore, to retrieve a list of relevant articles, using requests via APIs.

[1256] The server then analyzes the metadata for each document in the retrieved document list and calculates a relevance score, including the document's title, author, abstract, publication date, etc. The server uses its own algorithm to calculate the relevance score and stores the result in a list.

[1257] Next, the server selects the top 10 articles based on the scoring results. The IDs, links, and metadata of the selected articles are passed to the generation AI. The generation AI analyzes the abstracts, results, and discussion sections of these articles to generate summaries. The operation of the generation AI is controlled using the following prompts:

[1258] "Please write a summary of the Abstract, Results, and Discussion sections of this article."

[1259] The generated summary is formatted by the emotion engine to best suit the user's emotional state. For example, if the user is feeling anxious, the summary will be formatted to be concise and easy to understand. This formatted summary data is then sent from the server to the user's device. Data transmission is via the HTTP protocol.

[1260] Finally, the device analyzes the received summary data and displays it in an easy-to-understand manner for the user. Based on the analysis results of the emotion engine, the font size, color, layout, etc. are adjusted. This allows the user to receive information optimized for their emotional state. For example, if a user is feeling nervous, the summary data will be displayed in visually friendly colors, font size, and layout.

[1261] This series of processes allows users to quickly understand the gist of complex academic literature. The system not only ensures accuracy of information, but also reduces the psychological burden on users. This invention will be highly acclaimed as a new form of information provision.

[1262] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1263] Step 1:

[1264] A user enters the search keyword "COVID-19 vaccine effectiveness" into their device and clicks the search button. The device collects the user's facial expressions and voice through the camera and microphone and sends them to the emotion engine. Input: Search keyword, user's facial expression data, voice data. Output: Emotion data sent to the emotion engine.

[1265] Step 2:

[1266] The server receives the user's search keywords and sends a search query to academic databases. The server retrieves a list of relevant literature from databases such as PubMed and IEEE Xplore. Input: Search keywords. Output: List of relevant literature.

[1267] Step 3:

[1268] The server processes the retrieved literature list and analyzes the metadata of each literature (title, author, abstract, publication date, etc.). The server calculates a relevance score based on these metadata. Input: literature list. Output: literature list with relevance score.

[1269] Step 4:

[1270] The server selects the top 10 documents based on the scoring results. The IDs, links, and metadata of the selected documents are passed to the next process. Input: Document list with relevance scores. Output: Document list of the top 10 documents.

[1271] Step 5:

[1272] The server passes the abstract, results, and discussion sections of the selected papers to the generation AI, which generates summaries of each section. The generation AI is prompted with the following prompt: "Please create a summary of the abstract, results, and discussion sections of this paper." Input: Top 10 papers, prompt. Output: Summarized paper data.

[1273] Step 6:

[1274] The server uses an emotion engine to format the summary results to best suit the user's emotional state. For example, if the user is feeling anxious, the summary results will be adjusted to be concise and easy to understand. Input: Summarized literature data, emotion data. Output: Formatted summary data.

[1275] Step 7:

[1276] The server sends the formatted summary data to the user's terminal using the HTTP protocol. Input: Formatted summary data. Output: Data sent to the user's terminal.

[1277] Step 8:

[1278] The device analyzes the data received and displays it in an easy-to-understand manner for the user. Based on the analysis results of the emotion engine, it adjusts the font size, color, layout, etc. displayed. Input: Received data, analysis results of the emotion engine. Output: Displayed summary data.

[1279] (Application example 2)

[1280] 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."

[1281] Conventional information retrieval systems can search for and summarize the relevant literature a user desires, but they are unable to provide information that takes into account the user's psychological state and emotions, and therefore lack the functionality to help users understand the information or reduce their psychological burden. As a result, it is difficult to provide appropriate information to users who are feeling stressed or anxious, which can potentially impair their user experience. Furthermore, because information is displayed uniformly and is not adjusted based on the user's emotions, they are unable to flexibly respond to the needs of individual users.

[1282] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1283] In this invention, the server includes: a means for a user to input search keywords; a means for the server to retrieve relevant literature from an academic database; a means for the server to score the relevance of the literature and select the top literature; a means for the server to summarize the literature abstracts, results, and discussions using a generative AI; a means for the server to send the summarized data to the user's terminal; a means for recognizing the user's emotions; a means for adjusting the summary results based on the recognized emotions; and a means for the terminal to display the summary data received. This makes it possible to provide information in an optimal form according to the user's emotions, improving the user experience and reducing the user's psychological burden.

[1284] "User" refers to a human or end-user who searches for and uses information.

[1285] A "search keyword" is a specific word or phrase that a user inputs into a terminal, and refers to input data used to search for related information.

[1286] A "server" refers to a computer system that accepts search queries and performs database searches and data analysis.

[1287] An "academic database" refers to a database that stores academic papers and research materials, including, for example, scientific and technical papers and medical literature.

[1288] "Means for scoring relevance" refers to algorithms or calculation methods used to evaluate the content of retrieved documents and assign a score to their relevance to the search keywords.

