System

Abstract

A system is provided.SOLUTION: A system, comprising: means for receiving a search query input by a user; means for searching an academic database based on the search query to identify relevant academic papers; means for using a generative model to generate abstracts of the identified academic papers; means for analyzing research trends based on the generated abstracts and associated metadata; and means for displaying the abstracts and trend analysis results to the user.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 the past, researchers and students found it difficult to efficiently find relevant academic papers from the vast amount of available information. Even if they did find relevant papers, it took a great deal of time and effort to quickly understand and utilize their key discoveries and research trends. Therefore, to solve this problem, a system was needed that could efficiently search, summarize, and analyze trends in academic papers. [Means for solving the problem]

[0005] The present invention solves the above-mentioned problems by the following means: A system includes a means for receiving a search query entered by a user and a means for searching academic databases based on the search query to identify relevant academic papers. The system also includes a means for using a generative model to generate summaries of the identified academic papers, and a means for analyzing research trends based on the generated summaries and associated metadata. The system further includes a means for displaying the results of the summaries and trend analysis to the user. This system allows users to efficiently find relevant papers and quickly understand and utilize information through the summaries and trend analysis.

[0006] A "user" is a person who interacts with the system to enter search queries and receive results.

[0007] A "search query" is a specific topic or keyword that a user enters into the system.

[0008] An "academic database" is a collection of electronic information that includes academic papers, research results, and so on.

[0009] An "academic paper" is a document that describes academic research results or discoveries.

[0010] A "generative model" is an algorithm or program that uses natural language processing techniques to generate summaries of academic papers.

[0011] An "abstract" is a concise description of the main points and conclusions of an academic paper.

[0012] "Metadata" refers to supplementary information about academic papers, such as their title, author, publication year, abstract, and number of citations.

[0013] "Research trends" are information that indicates current popular topics, technological advances, and research directions in a particular research field.

[0014] "Trend analysis" is the process of analyzing research trends and patterns based on data from related academic papers.

[0015] "Display means" is an interface that allows the system to visually present information to the user. [Brief explanation of the drawings]

[0016] [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

[0017] 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.

[0018] First, the terms used in the following description will be explained.

[0019] 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).

[0020] 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.

[0021] 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.

[0022] 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.

[0023] 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."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 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.

[0027] 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).

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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."

[0037] The present invention relates to an AI system that efficiently finds highly relevant academic papers from a large number of papers and provides summaries and research trends. Specific embodiments for carrying out the present invention will be described below.

[0038] System configuration

[0039] This system consists of a user terminal, a server, a generative model, and an academic database.

[0040] 1. User terminal: Provides an interface for users to input topics and keywords.

[0041] 2. Server: Processes input queries, searches academic papers, generates summaries, and performs trend analysis.

[0042] 3. Generative model: Generate summaries of academic papers using natural language processing techniques.

[0043] 4. Academic database: A collection of data containing relevant academic papers.

[0044] Program processing

[0045] 1. User enters query (terminal):

[0046] A user inputs a specific topic or keyword, such as "application of generative modeling," through the interface of their device. When the user clicks the search button, the input query is sent to the server.

[0047] 2. The server receives the query:

[0048] The server receives and interprets queries submitted by users, and based on these queries, the server searches academic databases to find relevant academic papers.

[0049] 3. Searching academic databases:

[0050] The server queries academic databases to retrieve a list of relevant academic papers, returning metadata such as the paper's title, author, publication year, abstract, and number of citations.

[0051] 4. Summary generation using generative models:

[0052] The server then passes the retrieved list of papers to a generative model, which then uses natural language processing techniques to generate a summary of each paper, including the paper's main points and conclusions.

[0053] 5. Trend Analysis:

[0054] The server analyzes research trends based on the generated abstracts and associated metadata, taking into account the year of publication, number of citations, and key keywords of the papers. For example, it can generate a trend report showing that "Technology X" is attracting particular attention in recent research.

[0055] 6. Displaying the results:

[0056] The server formats the generated summary and trend analysis results and sends them back to the user's terminal, where the user's terminal displays the received information in a format that the user can visually confirm, such as by displaying a graph visualizing the trend analysis along with the summary list.

[0057] Specific examples

[0058] A user searches for "applications of generative models."

[0059] 1. Device: The user enters "application of generative models" in the search bar and clicks the search button.

[0060] 2. Server: Receives the query and searches academic databases to retrieve a list of relevant papers on "applications of technology X."

[0061] 3. Summary generation: The server uses the generative model to generate a summary of the paper, creating a summary statement such as, "This paper shows that technology X outperforms conventional technology."

[0062] 4. Trend analysis: The server generates a trend report based on the publication year, number of citations, and major keywords, stating that "the application of generative models has developed rapidly in recent years."

[0063] 5. Results display: A summary and trend report will be displayed on the user's terminal, allowing the user to conduct further research and discussion based on the information.

[0064] As described above, the system of the present invention efficiently searches for related academic papers and provides their summaries and trend information, thereby enabling researchers and students to quickly obtain the information they need.

[0065] The processing flow will be explained below.

[0066] Step 1:

[0067] The user inputs a search query using the interface of their device. For example, the user inputs "application of generative model" as the search query and clicks the search button.

[0068] Step 2:

[0069] The terminal sends the query entered by the user to the server. Specifically, the terminal sends the query as an HTTP request to the server.

[0070] Step 3:

[0071] The server analyzes the received query and executes a search query against the appropriate academic database, which retrieves a list of relevant papers.

[0072] Step 4:

[0073] The server analyzes the list of academic papers it has obtained and extracts the metadata for each paper (title, author, year of publication, abstract, number of citations, etc.).

[0074] Step 5:

[0075] The server passes the extracted metadata and full text of the papers to a generative model, which uses natural language processing techniques to generate a summary of each academic paper.

[0076] Step 6:

[0077] The generative model returns a summary to the server, including the main points and conclusions of each paper. For example, it generates a summary such as "The effectiveness of new generative model technology X has been demonstrated."

[0078] Step 7:

[0079] The server performs trend analysis based on the generated abstracts and metadata, taking into account the year of publication, number of citations, and key keywords of each paper to identify research trends.

[0080] Step 8:

[0081] Based on the results of the trend analysis, the server generates a trend report, such as "The application of generative models has developed rapidly in recent years, with technology X attracting particular attention."

[0082] Step 9:

[0083] The server sends the generated summary and trend report to the user's terminal, formatted in a user-friendly format.

[0084] Step 10:

[0085] The terminal displays the received summary and trend report to the user. The user interface displays a summary list and visualized trend graphs. The user can then conduct further research based on this information.

[0086] Through the above steps, users can efficiently find relevant academic papers, quickly understand summaries and trends, and effectively utilize information.

[0087] Example 1

[0088] 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."

[0089] Conventional academic paper search systems make it difficult to efficiently find relevant papers, and it is also difficult to quickly understand the content of each paper. Furthermore, they lack the functionality to analyze and provide research trends, making it difficult for researchers and students to quickly obtain the information they need. For this reason, there was a need to develop a system that could efficiently extract necessary knowledge from vast amounts of information and provide it in an easy-to-understand manner.

[0090] 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.

[0091] In this invention, the server includes a means for a user to input topics and keywords, a means for sending a query to the server, a means for receiving and analyzing the query, a means for searching academic databases and retrieving related academic papers, a means for generating summaries of the papers using a generative model, a means for analyzing research trends based on the generated summaries and related metadata, and a means for formatting the summaries and trend analysis results and transmitting them to a user terminal. This makes it possible to efficiently search for highly relevant academic papers and quickly provide their summaries and trend information.

[0092] - "Query" refers to the string of characters or keywords that a user enters to search for specific information.

[0093] "Server" refers to a computer system that receives queries from users, processes them, and provides appropriate search results, summaries, and trend analysis results.

[0094] "User Terminal" means the computer or mobile device used by a User to access the System, enter queries, and view results.

[0095] An "academic database" refers to a collection of data that contains and is used to search academic papers and related information.

[0096] A "generative model" refers to an algorithm or software that uses natural language processing techniques to automatically generate summaries of academic papers.

[0097] An "abstract" is a piece of text that concisely summarizes the main points and conclusions of an academic paper.

[0098] "Trend analysis" refers to the process of identifying recent trends and hot topics in a research field based on the metadata of retrieved academic papers.

[0099] "Metadata" refers to information about the data itself, such as the paper's title, author, publication year, abstract, and number of citations.

[0100] "HTTP request" refers to the protocol used by a user terminal to request data or services from a server.

[0101] "Natural language processing technology" refers to the techniques and methods that allow computers to understand, generate, and analyze human language.

[0102] This system efficiently finds related academic papers by allowing users to input topics and keywords, and provides summaries and research trends. Specifically, it consists of a user terminal, a server, a generative model, and an academic database.

[0103] User terminal

[0104] The user device provides an interface for users to input queries. This is implemented as a web or mobile application and is designed to allow users to easily input topics and keywords. For example, a user might input the query "application of generative AI model."

[0105] server

[0106] The server receives queries sent from user devices and analyzes their contents. Based on these queries, the server searches academic databases to find relevant academic papers. Academic databases often use SQL Server or Elasticsearch.

[0107] The obtained list of papers is passed to a generative model, which generates a summary of each paper. The generative model uses natural language processing techniques such as the Transformer or BERT model. The generated summaries are then stored on the server.

[0108] The server then analyzes the abstracts and associated metadata to identify research trends, taking into account factors such as the year of publication, number of citations, and key keywords of the papers. For example, it can generate a trend report that states, "The application of generative AI models has rapidly increased in recent years."

[0109] Trend analysis

[0110] The server formats the generated summary and trend analysis results and sends them to the user's terminal, where the user can visually check the results, such as a summary list or a graph visualizing the trend analysis.

[0111] Specific examples

[0112] The following is the series of processes that a user goes through when searching for "applications of generative models."

[0113] 1. User: The user enters "application of generative models" in the search bar and clicks the search button.

[0114] 2. Server: Receives the query and searches academic databases to obtain a list of relevant papers on "applications of generative models."

[0115] 3. Generative model: The server uses the generative model to generate a summary of the paper, creating a summary statement such as, "This paper demonstrates that generative models outperform conventional techniques."

[0116] 4. Trend analysis: The server generates a trend report based on the publication year, number of citations, and major keywords, stating that "the application of generative models has developed rapidly in recent years."

[0117] 5. Display of results: A summary and trend report is displayed on the user's terminal, allowing the user to conduct further research or discussion based on the information.

[0118] Example prompts to input to a generative AI model:

[0119] "Please search for the latest academic papers on applications of generative models and provide summaries and research trends."

[0120] This system makes it possible to efficiently search for relevant academic papers and quickly provide their summaries and trend information, enabling researchers and students to quickly and accurately obtain the information they need.

[0121] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0122] Step 1:

[0123] Query input (user)

[0124] A user enters a specific topic or keyword into the search field of a web application that serves as the interface for their device, for example, "applications of generative AI models," and clicks the search button. This action stacks the query locally on the device, ready to be sent.

[0125] Input: User-entered topics or keywords (e.g., "applications of generative AI models")

[0126] Output: Query data sent by the device to the server

[0127] Step 2:

[0128] Query sending (terminal)

[0129] The device generates an HTTP request containing the query entered by the user and sends it to the server's API endpoint, along with the user's session information and a timestamp.

[0130] Input: User-entered query, session information, timestamp

[0131] Output: HTTP request sent to the server

[0132] Step 3:

[0133] Query reception and interpretation (server)

[0134] The server analyzes the received HTTP request and extracts the query portion. It obtains the query "application of generative AI model" from the received request and normalizes the query using natural language processing. This analysis is performed using a text analysis library or similar.

[0135] Input: Query data in the HTTP request

[0136] Output: Normalized query

[0137] Step 4:

[0138] Academic database search (server)

[0139] The server searches academic databases using normalized queries, executing SQL or Elasticsearch queries to filter and retrieve relevant academic papers.

[0140] Input: Normalized query

[0141] Output: A search result list of relevant academic papers

[0142] Step 5:

[0143] Obtaining a list of papers (server)

[0144] The server creates a list of relevant papers based on the search results obtained from academic databases. Specifically, it collects metadata such as the paper title, author, publication year, abstract, and number of citations, and compiles them into a list.

[0145] Input: Search results from academic databases

[0146] Output: List of article metadata

[0147] Step 6:

[0148] Paper summary generation (server / generative model)

[0149] The server generates summaries of each paper using generative models, such as the Transformer and BERT models, which use natural language processing techniques, to generate summaries that include the main points and conclusions of each paper.

[0150] Input: Article metadata

[0151] Output: Generated paper abstract

[0152] Step 7:

[0153] Trend analysis (server)

[0154] The server analyzes research trends based on the generated abstracts and related metadata, taking into account the year of publication, number of citations, and key keywords of the paper, and generates a trend report such as, "The application of generative AI models has rapidly increased in recent years."

[0155] Input: Generated paper abstracts and metadata

[0156] Output: Trend report

[0157] Step 8:

[0158] Result formatting and sending (server)

[0159] The server formats the generated summary and trend analysis results and sends them to the user's device. The formatted information is encoded in, for example, JSON format and included in the HTTP response.

[0160] Input: Summary and trend analysis results

[0161] Output: HTTP response sent to the user device

[0162] Step 9:

[0163] Result display (terminal)

[0164] The terminal analyzes the results received from the server and displays them visually to the user, who can view summary lists and graphs visualizing trend analysis.

[0165] Input: HTTP response from the server

[0166] Output: On-screen summary list and trend analysis results

[0167] (Application example 1)

[0168] 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."

[0169] Currently, it takes a great deal of time and effort for researchers to efficiently find relevant papers from a large number of academic papers and analyze their summaries and research trends. Furthermore, there are only a limited number of systems that automatically generate paper summaries and display them visually, and convenience is particularly required on mobile devices such as smartphones. Furthermore, there is a lack of trend analysis to track research progress. To solve these problems, a system is needed that can quickly identify relevant academic papers and efficiently provide their summaries and trends.

[0170] 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.

[0171] In this invention, the server includes means for receiving a search query entered by a user, means for searching an academic database based on the search query to identify relevant academic papers, means for using a generative model to generate summaries of the identified academic papers, means for analyzing research trends based on the generated summaries and related data, means for visually displaying the summaries and trend analysis results to the user, means for passing the searched paper list to the generative model to generate summaries of each paper, and means for analyzing the number of citations and major keywords of each paper to detect research trends. This makes it possible to quickly identify relevant academic papers and efficiently provide their summaries and trends.

[0172] "Means for receiving a user-entered search query" refers to a device or software that provides an interface for a user to enter a topic or keyword they wish to search for and that is capable of receiving such input.

[0173] "Means for searching academic databases based on the search query and identifying relevant academic papers" refers to a device or software that searches academic databases based on a search query entered by a user and finds relevant academic papers.

[0174] "Means for using a generative model to generate summaries of the identified academic literature" refers to a device or software that uses AI technology known as a generative model to create summaries in order to automatically summarize the contents of the identified academic literature.

[0175] "Means for analyzing research trends based on the generated summaries and associated data" refers to devices or software that use the generated summaries and associated metadata to analyze movements and trends in a particular research field.

[0176] "Means for visually displaying the summary and trend analysis results to the user" refers to a device or software that displays the generated summary and trend analysis results on a screen in a format that is easy for the user to understand.

[0177] "Means for passing the searched paper list to a generative model and generating a summary of each paper" refers to a device or software that inputs the multiple paper lists obtained as search results into a generative model and generates a summary of each paper.

[0178] "Means for detecting research trends by analyzing the number of citations and major keywords of each document" refers to a device or software that analyzes the number of citations and major keywords of each paper and detects research trends based on that information.

[0179] The present invention relates to an AI system that efficiently finds highly relevant academic papers from a large number of papers and provides summaries and research trends. Specific embodiments for carrying out the present invention are described below.

[0180] System configuration

[0181] This system consists of a user terminal, a server, a generative model, and an academic database.

[0182] 1. User device: Provides an interface for users to input topics and keywords. Users can enter search queries through smartphone applications and visually check the results.

[0183] 2. Server: Processes input queries, searches academic papers, generates summaries, and performs trend analysis. The server is implemented using Node.js and Express.

[0184] 3. Generative model: Generate summaries of academic papers using natural language processing techniques. OpenAI's GPT-4 is used to generate summary sentences.

[0185] 4. Academic database: A collection of data containing related academic papers. Academic data can be obtained using APIs such as PubMed and arXiv.

[0186] System Operation

[0187] A user enters a specific keyword into the search bar of a smartphone app, for example, "application of generative models," and clicks the search button. This query is sent to the server.

[0188] The server calls the API of an academic database based on the received query to obtain a list of related papers, which it then passes to a generative model (GPT-4) to generate summaries of each paper.

[0189] The generated summaries are returned to the server, where a trend analysis is performed based on the metadata of each paper (number of citations, year of publication, main keywords, etc.). The progress of research and popular technologies are analyzed, and the results are generated as a trend report.

[0190] Finally, the server sends the summary and trend report to the user's terminal, where the user can visually check this information. For example, a summary list is displayed, and below it, visualized information such as trend graphs and keyword frequency rates can be viewed.

[0191] Example sentences

[0192] Example prompt sentence:

[0193] Summarize the following academic paper:

[0194] Title: Generative Models and Their Applications

[0195] Abstract: This paper explores the various applications of generative models, including their usage in image generation, natural language processing, and other fields...

[0196] This system allows researchers to quickly identify relevant academic papers and efficiently obtain their summaries and trends. For example, if a user searches for "applications of generative models," the system will display the latest related research papers along with their summaries. Furthermore, based on trend analysis, recent research trends can be visually confirmed, allowing researchers to efficiently grasp the progress of research.

[0197] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0198] Step 1:

[0199] The user enters a query

[0200] A user enters a topic or keyword into the search bar of a smartphone app, for example, "application of generative models." When the user clicks the search button, the query is sent from the device to the server.

[0201] Input: Search keywords entered by the user (e.g., "application of generative models")

[0202] Output: The search query sent to the server

[0203] Step 2:

[0204] The server receives the query and searches the academic database.

[0205] The server receives the search query sent from the device and calls the API of an academic database based on the query to search for related academic papers. The server uses the API of PubMed or arXiv, for example.

[0206] Input: The received search query

[0207] Data processing: Calling APIs of academic databases based on queries

[0208] Output: A list of academic papers obtained as search results

[0209] Step 3:

[0210] Pass the obtained list of academic papers to the generative model

[0211] The server passes each paper in the list of academic papers to a generative model (GPT-4) to generate a summary. The generative model analyzes the content of each paper and generates a summary that includes the main points.

[0212] Input: List of academic papers

[0213] Data Computation: Generating summaries using generative models

[0214] Output: Abstract of each paper

[0215] Step 4:

[0216] Conduct trend analysis based on the generated summaries and metadata

[0217] The server analyzes research trends based on the generated abstracts and the metadata of each paper (year of publication, number of citations, main keywords, etc.), for example, by analyzing the frequency of keyword appearances and the year of publication of the paper.

[0218] Input: Abstract and article metadata

[0219] Data Calculation: Trend Analysis through Metadata Analysis

[0220] Output: Trend report

[0221] Step 5:

[0222] Sends summary and trend reports to user terminals for display

[0223] The server formats the generated summary and trend report and sends them to the user's terminal, which receives and visually displays them, for example, visualizing the summary list along with trend graphs and keyword frequency rates.

[0224] Input: Abstract and trend report

[0225] Data processing: formatting summaries and trend reports

[0226] Output: Information displayed on the user's terminal (summary list, trend graph, etc.)

[0227] 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.

[0228] The present invention combines an emotion engine with a system that efficiently finds highly relevant academic papers from a large number of papers and provides summaries and research trends. This makes it possible to recognize the user's emotions and provide optimal information based on those emotions. Specific embodiments of the present invention are described below.

[0229] System configuration

[0230] This system consists of a user terminal, a server, a generative model, an academic database, and an emotion engine.

[0231] 1. User terminal: Provides an interface where users can input topics and keywords and obtain emotional information.

[0232] 2. Server: Processes queries and sentiment information, searches academic papers, generates summaries, and performs trend analysis.

[0233] 3. Generative model: Generate summaries of academic papers using natural language processing techniques.

[0234] 4. Academic database: A collection of data containing relevant academic papers.

[0235] 5. Emotion engine: A program that recognizes and analyzes emotions from user input, facial expressions, voice, etc.

[0236] Program processing

[0237] 1. User enters a query and sentiment is read (on device):

[0238] The user enters a search query such as "application of generative model" on the device. The device is also equipped with an emotion engine that reads the user's emotions in real time from their facial expressions, voice, and input content. The read emotion information is then sent to the server along with the query.

[0239] 2. The server receives the query and emotion information:

[0240] The server receives the search query and sentiment information sent by the user, and then executes the search query against the appropriate academic database based on the received query and sentiment information.

[0241] 3. Searching academic databases:

[0242] The server queries academic databases to retrieve a list of relevant academic papers, returning metadata such as the paper's title, author, publication year, abstract, and number of citations.

[0243] 4. Summary generation using generative models:

[0244] The server then passes the retrieved list of papers to a generative model, which uses natural language processing techniques to generate a summary of each paper, including the main points and conclusions of the paper.

[0245] 5. Trend analysis taking into account sentiment information:

[0246] The server analyzes research trends by taking into account emotional information along with the generated summaries and associated metadata. For example, the content of academic papers is filtered to match the user's interests and state. If the emotional engine recognizes the user's state of excitement, it will prioritize providing trend information on the latest technological innovations.

[0247] 6. Displaying the results:

[0248] The server then sends the generated summary and a trend report adjusted based on the sentiment information to the user's device. The device then displays the received information in a format that is easy for the user to understand, such as a graph visualizing the trend analysis along with the summary list.

[0249] Specific examples

[0250] The user searches for "applications of generative models." The user's emotions are also recognized.

[0251] 1. Device: The user enters "Application of generative model" and clicks the search button. At the same time, the emotion engine detects the user's excitement state.

[0252] 2. Server: Receives the query along with the emotion information and searches the academic database to obtain a list of relevant papers on "applications of technology X."

[0253] 3. Summary generation: The server uses the generative model to generate a summary of the paper, such as "This paper demonstrates that technology X outperforms conventional technologies."

[0254] 4. Sentiment Analysis: Taking into account the excitement level of users, trend analysis prioritizes trend information on the latest technological innovations.

[0255] 5. Result display: A summary and trend report is displayed on the user's terminal, highlighting the latest technological information taking into account the user's excitement.

[0256] Through these steps, users can efficiently find relevant academic papers, quickly understand summaries and trends, and effectively utilize information according to their emotional state.

[0257] The processing flow will be explained below.

[0258] Step 1:

[0259] The user enters a search query using the device's interface. For example, they enter "application of generative models" and click the search button. The device's built-in emotion engine then reads the user's emotions in real time from their facial expressions and voice.

[0260] Step 2:

[0261] The device sends the search query and emotion information entered by the user to the server. Specifically, the device sends the query and emotion information together as an HTTP request to the server.

[0262] Step 3:

[0263] The server analyzes the received query and sentiment information. Based on the query, the server searches academic databases to retrieve a list of relevant academic papers. The search results include metadata such as the paper title, author, publication year, abstract, and number of citations.

[0264] Step 4:

[0265] The server analyzes the list of academic papers it has acquired and passes each paper's metadata and full text to a generative model. The generative model then uses natural language processing technology to generate a summary of each paper. For example, it generates a summary such as, "This paper demonstrates that technology X outperforms conventional technology."

[0266] Step 5:

[0267] The server performs trend analysis based on the generated summary, metadata, and emotion information. For example, if the emotion engine recognizes the user's excitement, it will prioritize analyzing trend information related to the latest technological innovations.

[0268] Step 6:

[0269] As a result of the trend analysis, the server generates a trend report, such as "The application of generative models has developed rapidly in recent years, with technology X attracting particular attention."

[0270] Step 7:

[0271] The server sends the generated summary and trend report to the user's device, formatting it in a visually optimized format that takes into account emotional information, and then transmits the data.

[0272] Step 8:

[0273] The terminal displays the received summary and trend report to the user. The user interface displays summary lists and graphs that visualize trend analysis, highlighting important information that corresponds to the user's excitement level. The user can then use this information for further detailed research and discussion.