[1289] "Generative AI" refers to artificial intelligence that uses natural language processing technology to summarize literature and generate text.

[1290] An "abstract" refers to the summary section of a document, and mainly describes the background and purpose of the research.

[1291] "Results" refers to the results section of a document, which describes the results and findings of research or experiments.

[1292] "Discussion" refers to the section of the literature that considers the findings, including their interpretation and evaluation, and comparison with other literature.

[1293] "Summarization tools" refers to the ability to use generative AI to condense literature abstracts, results, and discussions, extracting key information into a concise format.

[1294] "Means for recognizing emotions" refers to technology for reading emotions from the user's facial expressions, voice, etc., and uses, for example, a camera or microphone.

[1295] "Emotion-based adjustment" refers to techniques for changing the presentation or content of information in consideration of the perceived emotions of a user.

[1296] "Terminal" refers to an interface device used by a user, including, for example, a smartphone, tablet, or computer.

[1297] The system of this invention consists of a series of processes: a user inputs search keywords, retrieves academic literature related to those keywords, uses a generative AI to create a summary, and then adjusts and displays the results based on the user's emotions. The main elements that make up the system are a user terminal, a server, an academic database, a generative AI, and an emotion recognition engine. A specific implementation of this system is described below.

[1298] A user enters search keywords into a device such as a smartphone or tablet. For example, they enter "latest technology news." The device's camera and microphone are used to collect the user's facial expressions and voice data, and emotional data is obtained. This emotional data is then analyzed using biometric authentication technology.

[1299] The server uses the received keywords to search academic databases. Examples of academic databases include databases that collect scientific and technical papers and medical literature. The server analyzes the retrieved documents and calculates a relevance score based on the metadata. As a result, documents with high scores are automatically selected.

[1300] The server then uses a generative AI to generate summaries of the selected literature. The generative AI automatically summarizes the abstracts, results, and discussions of the literature and extracts key information. The summarized data is further refined by an emotion recognition engine. Based on the user's emotional state, the summary results are formatted appropriately. For example, if the user is feeling stressed, the summary results will be displayed in a concise and easy-to-understand format.

[1301] The server then sends the adjusted summary to the user's device, where it analyzes the data and displays it in a user-friendly format, adjusting font size, color, and layout to improve visibility.

[1302] As a concrete example, consider a case where a user searches for "latest technology news" and their smartphone camera captures their stress state. If the emotion recognition engine determines that the user is stressed, the summary results will be displayed in a simpler and more understandable format.

[1303] An example of a prompt might be:

[1304] "Enter your topic of interest, such as 'latest technology news,' and start searching. Also, allow your camera to capture your facial expression data."

[1305] The system of the present invention provides appropriate information according to the user's emotions, thereby improving the user experience, making information comprehension more efficient, and reducing psychological burden.

[1306] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1307] Program processing steps

[1308] Step 1:

[1309] A user inputs search keywords into a device. The device's camera and microphone capture the user's facial expressions and voice. This input data includes, for example, the keyword "latest technology news," an image of the user's face, and a recording of their voice.

[1310] Step 2:

[1311] The device performs image processing and audio analysis to analyze the captured emotional data. Specifically, the device software uses a facial recognition algorithm to analyze facial expression data and an audio analysis algorithm to analyze tone of voice. The output is the user's current emotional state (e.g., "stressed" or "relaxed").

[1312] Step 3:

[1313] The search keywords and emotion data entered from the terminal are sent to the server, which then receives the keyword "latest technology news" and data about the user's emotional state.

[1314] Step 4:

[1315] The server searches academic databases. Using the received search keywords, the server sends a search query to multiple academic databases (e.g., news databases, paper databases, etc.). The output is a list of related literature and news articles.

[1316] Step 5:

[1317] The server scores the relevance of retrieved documents. An algorithm on the server analyzes the metadata and calculates a relevance score for each document. Documents with higher scores are ranked higher. The output of this process is a ranked list of highly relevant documents.

[1318] Step 6:

[1319] The server passes the abstracts, results, and discussions of the top-ranked articles to a generative AI, which then automatically generates summaries. The generative AI model (such as BART or T5) analyzes the articles and extracts the necessary information as a summary. The output is a summary.

[1320] Step 7:

[1321] The server uses an emotion recognition engine to adjust the summary result based on the user's emotional state. For example, if the user is stressed, the server converts the summary result into a more concise and understandable format. The output is an adjusted summary sentence.

[1322] Step 8:

[1323] The server sends the adjusted summary data to the terminal, which receives it using the HTTP protocol.

[1324] Step 9:

[1325] The terminal analyzes the received summary data and displays it in a format that is easy for the user to understand. The terminal arranges the summary data in an easy-to-read font, color, and layout, and displays the optimized summary result as the output for the user.

[1326] These steps allow users to receive information tailored to their emotional state and quickly understand the gist of complex literature or news articles.

[1327] 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.

[1328] 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.

[1329] 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.

[1330] 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.

[1331] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1332] 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.

[1333] 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).

[1334] 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.