[0274] As described above, the processing flow of the system combined with the emotion engine enables users to more efficiently find relevant academic papers, quickly understand summaries and trend information, and obtain the most appropriate information according to their emotional state.

[0275] Example 2

[0276] 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."

[0277] Conventional academic resource search systems search academic databases based on user-entered search queries to identify relevant materials, but they have the problem of being unable to provide information that takes into account the user's emotions and interests. This makes it impossible to provide appropriate information or analyze trends according to the user's emotional state, making it difficult to improve user satisfaction and convenience.

[0278] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0279] In this invention, the server includes means for receiving a search query entered by a user, means for searching an academic database based on the search query to identify relevant academic materials, means for using a generative model to generate summaries of the identified academic materials, means for analyzing research trends based on the summaries and associated metadata, means for recognizing user sentiment and adjusting the search query and trend analysis based on that information, and means for displaying the summaries and trend analysis results to the user, thereby providing optimal information according to the user's sentiment, improving the accuracy of search results and trend analysis and user satisfaction.

[0280] A "user" is an entity that uses this system to search for academic materials and obtain information.

[0281] A "search query" is a theme or keyword for academic materials to be searched that a user enters into a terminal.

[0282] An "academic database" is a collection of data that collects and manages academic materials, such as academic papers and research results.

[0283] "Academic materials" refers to documents such as papers, research reports, and presentations stored in academic databases.

[0284] A "generative model" is an algorithm or system that uses artificial intelligence and natural language processing techniques to analyze text data and generate a summary.

[0285] An "abstract" is a document that succinctly summarizes the main points and conclusions of academic material.

[0286] "Metadata" refers to data that indicates information related to academic materials, such as title, author, year of publication, and number of citations.

[0287] A "trend" refers to recent developments or popular trends in a particular academic field or topic.

[0288] "Emotion" refers to the user's psychological state and emotional response, and in this system, it is recognized from facial expressions, voice, input content, etc.

[0289] "Emotion information" is data about the user's emotional state obtained by the emotion engine.

[0290] An "emotion engine" is a program and system that recognizes and analyzes emotions from a user's facial expressions, voice, and input content.

[0291] A "terminal" is a device through which a user enters a search query and receives results, and includes, for example, a personal computer or smartphone.

[0292] The "server" is a central processing unit that processes search queries and sentiment information from users, searches academic databases, generates summaries, and performs trend analysis.

[0293] The present invention combines a system that efficiently finds highly relevant information from a large amount of academic materials and provides summaries and research trends with an emotion engine that recognizes the user's emotions. This makes it possible to provide optimal information tailored to the user's emotions. Specific embodiments of the present invention are described below.

[0294] System configuration

[0295] This system consists of a user terminal, a server, a generative model, an academic database, and an emotion engine.

[0296] 1. User Device:

[0297] It provides an interface for users to input search queries and obtain emotion information. For example, it is a PC or smartphone. The device is equipped with a camera and microphone, and has the ability to analyze the user's facial expressions and voice using an emotion engine.

[0298] 2. Server:

[0299] It is a central device that processes search queries and emotional information sent by users. Specifically, it receives queries and emotional information, searches academic databases, generates abstracts of papers using generative models, and performs trend analysis taking into account the emotional information.

[0300] 3. Generative Model:

[0301] These are algorithms and systems that use natural language processing techniques to generate summaries of academic materials, such as generative AI models using the Transformer architecture.

[0302] 4. Academic databases:

[0303] A collection of data that contains related scholarly materials, such as electronic journals and digital libraries.

[0304] 5. Emotion Engine:

[0305] It is a program and system that recognizes and analyzes emotions from user input, facial expressions, voice, etc. The emotion engine uses a regular camera and microphone to acquire and analyze emotional information in real time.

[0306] Program processing

[0307] When a user enters a search query, the device receives the query and simultaneously acquires the user's emotional information through the emotion engine. The acquired query and emotional information are then sent to the server, which then executes the search query against the academic database.

[0308] The list of academic materials obtained as search results is summarized by a generative model on the server, and trend analysis is then performed based on the generated summaries, metadata, and user sentiment information to uniquely provide the most relevant information.

[0309] Specific examples

[0310] When a user searches for "application of generative modeling," the user enters "application of generative modeling" into the device, and the emotion engine detects the user's excitement. The device then sends the search query and emotion information to the server.

[0311] The server searches academic databases based on the search query "application of generative models" to retrieve relevant materials. The retrieved materials are summarized using a generative model. For example, a summary such as "This research proposes a new application method using generative models" is generated.

[0312] The server then performs trend analysis to prioritize trend information on the latest technological innovations, taking into account the user's excitement level detected by the emotion engine. The results of the trend analysis and a summary are finally sent to the device.

[0313] The terminal receives this information and displays it to the user. The user interface displays summary lists and visualizes the latest technology trends in graphs and charts.

[0314] Prompt Sentence Examples

[0315] "Design a system to search the latest academic literature on the application of generative models and provide optimal summaries and trend information to excited users."

[0316] In this way, the system provides users with the most appropriate information based on their emotional state, allowing them to efficiently search and understand academic materials. It also helps users understand the latest research trends through trend analysis.

[0317] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0318] Step 1:

[0319] User enters search query and acquires emotion information (device)

[0320] A user types "application of generative models" into the search bar of their device and presses the Enter key. Pressing the Enter key triggers the device's built-in emotion engine, which uses the camera and microphone to analyze the user's facial expressions and voice in real time. The input data is the search query "application of generative models" and emotion information (e.g., excited state) acquired by the emotion engine. The emotion engine then outputs the results of the user's facial expression analysis and generates a data packet that is sent to the server along with the search query.

[0321] Step 2:

[0322] Send query and emotion information to the server (device)

[0323] The device sends the search query "Application of generative model" and the acquired emotion information together in a single data packet to the server. The transmission is triggered when the user presses the Enter key. The data packet contains the search keyword and the user's emotion (e.g., excitement) and is transferred to the server using a transmission protocol. The server is then ready to analyze the received data.

[0324] Step 3:

[0325] Receive query and emotion information (server)

[0326] The server receives the data packet sent from the terminal. The received data packet is decoded by the server's receiving module, and the search query "Application of generative model" and emotional information (e.g., excitement state) are separated. The server then passes the received search query to the analysis module, and the emotional information is passed to the emotion analysis module.

[0327] Step 4:

[0328] Search academic databases (server)

[0329] The server's query engine queries the received search query "application of generative models" to the academic database. The input data is the search query, and the database query execution engine executes the query. As a result, relevant paper data (metadata such as title, author, publication year, abstract, and number of citations) from the academic database is output as search results.

[0330] Step 5:

[0331] Generate a summary based on the search results (server)

[0332] The server passes search results obtained from the academic database to the generative model, which uses natural language processing techniques to generate summaries of the papers. The input data is a list of related papers and their metadata, which the generative model analyzes to create a concise summary. The output is a summary such as, "This research proposes a new application method using the generative model."

[0333] Step 6:

[0334] Conduct trend analysis taking into account emotional information (server)

[0335] The server performs trend analysis using the generated paper summaries and related metadata, as well as user sentiment information. The input data are the abstracts, metadata, and sentiment information. The sentiment analysis engine takes into account the user's state of excitement and prioritizes the extraction of trend information related to the latest technological innovations. The output is a trend report optimized for user sentiment.

[0336] Step 7:

[0337] Send the final result to the user terminal (server)

[0338] The server then combines the generated abstracts and the adjusted trend report into a single data packet and sends it to the user's device. The transmission protocol is HTTP or WebSocket. The data packet contains a list of abstracts and a trend report, allowing the user to receive the latest research trends and abstracts.

[0339] Step 8:

[0340] Display the results to the user (terminal)

[0341] The terminal receives the data packets sent from the server, analyzes the received summary statements and trend reports, and displays them. The user interface displays the summary list in an easy-to-read format and visualizes the latest technology trend information in graphs and charts. This is a step that allows users to efficiently obtain information and deepen their understanding.

[0342] (Application example 2)

[0343] 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."

[0344] Currently, security measures rely mainly on pre-defined rules or manual user intervention, making it difficult to implement appropriate measures that reflect the user's psychological state in real time. While existing systems identify relevant materials from large volumes of academic documents and generate summaries, they do not present search results or countermeasures that take the user's emotional state into account. As a result, it is difficult for users to obtain optimal information and countermeasures in situations where they feel stressed or tense.

[0345] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving a search query entered by a user, means for searching an academic database based on the search query and identifying relevant academic materials, means for using a generative model to generate a summary of the identified academic materials, means for analyzing research trends based on the generated summary and related information, means for displaying the summary and trend analysis results to the user, and means for recognizing the user's emotional state using an emotion engine and proposing and implementing search results and security measures based on that. This makes it possible to provide optimal information and security measures in real time according to the user's psychological state.

[0346] A "search query" is a keyword or phrase that a user enters into a search system.

[0347] An "academic database" is a collection of information that contains academic materials.

[0348] "Academic materials" are documents containing academic information, such as research and academic papers.

[0349] A "generative model" is an algorithm that uses natural language processing techniques to automatically perform specific tasks.

[0350] "Natural language processing technology" is an artificial intelligence technology that has the ability to understand and generate human language.

[0351] An "abstract" is a sentence that succinctly summarizes the main points and conclusions of a document.

[0352] "Related information" refers to the metadata and contextual information that accompanies scholarly documents and abstracts.

[0353] "Trend analysis" is the process of identifying and analyzing patterns and trends based on data.

[0354] The "emotion engine" is a program that recognizes the user's emotional state from their facial expressions, voice, and input content.

[0355] An "emotional state" is the psychological state that a user is feeling at that moment.

[0356] "Search Results" are lists of relevant information returned by the system based on a search query.

[0357] "Security measures" are specific measures and procedures to ensure user safety.

[0358] "Suggestion and execution" means presenting the optimal option to the user and automatically implementing that option if necessary.

[0359] This invention is a system that recognizes the emotional state of a user in real time and then proposes and implements relevant information and security measures based on that information. Below, we will explain in detail the program and processing of the system that realizes this application example.

[0360] System Configuration

[0361] This system consists of a user terminal, a server, a generative model, an academic database, and an emotion engine.

[0362] 1. User Device

[0363] It provides an interface for users to enter search queries and obtain emotion information.

[0364] Hardware used: Smartphone, smart glasses

[0365] Software used: React Native and ARKit as front-end technologies

[0366] 2. Server

[0367] It processes query and sentiment information to search, summarize, and analyze trends in academic materials.

[0368] Software used: Node.js, Express.js, MongoDB, PostgreSQL

[0369] 3. Generative Model

[0370] Generate summaries of academic materials using natural language processing techniques.

[0371] Software used: Google Cloud Natural Language API, OpenAI GPT-3

[0372] 4. Academic Databases

[0373] A collection of data containing related academic materials.

[0374] Software used: MongoDB, PostgreSQL

[0375] 5. Emotion Engine

[0376] A program that recognizes and analyzes emotions from the user's facial expressions, voice, and input content.

[0377] Software used: IBM Watson, Microsoft Azure Cognitive Services

[0378] A natural language description of the process

[0379] Acquiring emotion data

[0380] The user device captures the user's facial expressions using a camera on a smartphone or smart glasses, and records their voice using a microphone. This allows for real-time emotional data acquisition, and the emotional data from the facial and vocal signals is analyzed using an emotion engine (IBM Watson or Microsoft Azure Cognitive Services).

[0381] Determining the risk level

[0382] The server evaluates the emotional data obtained from the emotion engine in real time, and if negative emotions such as tension or anxiety are strong, it judges the risk level to be high.

[0383] Proposing and implementing measures

[0384] If the risk level is determined to be high, the server will send a push notification to the user suggesting specific countermeasures. For example, it may suggest security measures such as "locking the front door." If the user allows "auto-execution," the specified security measures will be automatically executed. At this time, the server performs operations through an API that controls smart home devices (Google Home, Amazon Alexa, etc.).

[0385] Specific examples

[0386] When the user returns home late at night, the user device detects a state of tension. The server determines the security risk level to be high and notifies the user as follows: "You are feeling tense. Do you want to lock the front door?" and "Do you want to allow automatic locking?"

[0387] Prompt Sentence Examples

[0388] "If users are nervous, suggest security measures."

[0389] "Please judge the risk level based on the following emotional data. Emotional data: [Facial expression: 'Tension', Audio: 'Anxiety']"

[0390] This makes it possible to provide optimal information and security measures in real time according to the user's psychological state.

[0391] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0392] Step 1:

[0393] A user enters a search query.

[0394] Input: The user enters a "search query" into the device.

[0395] How it works: The device receives a user's search query and prepares to process it.

[0396] Output: Search query data from the user

[0397] Step 2:

[0398] The emotion engine reads the user's emotions in real time.

[0399] Input: Video and audio data captured from the camera and microphone on the user's device.

[0400] How it works: The device sends video and audio data to the emotion engine, which then analyzes it.

[0401] Output: Emotional state data (e.g., tension, anxiety)

[0402] Step 3:

[0403] The server receives the search query and the emotional state.

[0404] Input: User search query data and emotional state data.

[0405] Operation: The device sends this data to the server.

[0406] Output: Search query data and emotional state data are passed to the server.

[0407] Step 4:

[0408] The server performs searches against academic databases.

[0409] Input: Search query data.

[0410] How it works: The server searches academic databases to identify relevant academic materials.

[0411] Output: List of related academic materials

[0412] Step 5:

[0413] A generative model generates summaries of relevant academic material.

[0414] Input: List of relevant academic resources.

[0415] How it works: The server passes a list of materials to the generative model, which then generates a summary using natural language processing techniques.

[0416] Output: Generated summary

[0417] Step 6:

[0418] The server analyzes research trends based on the generated abstracts and associated metadata.

[0419] Input: Generated summary sentences and associated metadata, emotional state data.

[0420] How it works: The server aggregates this data and uses trend analysis algorithms to analyze trends.

[0421] Output: Trend analysis results

[0422] Step 7:

[0423] Determine security risk levels based on emotional states.

[0424] Input: Emotional state data.

[0425] Operation: The server evaluates the emotional state data and determines the risk level.

[0426] Output: Risk level data (e.g. high, medium, low)

[0427] Step 8:

[0428] If the security risk level is determined to be high, countermeasures will be proposed.

[0429] Input: Risk level data.

[0430] What it does: If the server determines a high risk, it will send a notification to the user suggesting specific security measures.

[0431] Output: Specific action suggestions (e.g., "Would you like to lock your front door?")

[0432] Step 9:

[0433] If the user allows automatic execution, the specified security measures are implemented.

[0434] Input: User authorization data.

[0435] How it works: The server receives the user's permission and executes the measures through the control API of the smart home device.

[0436] Output: Security measures taken (e.g., front door automatically locks)

[0437] Through the above steps, optimal information provision and security measures are automatically implemented in real time according to the user's emotions.

[0438] 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.

[0439] 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.

[0440] 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.

[0441] [Second embodiment]

[0442] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0443] 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.

[0444] 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).

[0445] 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.

[0446] 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.

[0447] 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).

[0448] 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.

[0449] 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.

[0450] 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.

[0451] 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.

[0452] 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.

[0453] 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."

[0454] The present invention relates to an AI system that efficiently finds highly relevant academic papers from a large number of papers and provides their summaries and research trends. Specific embodiments for carrying out the present invention will be described below.

[0455] System configuration

[0456] This system consists of a user terminal, a server, a generative model, and an academic database.

[0457] 1. User terminal: Provides an interface for users to input topics and keywords.

[0458] 2. Server: Processes input queries, searches academic papers, generates summaries, and performs trend analysis.

[0459] 3. Generative model: Generate summaries of academic papers using natural language processing techniques.

[0460] 4. Academic database: A collection of data containing relevant academic papers.

[0461] Program processing

[0462] 1. User enters query (terminal):

[0463] A user inputs a specific topic or keyword, such as "application of generative modeling," through the interface of their device. When the user clicks the search button, the input query is sent to the server.

[0464] 2. The server receives the query:

[0465] The server receives and interprets queries submitted by users, and based on these queries, the server searches academic databases to find relevant academic papers.

[0466] 3. Searching academic databases:

[0467] The server queries academic databases to retrieve a list of relevant academic papers, returning metadata such as the paper's title, author, publication year, abstract, and number of citations.

[0468] 4. Summary generation using generative models:

[0469] The server then passes the retrieved list of papers to a generative model, which then uses natural language processing techniques to generate a summary of each paper, including the paper's main points and conclusions.

[0470] 5. Trend Analysis:

[0471] The server analyzes research trends based on the generated abstracts and associated metadata, taking into account the year of publication, number of citations, and key keywords of the papers. For example, a trend report can be generated showing that "Technology X" is attracting particular attention in recent research.

[0472] 6. Displaying the results:

[0473] The server formats the generated summary and trend analysis results and sends them back to the user's terminal, where the user's terminal displays the received information in a format that the user can visually confirm, such as by displaying a graph that visualizes the trend analysis along with the summary list.

[0474] Specific examples

[0475] A user searches for "applications of generative models."

[0476] 1. Device: The user enters "application of generative models" in the search bar and clicks the search button.

[0477] 2. Server: Receives the query and searches academic databases to retrieve a list of relevant papers on "applications of technology X."

[0478] 3. Summary generation: The server uses the generative model to generate a summary of the paper, creating a summary statement such as, "This paper shows that technology X outperforms conventional technology."

[0479] 4. Trend analysis: The server generates a trend report based on the publication year, number of citations, and major keywords, stating that "the application of generative models has developed rapidly in recent years."

[0480] 5. Results display: A summary and trend report will be displayed on the user's terminal, allowing the user to conduct further research and discussion based on the information.

[0481] As described above, the system of the present invention efficiently searches for related academic papers and provides their summaries and trend information, thereby enabling researchers and students to quickly obtain the information they need.

[0482] The processing flow will be explained below.

[0483] Step 1:

[0484] The user inputs a search query using the interface of their device. For example, the user inputs "application of generative model" as the search query and clicks the search button.

[0485] Step 2:

[0486] The terminal sends the query entered by the user to the server. Specifically, the terminal sends the query as an HTTP request to the server.

[0487] Step 3:

[0488] The server analyzes the received query and executes a search query against the appropriate academic database, which retrieves a list of relevant papers.

[0489] Step 4:

[0490] The server analyzes the list of academic papers it has obtained and extracts the metadata for each paper (title, author, year of publication, abstract, number of citations, etc.).

[0491] Step 5:

[0492] The server passes the extracted metadata and full text of the papers to a generative model, which uses natural language processing techniques to generate a summary of each academic paper.

[0493] Step 6:

[0494] The generative model returns a summary of each paper to the server, including its main points and conclusions. For example, it generates a summary such as, "The effectiveness of new generative model technology X has been demonstrated."

[0495] Step 7:

[0496] The server performs trend analysis based on the generated abstracts and metadata, taking into account the year of publication, number of citations, and main keywords of the papers to identify research trends.

[0497] Step 8:

[0498] Based on the results of the trend analysis, the server generates a trend report, such as "The application of generative models has developed rapidly in recent years, with technology X attracting particular attention."

[0499] Step 9:

[0500] The server sends the generated summary and trend report to the user's terminal, formatted in a user-friendly format.

[0501] Step 10:

[0502] The terminal displays the received summary and trend report to the user. The user interface displays a summary list and visualized trend graphs. The user can then conduct further research based on this information.

[0503] Through the above steps, users can efficiently find relevant academic papers, quickly understand summaries and trends, and effectively utilize information.

[0504] Example 1

[0505] 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."

[0506] Conventional academic paper search systems make it difficult to efficiently find relevant papers, and it is also difficult to quickly understand the content of each paper. Furthermore, they lack the functionality to analyze and provide research trends, making it difficult for researchers and students to quickly obtain the information they need. For this reason, there was a need to develop a system that could efficiently extract necessary knowledge from vast amounts of information and provide it in an easy-to-understand manner.

[0507] 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.

[0508] In this invention, the server includes a means for a user to input topics and keywords, a means for sending a query to the server, a means for receiving and analyzing the query, a means for searching academic databases and retrieving related academic papers, a means for generating summaries of the papers using a generative model, a means for analyzing research trends based on the generated summaries and related metadata, and a means for formatting the summaries and trend analysis results and transmitting them to a user terminal. This makes it possible to efficiently search for highly relevant academic papers and quickly provide their summaries and trend information.

[0509] - "Query" refers to the string of characters or keywords that a user enters to search for specific information.

[0510] "Server" refers to a computer system that receives queries from users, processes them, and provides appropriate search results, summaries, and trend analysis results.

[0511] "User Terminal" means the computer or mobile device used by a User to access the System, enter queries, and view results.

[0512] An "academic database" refers to a collection of data that contains and is used to search academic papers and related information.

[0513] A "generative model" refers to an algorithm or software that uses natural language processing techniques to automatically generate summaries of academic papers.

[0514] An "abstract" is a piece of text that concisely summarizes the main points and conclusions of an academic paper.

[0515] "Trend analysis" refers to the process of identifying recent trends and hot topics in a research field based on the metadata of retrieved academic papers.

[0516] "Metadata" refers to information about the data itself, such as the paper's title, author, publication year, abstract, and number of citations.

[0517] "HTTP request" refers to the protocol used by a user terminal to request data or services from a server.

[0518] "Natural language processing technology" refers to the techniques and methods that allow computers to understand, generate, and analyze human language.

[0519] This system efficiently finds related academic papers by allowing users to input topics and keywords, and provides summaries and research trends. Specifically, it consists of a user terminal, a server, a generative model, and an academic database.

[0520] User terminal

[0521] The user device provides an interface for users to input queries. This is implemented as a web or mobile application and is designed to allow users to easily input topics and keywords. For example, a user might input the query "application of generative AI model."

[0522] server

[0523] The server receives queries sent from user devices and analyzes their contents. Based on these queries, the server searches academic databases to find relevant academic papers. Academic databases often use SQL Server or Elasticsearch.

[0524] The obtained list of papers is passed to a generative model, which generates a summary of each paper. The generative model uses natural language processing techniques such as the Transformer or BERT model. The generated summaries are then stored on the server.

[0525] The server then analyzes the abstracts and associated metadata to identify research trends, taking into account factors such as the year of publication, number of citations, and key keywords of the papers. For example, it can generate a trend report that states, "The application of generative AI models has rapidly increased in recent years."

[0526] Trend analysis

[0527] The server formats the generated summary and trend analysis results and sends them to the user's terminal, where the user can visually check the results, such as a summary list or a graph visualizing the trend analysis.

[0528] Specific examples

[0529] The following is the series of processes that a user goes through when searching for "applications of generative models."

[0530] 1. User: The user enters "application of generative models" in the search bar and clicks the search button.

[0531] 2. Server: Receives the query and searches academic databases to obtain a list of relevant papers on "applications of generative models."

[0532] 3. Generative model: The server uses the generative model to generate a summary of the paper, creating a summary statement such as, "This paper demonstrates that generative models outperform conventional techniques."

[0533] 4. Trend analysis: The server generates a trend report based on the publication year, number of citations, and major keywords, stating that "the application of generative models has developed rapidly in recent years."

[0534] 5. Display of results: A summary and trend report is displayed on the user's terminal, allowing the user to conduct further research or discussion based on the information.

[0535] Example prompts to input to a generative AI model:

[0536] "Please search for the latest academic papers on applications of generative models and provide summaries and research trends."

[0537] This system makes it possible to efficiently search for relevant academic papers and quickly provide their summaries and trend information, enabling researchers and students to quickly and accurately obtain the information they need.

[0538] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0539] Step 1:

[0540] Query input (user)

[0541] A user enters a specific topic or keyword into the search field of a web application that serves as the interface for their device, for example, "applications of generative AI models," and clicks the search button. This action stacks the query locally on the device, ready to be sent.

[0542] Input: User-entered topics or keywords (e.g., "applications of generative AI models")

[0543] Output: Query data sent by the device to the server

[0544] Step 2:

[0545] Query sending (terminal)

[0546] The device generates an HTTP request containing the query entered by the user and sends it to the server's API endpoint, along with the user's session information and a timestamp.