[1335] 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."

[1336] 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.

[1337] 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).

[1338] 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.

[1339] 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.

[1340] 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.

[1341] 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.

[1342] 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.

[1343] 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.

[1344] 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.

[1345] 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.

[1346] 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.

[1347] 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.

[1348] The following is further disclosed regarding the above embodiment.

[1349] (Claim 1)

[1350] a means for a user to input search keywords;

[1351] A means for the server to retrieve relevant documents from an academic database;

[1352] a means for the server to score the relevance of documents and select the top documents;

[1353] A server-generated AI-based method for summarizing abstracts, results, and discussions of literature;

[1354] means for transmitting the summarized data by the server to a user's terminal;

[1355] means for displaying the summary data received by the terminal;

[1356] A system including:

[1357] (Claim 2)

[1358] 10. The system of claim 1, further comprising means for formatting the summary results so that they are displayed in an easy-to-view format.

[1359] (Claim 3)

[1360] 10. The system of claim 1, further comprising means for analyzing metadata to calculate a relevance score when scoring the relevance of documents.

[1361] "Example 1"

[1362] (Claim 1)

[1363] a means for a user to input search keywords;

[1364] A means for the server to retrieve relevant documents from an academic database;

[1365] A means for the server to analyze the metadata of documents, evaluate their relevance, and select top documents;

[1366] A means for the server to use a generative AI model to summarize the abstracts, results, and discussions of the literature;

[1367] means for transmitting the summarized data by the server to a user's terminal;

[1368] means for displaying the summary data received by the terminal;

[1369] A system including:

[1370] (Claim 2)

[1371] 10. The system of claim 1, further comprising means for formatting the summary results so that they are displayed in an easy-to-view format.

[1372] (Claim 3)

[1373] 10. The system of claim 1, further comprising means for analyzing the metadata and calculating a relevance score when assessing the relevance of a document.

[1374] "Application Example 1"

[1375] (Claim 1)

[1376] a means for a user to input search keywords;

[1377] a means for the server to retrieve relevant information from a database;

[1378] a means for the server to score the relevance of the information and select the top information;

[1379] A means for the server to summarize information using a generating AI;

[1380] means for transmitting the summarized data by the server to a user's terminal;

[1381] means for displaying the summary data received by the terminal;

[1382] A system including:

[1383] (Claim 2)

[1384] 10. The system of claim 1, further comprising means for formatting the summary results so that they are displayed in an easy-to-view format.

[1385] (Claim 3)

[1386] 10. The system of claim 1, further comprising means for analyzing metadata to calculate a relevance score when scoring the relevance of information.

[1387] "Example 2: Combining Emotion Engines"

[1388] (Claim 1)

[1389] a means for a user to input search keywords;

[1390] A means for the terminal to collect the user's emotions from facial expressions and voice;

[1391] A means for the server to retrieve relevant documents from an academic database;

[1392] a means for the server to score the relevance of documents and select the top documents;

[1393] A server-generated AI-based method for summarizing abstracts, results, and discussions of literature;

[1394] A means for formatting the summarized data based on the emotion data of the user by the server;

[1395] means for transmitting the summarized data by the server to a user's terminal;

[1396] means for displaying the summary data received by the terminal;

[1397] A system including:

[1398] (Claim 2)

[1399] 10. The system of claim 1, further comprising means for formatting the summary results so that they are displayed in a format that best suits the emotional state of the user.

[1400] (Claim 3)

[1401] 2. The system of claim 1, further comprising means for analyzing the acquired metadata to calculate a relevance score when scoring the relevance of documents.

[1402] "Application example 2 when combining emotion engines"

[1403] (Claim 1)

[1404] a means for a user to input search keywords;

[1405] A means for the server to retrieve relevant documents from an academic database;

[1406] a means for the server to score the relevance of documents and select the top documents;

[1407] A server-generated AI-based method for summarizing abstracts, results, and discussions of literature;

[1408] means for transmitting the summarized data by the server to a user's terminal;

[1409] means for recognizing a user's emotion;

[1410] means for adjusting the summary results based on the perceived sentiment;

[1411] means for displaying the summary data received by the terminal;

[1412] A system including:

[1413] (Claim 2)

[1414] 10. The system of claim 1, further comprising means for formatting the summary results so that they are displayed in an easy-to-view format.

[1415] (Claim 3)

[1416] 10. The system of claim 1, further comprising means for analyzing metadata to calculate a relevance score when scoring the relevance of documents. [Explanation of symbols]

[1417] 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 a user to input search keywords; A means for the server to retrieve relevant documents from an academic database; a means for the server to score the relevance of documents and select the top documents; A server-generated AI-based method for summarizing abstracts, results, and discussions of literature; means for transmitting the summarized data by the server to a user's terminal; means for displaying the summary data received by the terminal; A system including:

2. 10. The system of claim 1, further comprising means for formatting the summary results so that they are displayed in an easy-to-read format.

3. 10. The system of claim 1, further comprising means for analyzing metadata to calculate a relevance score when scoring documents for relevance.

Images