[0547] Input: User-entered query, session information, timestamp

[0548] Output: HTTP request sent to the server

[0549] Step 3:

[0550] Query reception and interpretation (server)

[0551] The server analyzes the received HTTP request and extracts the query portion. It obtains the query "application of generative AI model" from the received request and normalizes the query using natural language processing. This analysis is performed using a text analysis library or similar.

[0552] Input: Query data in the HTTP request

[0553] Output: Normalized query

[0554] Step 4:

[0555] Academic database search (server)

[0556] The server searches academic databases using normalized queries, executing SQL or Elasticsearch queries to filter and retrieve relevant academic papers.

[0557] Input: Normalized query

[0558] Output: A search result list of relevant academic papers

[0559] Step 5:

[0560] Obtaining a list of papers (server)

[0561] The server creates a list of relevant papers based on the search results obtained from academic databases. Specifically, it collects metadata such as the paper's title, author, publication year, abstract, and number of citations, and compiles it into a list.

[0562] Input: Search results from academic databases

[0563] Output: List of article metadata

[0564] Step 6:

[0565] Paper summary generation (server / generative model)

[0566] The server generates summaries of each paper using generative models, such as the Transformer and BERT models, which use natural language processing techniques, to generate summaries that include the main points and conclusions of each paper.

[0567] Input: Article metadata

[0568] Output: Generated paper abstract

[0569] Step 7:

[0570] Trend analysis (server)

[0571] The server analyzes research trends based on the generated abstracts and related metadata, taking into account the year of publication, number of citations, and key keywords of the paper, and generates a trend report such as, "The application of generative AI models has rapidly increased in recent years."

[0572] Input: Generated paper abstracts and metadata

[0573] Output: Trend report

[0574] Step 8:

[0575] Result formatting and sending (server)

[0576] The server formats the generated summary and trend analysis results and sends them to the user's device. The formatted information is encoded in, for example, JSON format and included in the HTTP response.

[0577] Input: Summary and trend analysis results

[0578] Output: HTTP response sent to the user device

[0579] Step 9:

[0580] Result display (terminal)

[0581] The terminal analyzes the results received from the server and displays them visually to the user, who can view summary lists and graphs visualizing trend analysis.

[0582] Input: HTTP response from the server

[0583] Output: On-screen summary list and trend analysis results

[0584] (Application example 1)

[0585] 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."

[0586] Currently, it takes a great deal of time and effort for researchers to efficiently find relevant papers from a large number of academic papers and analyze their summaries and research trends. Furthermore, there are only a limited number of systems that automatically generate paper summaries and display them visually, and convenience is particularly required on mobile devices such as smartphones. Furthermore, there is a lack of trend analysis to track research progress. To solve these problems, a system is needed that can quickly identify relevant academic papers and efficiently provide their summaries and trends.

[0587] 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.

[0588] In this invention, the server includes means for receiving a search query entered by a user, means for searching an academic database based on the search query to identify relevant academic papers, means for using a generative model to generate summaries of the identified academic papers, means for analyzing research trends based on the generated summaries and related data, means for visually displaying the summaries and trend analysis results to the user, means for passing the searched paper list to the generative model to generate summaries of each paper, and means for analyzing the number of citations and major keywords of each paper to detect research trends. This makes it possible to quickly identify relevant academic papers and efficiently provide their summaries and trends.

[0589] "Means for receiving a user-entered search query" refers to a device or software that provides an interface for a user to enter a topic or keyword they wish to search for and that is capable of receiving such input.

[0590] "Means for searching academic databases based on the search query and identifying relevant academic papers" refers to a device or software that searches academic databases based on a search query entered by a user and finds relevant academic papers.

[0591] "Means for using a generative model to generate summaries of the identified academic literature" refers to a device or software that uses AI technology known as a generative model to create summaries in order to automatically summarize the contents of the identified academic literature.

[0592] "Means for analyzing research trends based on the generated abstracts and associated data" refers to devices or software that use the generated abstracts and associated metadata to analyze movements and trends in a particular research field.

[0593] "Means for visually displaying the summary and trend analysis results to the user" refers to a device or software that displays the generated summary and trend analysis results on a screen in a format that is easy for the user to understand.

[0594] "Means for passing the searched paper list to a generative model and generating a summary of each paper" refers to a device or software that inputs the list of multiple papers obtained as search results into a generative model and generates a summary of each paper.

[0595] "Means for detecting research trends by analyzing the number of citations and major keywords of each document" refers to a device or software that analyzes the number of citations and major keywords of each paper and detects research trends based on that information.

[0596] The present invention relates to an AI system that efficiently finds highly relevant academic papers from a large number of papers and provides summaries and research trends. Specific embodiments for carrying out the present invention are described below.

[0597] System configuration

[0598] This system consists of a user terminal, a server, a generative model, and an academic database.

[0599] 1. User device: Provides an interface for users to input topics and keywords. Users can enter search queries through smartphone applications and visually check the results.

[0600] 2. Server: Processes input queries, searches academic papers, generates summaries, and performs trend analysis. The server is implemented using Node.js and Express.

[0601] 3. Generative model: Generate summaries of academic papers using natural language processing techniques. OpenAI's GPT-4 is used to generate summary sentences.

[0602] 4. Academic database: A collection of data containing related academic papers. Academic data can be obtained using APIs such as PubMed and arXiv.

[0603] System Operation

[0604] A user enters a specific keyword into the search bar of a smartphone app, for example, "application of generative models," and clicks the search button. This query is sent to the server.

[0605] The server calls the API of an academic database based on the received query to obtain a list of related papers, which it then passes to a generative model (GPT-4) to generate summaries of each paper.

[0606] The generated summaries are returned to the server, where a trend analysis is performed based on the metadata of each paper (number of citations, year of publication, main keywords, etc.). The progress of research and popular technologies are analyzed, and the results are generated as a trend report.

[0607] Finally, the server sends the summary and trend report to the user's terminal, where the user can visually check this information. For example, a summary list is displayed, and below it, visualized information such as trend graphs and keyword frequency rates can be viewed.

[0608] Example sentences

[0609] Example prompt sentence:

[0610] Summarize the following academic paper:

[0611] Title: Generative Models and Their Applications

[0612] Abstract: This paper explores the various applications of generative models, including their usage in image generation, natural language processing, and other fields...

[0613] This system allows researchers to quickly identify relevant academic papers and efficiently obtain their summaries and trends. For example, if a user searches for "applications of generative models," the system will display the latest related research papers along with their summaries. Furthermore, based on trend analysis, recent research trends can be visually confirmed, allowing researchers to efficiently grasp the progress of research.

[0614] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0615] Step 1:

[0616] The user enters a query

[0617] A user enters a topic or keyword into the search bar of a smartphone app, for example, "application of generative models." When the user clicks the search button, the query is sent from the device to the server.

[0618] Input: Search keywords entered by the user (e.g., "application of generative models")

[0619] Output: The search query sent to the server

[0620] Step 2:

[0621] The server receives the query and searches the academic database.

[0622] The server receives the search query sent from the device and calls the API of an academic database based on the query to search for related academic papers. The server uses the API of PubMed or arXiv, for example.

[0623] Input: The received search query

[0624] Data processing: Calling APIs of academic databases based on queries

[0625] Output: A list of academic papers obtained as search results

[0626] Step 3:

[0627] Pass the obtained list of academic papers to the generative model

[0628] The server passes each paper in the list of academic papers to a generative model (GPT-4) to generate a summary. The generative model analyzes the content of each paper and generates a summary containing the main points.

[0629] Input: List of academic papers

[0630] Data Computation: Generating summaries using generative models

[0631] Output: Abstract of each paper

[0632] Step 4:

[0633] Conduct trend analysis based on the generated summaries and metadata

[0634] The server analyzes research trends based on the generated abstracts and each paper's metadata (year of publication, number of citations, major keywords, etc.), for example, by analyzing the frequency of keyword appearances and the year of publication of the paper.

[0635] Input: Abstract and article metadata

[0636] Data Calculation: Trend Analysis through Metadata Analysis

[0637] Output: Trend report

[0638] Step 5:

[0639] Sends summary and trend reports to user terminals for display

[0640] The server formats the generated summary and trend report and sends them to the user's terminal, which receives and visually displays them, for example, visualizing the summary list along with trend graphs and keyword frequency rates.

[0641] Input: Abstract and trend report

[0642] Data processing: formatting summaries and trend reports

[0643] Output: Information displayed on the user's terminal (summary list, trend graph, etc.)

[0644] 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.

[0645] The present invention combines an emotion engine with a system that efficiently finds highly relevant academic papers from a large number of papers and provides summaries and research trends. This makes it possible to recognize the user's emotions and provide optimal information based on those emotions. Specific embodiments of the present invention are described below.

[0646] System configuration

[0647] This system consists of a user terminal, a server, a generative model, an academic database, and an emotion engine.

[0648] 1. User terminal: Provides an interface where users can input topics and keywords and obtain emotional information.

[0649] 2. Server: Processes queries and sentiment information, searches academic papers, generates summaries, and performs trend analysis.

[0650] 3. Generative model: Generate summaries of academic papers using natural language processing techniques.

[0651] 4. Academic database: A collection of data containing relevant academic papers.

[0652] 5. Emotion engine: A program that recognizes and analyzes emotions from user input, facial expressions, voice, etc.

[0653] Program processing

[0654] 1. User enters a query and sentiment is read (on device):

[0655] The user enters a search query such as "application of generative model" on the device. The device is also equipped with an emotion engine that reads the user's emotions in real time from their facial expressions, voice, and input content. The read emotion information is then sent to the server along with the query.

[0656] 2. The server receives the query and emotion information:

[0657] The server receives the search query and sentiment information sent by the user, and then executes the search query against the appropriate academic database based on the received query and sentiment information.

[0658] 3. Searching academic databases:

[0659] The server queries academic databases to retrieve a list of relevant academic papers, returning metadata such as the paper's title, author, publication year, abstract, and number of citations.

[0660] 4. Summary generation using generative models:

[0661] The server then passes the retrieved list of papers to a generative model, which uses natural language processing techniques to generate a summary of each paper, including the main points and conclusions of the paper.

[0662] 5. Trend analysis taking into account sentiment information:

[0663] The server analyzes research trends by taking into account emotional information along with the generated summaries and associated metadata. For example, the content of academic papers is filtered to match the user's interests and state. If the emotional engine recognizes the user's state of excitement, it will prioritize providing trend information on the latest technological innovations.

[0664] 6. Displaying the results:

[0665] The server then sends the generated summary and a trend report adjusted based on the sentiment information to the user's device. The device then displays the received information in a format that is easy for the user to understand, such as a graph visualizing the trend analysis along with the summary list.

[0666] Specific examples

[0667] The user searches for "applications of generative models." The user's emotions are also recognized.

[0668] 1. Device: The user enters "Application of generative model" and clicks the search button. At the same time, the emotion engine detects the user's excitement state.

[0669] 2. Server: Receives the query along with the emotion information and searches the academic database to obtain a list of relevant papers on "applications of technology X."

[0670] 3. Summary generation: The server uses the generative model to generate a summary of the paper, such as "This paper demonstrates that technology X outperforms conventional technologies."

[0671] 4. Sentiment Analysis: Taking into account the excitement level of users, trend analysis prioritizes trend information on the latest technological innovations.

[0672] 5. Result display: A summary and trend report is displayed on the user's terminal, highlighting the latest technological information taking into account the user's excitement.

[0673] Through these steps, users can efficiently find relevant academic papers, quickly understand summaries and trends, and effectively utilize information according to their emotional state.

[0674] The processing flow will be explained below.

[0675] Step 1:

[0676] The user enters a search query using the device's interface. For example, they enter "application of generative models" and click the search button. The device's built-in emotion engine then reads the user's emotions in real time from their facial expressions and voice.

[0677] Step 2:

[0678] The device sends the search query and emotion information entered by the user to the server. Specifically, the device sends the query and emotion information together as an HTTP request to the server.

[0679] Step 3:

[0680] The server analyzes the received query and sentiment information. Based on the query, the server searches academic databases to retrieve a list of relevant academic papers. The search results include metadata such as the paper title, author, publication year, abstract, and number of citations.

[0681] Step 4:

[0682] The server analyzes the list of academic papers it has acquired and passes each paper's metadata and full text to a generative model. The generative model then uses natural language processing technology to generate a summary of each paper. For example, it generates a summary such as, "This paper demonstrates that technology X outperforms conventional technology."

[0683] Step 5:

[0684] The server performs trend analysis based on the generated summary, metadata, and emotion information. For example, if the emotion engine recognizes the user's excitement, it will prioritize analyzing trend information related to the latest technological innovations.

[0685] Step 6:

[0686] As a result of the trend analysis, the server generates a trend report, such as "The application of generative models has developed rapidly in recent years, with technology X attracting particular attention."

[0687] Step 7:

[0688] The server sends the generated summary and trend report to the user's device, formats it in a visually optimized format that takes into account emotional information, and then transmits the data.

[0689] Step 8:

[0690] The terminal displays the received summary and trend report to the user. The user interface displays summary lists and graphs that visualize trend analysis, highlighting important information that corresponds to the user's excitement level. The user can then use this information for further detailed research and discussion.

[0691] As described above, the processing flow of the system combined with the emotion engine enables users to more efficiently find relevant academic papers, quickly understand summaries and trend information, and obtain the most appropriate information according to their emotional state.

[0692] Example 2

[0693] 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."

[0694] Conventional academic resource search systems search academic databases based on user-entered search queries to identify relevant materials, but they have the problem of being unable to provide information that takes into account the user's emotions and interests. This makes it impossible to provide appropriate information or analyze trends according to the user's emotional state, making it difficult to improve user satisfaction and convenience.

[0695] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0696] In this invention, the server includes means for receiving a search query entered by a user, means for searching an academic database based on the search query to identify relevant academic materials, means for using a generative model to generate summaries of the identified academic materials, means for analyzing research trends based on the summaries and associated metadata, means for recognizing user sentiment and adjusting the search query and trend analysis based on that information, and means for displaying the summaries and trend analysis results to the user, thereby providing optimal information according to the user's sentiment, improving the accuracy of search results and trend analysis and user satisfaction.

[0697] A "user" is an entity that uses this system to search for academic materials and obtain information.

[0698] A "search query" is a theme or keyword for academic materials to be searched that a user enters into a terminal.

[0699] An "academic database" is a collection of data that collects and manages academic materials, such as academic papers and research results.

[0700] "Academic materials" refers to documents such as papers, research reports, and presentations stored in academic databases.

[0701] A "generative model" is an algorithm or system that uses artificial intelligence and natural language processing techniques to analyze text data and generate a summary.

[0702] An "abstract" is a document that succinctly summarizes the main points and conclusions of academic material.

[0703] "Metadata" refers to data that indicates information related to academic materials, such as title, author, year of publication, and number of citations.

[0704] A "trend" refers to recent developments or popular trends in a particular academic field or topic.

[0705] "Emotion" refers to the user's psychological state and emotional response, and in this system, it is recognized from facial expressions, voice, input content, etc.

[0706] "Emotion information" is data about the user's emotional state obtained by the emotion engine.

[0707] An "emotion engine" is a program and system that recognizes and analyzes emotions from a user's facial expressions, voice, and input content.

[0708] A "terminal" is a device through which a user enters a search query and receives results, and includes, for example, a personal computer or smartphone.

[0709] The "server" is a central processing unit that processes search queries and sentiment information from users, searches academic databases, generates summaries, and performs trend analysis.

[0710] The present invention combines a system that efficiently finds highly relevant information from a large amount of academic materials and provides summaries and research trends with an emotion engine that recognizes the user's emotions. This makes it possible to provide optimal information tailored to the user's emotions. Specific embodiments of the present invention are described below.

[0711] System configuration

[0712] This system consists of a user terminal, a server, a generative model, an academic database, and an emotion engine.

[0713] 1. User Device:

[0714] It provides an interface for users to input search queries and obtain emotion information. For example, it is a PC or smartphone. The device is equipped with a camera and microphone, and has the ability to analyze the user's facial expressions and voice using an emotion engine.

[0715] 2. Server:

[0716] It is a central device that processes search queries and emotional information sent by users. Specifically, it receives queries and emotional information, searches academic databases, generates abstracts of papers using generative models, and performs trend analysis taking into account the emotional information.

[0717] 3. Generative Model:

[0718] These are algorithms and systems that use natural language processing techniques to generate summaries of academic materials, such as generative AI models using the Transformer architecture.

[0719] 4. Academic databases:

[0720] A collection of data that contains related scholarly materials, such as electronic journals and digital libraries.

[0721] 5. Emotion Engine:

[0722] It is a program and system that recognizes and analyzes emotions from user input, facial expressions, voice, etc. The emotion engine uses a regular camera and microphone to acquire and analyze emotional information in real time.

[0723] Program processing

[0724] When a user enters a search query, the device receives the query and simultaneously acquires the user's emotional information through the emotion engine. The acquired query and emotional information are then sent to the server, which then executes the search query against the academic database.

[0725] The list of academic materials obtained as search results is summarized by a generative model on the server, and trend analysis is then performed based on the generated summaries, metadata, and user sentiment information to uniquely provide the most relevant information.

[0726] Specific examples

[0727] When a user searches for "application of generative modeling," the user enters "application of generative modeling" into the device, and the emotion engine detects the user's excitement. The device then sends the search query and emotion information to the server.

[0728] The server searches academic databases based on the search query "application of generative models" to retrieve relevant materials. The retrieved materials are summarized using a generative model. For example, a summary such as "This research proposes a new application method using generative models" is generated.

[0729] The server then performs trend analysis to prioritize the provision of trend information on the latest technological innovations, taking into account the user's excitement state detected by the emotion engine. The results of the trend analysis and a summary are finally sent to the terminal.

[0730] The terminal receives this information and displays it to the user. The user interface displays summary lists and visualizes the latest technology trends in graphs and charts.

[0731] Prompt Sentence Examples

[0732] "Design a system to search the latest academic literature on the application of generative models and provide optimal summaries and trend information to excited users."

[0733] In this way, the system provides users with the most appropriate information based on their emotional state, allowing them to efficiently search and understand academic materials. It also helps users understand the latest research trends through trend analysis.

[0734] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0735] Step 1:

[0736] User enters search query and acquires emotion information (device)

[0737] A user types "application of generative models" into the search bar of their device and presses the Enter key. Pressing the Enter key triggers the device's built-in emotion engine, which uses the camera and microphone to analyze the user's facial expressions and voice in real time. The input data is the search query "application of generative models" and emotion information (e.g., excited state) acquired by the emotion engine. The emotion engine then outputs the results of the user's facial expression analysis and generates a data packet that is sent to the server along with the search query.

[0738] Step 2:

[0739] Send query and emotion information to the server (device)

[0740] The device sends the search query "Application of generative model" and the acquired emotion information together in a single data packet to the server. The transmission is triggered when the user presses the Enter key. The data packet contains the search keyword and the user's emotion (e.g., excitement) and is transferred to the server using a transmission protocol. The server is then ready to analyze the received data.

[0741] Step 3:

[0742] Receive query and emotion information (server)

[0743] The server receives the data packet sent from the terminal. The received data packet is decoded by the server's receiving module, and the search query "Application of generative model" and emotional information (e.g., excitement state) are separated. The server then passes the received search query to the analysis module, and the emotional information is passed to the emotion analysis module.

[0744] Step 4:

[0745] Search academic databases (server)

[0746] The server's query engine queries the received search query "application of generative models" to the academic database. The input data is the search query, and the database query execution engine executes the query. As a result, relevant paper data (metadata such as title, author, publication year, abstract, and number of citations) from the academic database is output as search results.

[0747] Step 5:

[0748] Generate a summary based on the search results (server)

[0749] The server passes search results obtained from the academic database to the generative model, which uses natural language processing techniques to generate summaries of the papers. The input data is a list of related papers and their metadata, which the generative model analyzes to create a concise summary. The output is a summary such as, "This research proposes a new application method using the generative model."

[0750] Step 6:

[0751] Conduct trend analysis taking into account emotional information (server)

[0752] The server performs trend analysis using the generated paper summaries and related metadata, as well as user sentiment information. The input data are the abstracts, metadata, and sentiment information. The sentiment analysis engine takes into account the user's state of excitement and prioritizes the extraction of trend information related to the latest technological innovations. The output is a trend report optimized for user sentiment.

[0753] Step 7:

[0754] Send the final result to the user terminal (server)

[0755] The server then combines the generated abstracts and the adjusted trend report into a single data packet and sends it to the user's device. The transmission protocol is HTTP or WebSocket. The data packet contains a list of abstracts and a trend report, allowing the user to receive the latest research trends and abstracts.

[0756] Step 8:

[0757] Display the results to the user (terminal)

[0758] The terminal receives the data packets sent from the server, analyzes the received summary statements and trend reports, and displays them. The user interface displays the summary list in an easy-to-read format and visualizes the latest technology trend information in graphs and charts. This is a step that allows users to efficiently obtain information and deepen their understanding.

[0759] (Application example 2)

[0760] 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."

[0761] Currently, security measures rely mainly on pre-defined rules or manual user intervention, making it difficult to implement appropriate measures that reflect the user's psychological state in real time. While existing systems identify relevant materials from large volumes of academic documents and generate summaries, they do not present search results or countermeasures that take the user's emotional state into account. As a result, it is difficult for users to obtain optimal information and countermeasures in situations where they feel stressed or tense.

[0762] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving a search query entered by a user, means for searching an academic database based on the search query and identifying relevant academic materials, means for using a generative model to generate a summary of the identified academic materials, means for analyzing research trends based on the generated summary and related information, means for displaying the summary and trend analysis results to the user, and means for recognizing the user's emotional state using an emotion engine and proposing and implementing search results and security measures based on that. This makes it possible to provide optimal information and security measures in real time according to the user's psychological state.

[0763] A "search query" is a keyword or phrase that a user enters into a search system.

[0764] An "academic database" is a collection of information that contains academic materials.

[0765] "Academic materials" are documents containing academic information, such as research and academic papers.

[0766] A "generative model" is an algorithm that uses natural language processing techniques to automatically perform specific tasks.

[0767] "Natural language processing technology" is an artificial intelligence technology that has the ability to understand and generate human language.

[0768] An "abstract" is a sentence that succinctly summarizes the main points and conclusions of a document.

[0769] "Related information" refers to the metadata and contextual information that accompanies scholarly documents and abstracts.

[0770] "Trend analysis" is the process of identifying and analyzing patterns and trends based on data.

[0771] The "emotion engine" is a program that recognizes the user's emotional state from their facial expressions, voice, and input content.

[0772] An "emotional state" is the psychological state that a user is feeling at that moment.

[0773] "Search Results" are lists of relevant information returned by the system based on a search query.

[0774] "Security measures" are specific measures and procedures to ensure user safety.

[0775] "Suggestion and execution" means presenting the optimal option to the user and automatically implementing that option if necessary.

[0776] This invention is a system that recognizes a user's emotional state in real time and then proposes and implements relevant information and security measures based on that information. Below, we will explain in detail the program and processing of the system that realizes this application example.

[0777] System Configuration

[0778] This system consists of a user terminal, a server, a generative model, an academic database, and an emotion engine.

[0779] 1. User Device

[0780] It provides an interface for users to enter search queries and obtain emotion information.

[0781] Hardware used: Smartphone, smart glasses

[0782] Software used: React Native and ARKit as front-end technologies

[0783] 2. Server

[0784] It processes query and sentiment information to search, summarize, and analyze trends in academic materials.

[0785] Software used: Node.js, Express.js, MongoDB, PostgreSQL

[0786] 3. Generative Model

[0787] Generate summaries of academic materials using natural language processing techniques.

[0788] Software used: Google Cloud Natural Language API, OpenAI GPT-3

[0789] 4. Academic Databases

[0790] A collection of data containing related academic materials.

[0791] Software used: MongoDB, PostgreSQL

[0792] 5. Emotion Engine

[0793] A program that recognizes and analyzes emotions from the user's facial expressions, voice, and input content.

[0794] Software used: IBM Watson, Microsoft Azure Cognitive Services

[0795] A natural language description of the process

[0796] Acquiring emotion data

[0797] The user device captures the user's facial expressions using a camera on a smartphone or smart glasses, and records their voice using a microphone. This allows for real-time emotional data acquisition, and the emotional data from the facial and vocal signals is analyzed using an emotion engine (IBM Watson or Microsoft Azure Cognitive Services).

[0798] Determining the risk level

[0799] The server evaluates the emotional data obtained from the emotion engine in real time, and if negative emotions such as tension or anxiety are strong, it judges the risk level to be high.

[0800] Proposing and implementing measures

[0801] If the risk level is determined to be high, the server will send a push notification to the user suggesting specific countermeasures. For example, it may suggest security measures such as "locking the front door." If the user allows "auto-execution," the specified security measures will be automatically executed. At this time, the server performs operations through an API that controls smart home devices (Google Home, Amazon Alexa, etc.).

[0802] Specific examples

[0803] When the user returns home late at night, the user device detects a state of tension. The server determines the security risk level to be high and notifies the user as follows: "You are feeling tense. Do you want to lock the front door?" and "Do you want to allow automatic locking?"

[0804] Prompt Sentence Examples

[0805] "If users are nervous, suggest security measures."

[0806] "Please judge the risk level based on the following emotional data. Emotional data: [Facial expression: 'Tension', Audio: 'Anxiety']"

[0807] This makes it possible to provide optimal information and security measures in real time according to the user's psychological state.

[0808] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0809] Step 1:

[0810] The user enters a search query.

[0811] Input: The user enters a "search query" into the device.

[0812] How it works: The device receives a user's search query and prepares to process it.

[0813] Output: Search query data from the user

[0814] Step 2:

[0815] The emotion engine reads the user's emotions in real time.

[0816] Input: Video and audio data captured from the camera and microphone on the user's device.

[0817] How it works: The device sends video and audio data to the emotion engine, which then analyzes it.

[0818] Output: Emotional state data (e.g., tension, anxiety)

[0819] Step 3:

[0820] The server receives the search query and the emotional state.

[0821] Input: User search query data and emotional state data.

[0822] Operation: The device sends this data to the server.

[0823] Output: Search query data and emotional state data are passed to the server.

[0824] Step 4:

[0825] The server performs searches against academic databases.

[0826] Input: Search query data.

[0827] How it works: The server searches academic databases to identify relevant academic materials.

[0828] Output: List of related academic materials

[0829] Step 5:

[0830] A generative model generates summaries of relevant academic material.

[0831] Input: List of relevant academic resources.

[0832] How it works: The server passes a list of materials to the generative model, which then generates a summary using natural language processing techniques.

[0833] Output: Generated summary

[0834] Step 6:

[0835] The server analyzes research trends based on the generated abstracts and associated metadata.

[0836] Input: Generated summary sentences and associated metadata, emotional state data.

[0837] How it works: The server aggregates this data and uses trend analysis algorithms to analyze trends.

[0838] Output: Trend analysis results

[0839] Step 7:

[0840] Determine security risk levels based on emotional states.

[0841] Input: Emotional state data.

[0842] Operation: The server evaluates the emotional state data and determines the risk level.

[0843] Output: Risk level data (e.g. high, medium, low)

[0844] Step 8:

[0845] If the security risk level is determined to be high, countermeasures will be proposed.

[0846] Input: Risk level data.

[0847] What it does: If the server determines a high risk, it will send a notification to the user suggesting specific security measures.

[0848] Output: Specific action suggestions (e.g., "Would you like to lock your front door?")

[0849] Step 9:

[0850] If the user allows automatic execution, the specified security measures are implemented.

[0851] Input: User authorization data.

[0852] How it works: The server receives the user's permission and executes the measures through the control API of the smart home device.

[0853] Output: Security measures taken (e.g., front door automatically locks)

[0854] Through the above steps, optimal information provision and security measures are automatically implemented in real time according to the user's emotions.

[0855] 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.

[0856] 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.

[0857] 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.

[0858] [Third embodiment]

[0859] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0860] 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.

[0861] 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).

[0862] 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.

[0863] 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.

[0864] 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).

[0865] 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.

[0866] 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.

[0867] 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.

[0868] 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.

[0869] 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.

[0870] 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."

[0871] The present invention relates to an AI system that efficiently finds highly relevant academic papers from a large number of papers and provides their summaries and research trends. Specific embodiments for carrying out the present invention will be described below.

[0872] System configuration

[0873] This system consists of a user terminal, a server, a generative model, and an academic database.

[0874] 1. User terminal: Provides an interface for users to input topics and keywords.

[0875] 2. Server: Processes input queries, searches academic papers, generates summaries, and performs trend analysis.

[0876] 3. Generative model: Generate summaries of academic papers using natural language processing techniques.

[0877] 4. Academic database: A collection of data containing relevant academic papers.

[0878] Program processing

[0879] 1. User enters query (terminal):

[0880] A user inputs a specific topic or keyword, such as "application of generative modeling," through the interface of their device. When the user clicks the search button, the input query is sent to the server.

[0881] 2. The server receives the query:

[0882] The server receives and interprets queries submitted by users, and based on these queries, the server searches academic databases to find relevant academic papers.

[0883] 3. Searching academic databases:

[0884] The server queries academic databases to retrieve a list of relevant academic papers, returning metadata such as the paper's title, author, publication year, abstract, and number of citations.

[0885] 4. Summary generation using generative models:

[0886] The server then passes the retrieved list of papers to a generative model, which then uses natural language processing techniques to generate a summary of each paper, including the paper's main points and conclusions.

[0887] 5. Trend Analysis:

[0888] The server analyzes research trends based on the generated abstracts and associated metadata, taking into account the year of publication, number of citations, and key keywords of the papers. For example, a trend report can be generated showing that "Technology X" is attracting particular attention in recent research.

[0889] 6. Displaying the results:

[0890] The server formats the generated summary and trend analysis results and sends them back to the user's terminal, where the user's terminal displays the received information in a format that the user can visually confirm, such as by displaying a graph that visualizes the trend analysis along with the summary list.

[0891] Specific examples

[0892] A user searches for "applications of generative models."

[0893] 1. Device: The user enters "application of generative models" in the search bar and clicks the search button.

[0894] 2. Server: Receives the query and searches academic databases to retrieve a list of relevant papers on "applications of technology X."

[0895] 3. Summary generation: The server uses the generative model to generate a summary of the paper, creating a summary statement such as, "This paper shows that technology X outperforms conventional technology."

[0896] 4. Trend analysis: The server generates a trend report based on the publication year, number of citations, and major keywords, stating that "the application of generative models has developed rapidly in recent years."

[0897] 5. Results display: A summary and trend report will be displayed on the user's terminal, allowing the user to conduct further research and discussion based on the information.

[0898] As described above, the system of the present invention efficiently searches for related academic papers and provides their summaries and trend information, thereby enabling researchers and students to quickly obtain the information they need.

[0899] The processing flow will be explained below.

[0900] Step 1:

[0901] The user inputs a search query using the interface of their device. For example, the user inputs "application of generative model" as the search query and clicks the search button.

[0902] Step 2:

[0903] The terminal sends the query entered by the user to the server. Specifically, the terminal sends the query as an HTTP request to the server.

[0904] Step 3:

[0905] The server analyzes the received query and executes a search query against the appropriate academic database, which retrieves a list of relevant papers.

[0906] Step 4:

[0907] The server analyzes the list of academic papers it has obtained and extracts the metadata for each paper (title, author, year of publication, abstract, number of citations, etc.).

[0908] Step 5:

[0909] The server passes the extracted metadata and full text of the papers to a generative model, which uses natural language processing techniques to generate a summary of each academic paper.

[0910] Step 6:

[0911] The generative model returns a summary of each paper to the server, including its main points and conclusions. For example, it generates a summary such as, "The effectiveness of new generative model technology X has been demonstrated."

[0912] Step 7:

[0913] The server performs trend analysis based on the generated abstracts and metadata, taking into account the year of publication, number of citations, and main keywords of the papers to identify research trends.

[0914] Step 8:

[0915] Based on the results of the trend analysis, the server generates a trend report, such as "The application of generative models has developed rapidly in recent years, with technology X attracting particular attention."

[0916] Step 9:

[0917] The server sends the generated summary and trend report to the user's terminal, formatted in a user-friendly format.

[0918] Step 10:

[0919] The terminal displays the received summary and trend report to the user. The user interface displays a summary list and visualized trend graphs. The user can then conduct further research based on this information.

[0920] Through the above steps, users can efficiently find relevant academic papers, quickly understand summaries and trends, and effectively utilize information.

[0921] Example 1

[0922] 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."

[0923] Conventional academic paper search systems make it difficult to efficiently find relevant papers, and it is also difficult to quickly understand the content of each paper. Furthermore, they lack the functionality to analyze and provide research trends, making it difficult for researchers and students to quickly obtain the information they need. For this reason, there was a need to develop a system that could efficiently extract necessary knowledge from vast amounts of information and provide it in an easy-to-understand manner.

[0924] 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.

[0925] In this invention, the server includes a means for a user to input topics and keywords, a means for sending a query to the server, a means for receiving and analyzing the query, a means for searching academic databases and retrieving related academic papers, a means for generating summaries of the papers using a generative model, a means for analyzing research trends based on the generated summaries and related metadata, and a means for formatting the summaries and trend analysis results and transmitting them to a user terminal. This makes it possible to efficiently search for highly relevant academic papers and quickly provide their summaries and trend information.

[0926] - "Query" refers to the string of characters or keywords that a user enters to search for specific information.

[0927] "Server" refers to a computer system that receives queries from users, processes them, and provides appropriate search results, summaries, and trend analysis results.

[0928] "User Terminal" means the computer or mobile device used by a User to access the System, enter queries, and view results.

[0929] An "academic database" refers to a collection of data that contains and is used to search academic papers and related information.

[0930] A "generative model" refers to an algorithm or software that uses natural language processing techniques to automatically generate summaries of academic papers.

[0931] An "abstract" is a piece of text that concisely summarizes the main points and conclusions of an academic paper.

[0932] "Trend analysis" refers to the process of identifying recent trends and hot topics in a research field based on the metadata of retrieved academic papers.

[0933] "Metadata" refers to information about the data itself, such as the paper's title, author, publication year, abstract, and number of citations.

[0934] "HTTP request" refers to the protocol used by a user terminal to request data or services from a server.

[0935] "Natural language processing technology" refers to the techniques and methods that allow computers to understand, generate, and analyze human language.

[0936] This system efficiently finds related academic papers by allowing users to input topics and keywords, and provides summaries and research trends. Specifically, it consists of a user terminal, a server, a generative model, and an academic database.

[0937] User terminal

[0938] The user device provides an interface for users to input queries. This is implemented as a web or mobile application and is designed to allow users to easily input topics and keywords. For example, a user might input the query "application of generative AI model."

[0939] server

[0940] The server receives queries sent from user devices and analyzes their contents. Based on these queries, the server searches academic databases to find relevant academic papers. Academic databases often use SQL Server or Elasticsearch.

[0941] The obtained list of papers is passed to a generative model, which generates a summary of each paper. The generative model uses natural language processing techniques such as the Transformer or BERT model. The generated summaries are then stored on the server.

[0942] The server then analyzes the abstracts and associated metadata to identify research trends, taking into account factors such as the year of publication, number of citations, and key keywords of the papers. For example, it can generate a trend report that states, "The application of generative AI models has rapidly increased in recent years."

[0943] Trend analysis

[0944] The server formats the generated summary and trend analysis results and sends them to the user's terminal, where the user can visually check the results, such as a summary list or a graph visualizing the trend analysis.

[0945] Specific examples

[0946] The following is the series of processes that a user goes through when searching for "applications of generative models."

[0947] 1. User: The user enters "application of generative models" in the search bar and clicks the search button.

[0948] 2. Server: Receives the query and searches academic databases to obtain a list of relevant papers on "applications of generative models."

[0949] 3. Generative model: The server uses the generative model to generate a summary of the paper, creating a summary statement such as, "This paper demonstrates that generative models outperform conventional techniques."

[0950] 4. Trend analysis: The server generates a trend report based on the publication year, number of citations, and major keywords, stating that "the application of generative models has developed rapidly in recent years."

[0951] 5. Display of results: A summary and trend report is displayed on the user's terminal, allowing the user to conduct further research or discussion based on the information.

[0952] Example prompts to input to a generative AI model:

[0953] "Please search for the latest academic papers on applications of generative models and provide summaries and research trends."

[0954] This system makes it possible to efficiently search for relevant academic papers and quickly provide their summaries and trend information, enabling researchers and students to quickly and accurately obtain the information they need.

[0955] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0956] Step 1:

[0957] Query input (user)

[0958] A user enters a specific topic or keyword into the search field of a web application that serves as the interface for their device, for example, "applications of generative AI models," and clicks the search button. This action stacks the query locally on the device, ready to be sent.

[0959] Input: User-entered topics or keywords (e.g., "applications of generative AI models")

[0960] Output: Query data sent by the device to the server

[0961] Step 2:

[0962] Query sending (terminal)

[0963] The device generates an HTTP request containing the query entered by the user and sends it to the server's API endpoint, along with the user's session information and a timestamp.

[0964] Input: User-entered query, session information, timestamp

[0965] Output: HTTP request sent to the server

[0966] Step 3:

[0967] Query reception and interpretation (server)

[0968] The server analyzes the received HTTP request and extracts the query portion. It obtains the query "application of generative AI model" from the received request and normalizes the query using natural language processing. This analysis is performed using a text analysis library or similar.

[0969] Input: Query data in the HTTP request

[0970] Output: Normalized query

[0971] Step 4:

[0972] Academic database search (server)

[0973] The server searches academic databases using normalized queries, executing SQL or Elasticsearch queries to filter and retrieve relevant academic papers.

[0974] Input: Normalized query

[0975] Output: A search result list of relevant academic papers

[0976] Step 5:

[0977] Obtaining a list of papers (server)

[0978] The server creates a list of relevant papers based on the search results obtained from academic databases. Specifically, it collects metadata such as the paper's title, author, publication year, abstract, and number of citations, and compiles it into a list.

[0979] Input: Search results from academic databases

[0980] Output: List of article metadata

[0981] Step 6:

[0982] Paper summary generation (server / generative model)

[0983] The server generates summaries of each paper using generative models, such as the Transformer and BERT models, which use natural language processing techniques, to generate summaries that include the main points and conclusions of each paper.

[0984] Input: Article metadata

[0985] Output: Generated paper abstract

[0986] Step 7:

[0987] Trend analysis (server)

[0988] The server analyzes research trends based on the generated abstracts and related metadata, taking into account the year of publication, number of citations, and key keywords of the paper, and generates a trend report such as, "The application of generative AI models has rapidly increased in recent years."

[0989] Input: Generated paper abstracts and metadata

[0990] Output: Trend report

[0991] Step 8:

[0992] Result formatting and sending (server)

[0993] The server formats the generated summary and trend analysis results and sends them to the user's device. The formatted information is encoded in, for example, JSON format and included in the HTTP response.

[0994] Input: Summary and trend analysis results

[0995] Output: HTTP response sent to the user device

[0996] Step 9:

[0997] Result display (terminal)

[0998] The terminal analyzes the results received from the server and displays them visually to the user, who can view summary lists and graphs visualizing trend analysis.

[0999] Input: HTTP response from the server

[1000] Output: On-screen summary list and trend analysis results

[1001] (Application example 1)

[1002] 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."

[1003] Currently, it takes a great deal of time and effort for researchers to efficiently find relevant papers from a large number of academic papers and analyze their summaries and research trends. Furthermore, there are only a limited number of systems that automatically generate paper summaries and display them visually, and convenience is particularly required on mobile devices such as smartphones. Furthermore, there is a lack of trend analysis to track research progress. To solve these problems, a system is needed that can quickly identify relevant academic papers and efficiently provide their summaries and trends.

[1004] 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.

[1005] In this invention, the server includes means for receiving a search query entered by a user, means for searching an academic database based on the search query to identify relevant academic papers, means for using a generative model to generate summaries of the identified academic papers, means for analyzing research trends based on the generated summaries and related data, means for visually displaying the summaries and trend analysis results to the user, means for passing the searched paper list to the generative model to generate summaries of each paper, and means for analyzing the number of citations and major keywords of each paper to detect research trends. This makes it possible to quickly identify relevant academic papers and efficiently provide their summaries and trends.

[1006] "Means for receiving a user-entered search query" refers to a device or software that provides an interface for a user to enter a topic or keyword they wish to search for and that is capable of receiving such input.

[1007] "Means for searching academic databases based on the search query and identifying relevant academic papers" refers to a device or software that searches academic databases based on a search query entered by a user and finds relevant academic papers.

[1008] "Means for using a generative model to generate summaries of the identified academic literature" refers to a device or software that uses AI technology known as a generative model to create summaries in order to automatically summarize the contents of the identified academic literature.

[1009] "Means for analyzing research trends based on the generated abstracts and associated data" refers to devices or software that use the generated abstracts and associated metadata to analyze movements and trends in a particular research field.

[1010] "Means for visually displaying the summary and trend analysis results to the user" refers to a device or software that displays the generated summary and trend analysis results on a screen in a format that is easy for the user to understand.

[1011] "Means for passing the searched paper list to a generative model and generating a summary of each paper" refers to a device or software that inputs the list of multiple papers obtained as search results into a generative model and generates a summary of each paper.

[1012] "Means for detecting research trends by analyzing the number of citations and major keywords of each document" refers to a device or software that analyzes the number of citations and major keywords of each paper and detects research trends based on that information.

[1013] The present invention relates to an AI system that efficiently finds highly relevant academic papers from a large number of papers and provides summaries and research trends. Specific embodiments for carrying out the present invention are described below.

[1014] System configuration

[1015] This system consists of a user terminal, a server, a generative model, and an academic database.

[1016] 1. User device: Provides an interface for users to input topics and keywords. Users can enter search queries through smartphone applications and visually check the results.

[1017] 2. Server: Processes input queries, searches academic papers, generates summaries, and performs trend analysis. The server is implemented using Node.js and Express.

[1018] 3. Generative model: Generate summaries of academic papers using natural language processing techniques. OpenAI's GPT-4 is used to generate summary sentences.

[1019] 4. Academic database: A collection of data containing related academic papers. Academic data can be obtained using APIs such as PubMed and arXiv.

[1020] System Operation

[1021] A user enters a specific keyword into the search bar of a smartphone app, for example, "application of generative models," and clicks the search button. This query is sent to the server.

[1022] The server calls the API of an academic database based on the received query to obtain a list of related papers, which it then passes to a generative model (GPT-4) to generate summaries of each paper.

[1023] The generated summaries are returned to the server, where a trend analysis is performed based on the metadata of each paper (number of citations, year of publication, main keywords, etc.). The progress of research and popular technologies are analyzed, and the results are generated as a trend report.

[1024] Finally, the server sends the summary and trend report to the user's terminal, where the user can visually check this information. For example, a summary list is displayed, and below it, visualized information such as trend graphs and keyword frequency rates can be viewed.

[1025] Example sentences

[1026] Example prompt sentence:

[1027] Summarize the following academic paper:

[1028] Title: Generative Models and Their Applications

[1029] Abstract: This paper explores the various applications of generative models, including their usage in image generation, natural language processing, and other fields...

[1030] This system allows researchers to quickly identify relevant academic papers and efficiently obtain their summaries and trends. For example, if a user searches for "applications of generative models," the system will display the latest related research papers along with their summaries. Furthermore, based on trend analysis, recent research trends can be visually confirmed, allowing researchers to efficiently grasp the progress of research.

[1031] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1032] Step 1:

[1033] The user enters a query

[1034] A user enters a topic or keyword into the search bar of a smartphone app, for example, "application of generative models." When the user clicks the search button, the query is sent from the device to the server.

[1035] Input: Search keywords entered by the user (e.g., "application of generative models")

[1036] Output: The search query sent to the server

[1037] Step 2:

[1038] The server receives the query and searches the academic database.

[1039] The server receives the search query sent from the device and calls the API of an academic database based on the query to search for related academic papers. The server uses the API of PubMed or arXiv, for example.

[1040] Input: The received search query

[1041] Data processing: Calling APIs of academic databases based on queries

[1042] Output: A list of academic papers obtained as search results

[1043] Step 3:

[1044] Pass the obtained list of academic papers to the generative model

[1045] The server passes each paper in the list of academic papers to a generative model (GPT-4) to generate a summary. The generative model analyzes the content of each paper and generates a summary containing the main points.

[1046] Input: List of academic papers

[1047] Data Computation: Generating summaries using generative models

[1048] Output: Abstract of each paper

[1049] Step 4:

[1050] Conduct trend analysis based on the generated summaries and metadata

[1051] The server analyzes research trends based on the generated abstracts and each paper's metadata (year of publication, number of citations, major keywords, etc.), for example, by analyzing the frequency of keyword appearances and the year of publication of the paper.

[1052] Input: Abstract and article metadata

[1053] Data Calculation: Trend Analysis through Metadata Analysis

[1054] Output: Trend report

[1055] Step 5:

[1056] Sends summary and trend reports to user terminals for display

[1057] The server formats the generated summary and trend report and sends them to the user's terminal, which receives and visually displays them, for example, visualizing the summary list along with trend graphs and keyword frequency rates.

[1058] Input: Abstract and trend report

[1059] Data processing: formatting summaries and trend reports

[1060] Output: Information displayed on the user's terminal (summary list, trend graph, etc.)

[1061] 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.

[1062] The present invention combines an emotion engine with a system that efficiently finds highly relevant academic papers from a large number of papers and provides summaries and research trends. This makes it possible to recognize the user's emotions and provide optimal information based on those emotions. Specific embodiments of the present invention are described below.

[1063] System configuration

[1064] This system consists of a user terminal, a server, a generative model, an academic database, and an emotion engine.

[1065] 1. User terminal: Provides an interface where users can input topics and keywords and obtain emotional information.

[1066] 2. Server: Processes queries and sentiment information, searches academic papers, generates summaries, and performs trend analysis.

[1067] 3. Generative model: Generate summaries of academic papers using natural language processing techniques.

[1068] 4. Academic database: A collection of data containing relevant academic papers.

[1069] 5. Emotion engine: A program that recognizes and analyzes emotions from user input, facial expressions, voice, etc.

[1070] Program processing

[1071] 1. User enters a query and sentiment is read (on device):

[1072] The user enters a search query such as "application of generative model" on the device. The device is also equipped with an emotion engine that reads the user's emotions in real time from their facial expressions, voice, and input content. The read emotion information is then sent to the server along with the query.

[1073] 2. The server receives the query and emotion information:

[1074] The server receives the search query and sentiment information sent by the user, and then executes the search query against the appropriate academic database based on the received query and sentiment information.

[1075] 3. Searching academic databases:

[1076] The server queries academic databases to retrieve a list of relevant academic papers, returning metadata such as the paper's title, author, publication year, abstract, and number of citations.

[1077] 4. Summary generation using generative models:

[1078] The server then passes the retrieved list of papers to a generative model, which uses natural language processing techniques to generate a summary of each paper, including the main points and conclusions of the paper.

[1079] 5. Trend analysis taking into account sentiment information:

[1080] The server analyzes research trends by taking into account emotional information along with the generated summaries and associated metadata. For example, the content of academic papers is filtered to match the user's interests and state. If the emotional engine recognizes the user's state of excitement, it will prioritize providing trend information on the latest technological innovations.

[1081] 6. Displaying the results:

[1082] The server then sends the generated summary and a trend report adjusted based on the sentiment information to the user's device. The device then displays the received information in a format that is easy for the user to understand, such as a graph visualizing the trend analysis along with the summary list.

[1083] Specific examples

[1084] The user searches for "applications of generative models." The user's emotions are also recognized.

[1085] 1. Device: The user enters "Application of generative model" and clicks the search button. At the same time, the emotion engine detects the user's excitement state.

[1086] 2. Server: Receives the query along with the emotion information and searches the academic database to obtain a list of relevant papers on "applications of technology X."

[1087] 3. Summary generation: The server uses the generative model to generate a summary of the paper, such as "This paper demonstrates that technology X outperforms conventional technologies."

[1088] 4. Sentiment Analysis: Taking into account the excitement level of users, trend analysis prioritizes trend information on the latest technological innovations.

[1089] 5. Result display: A summary and trend report is displayed on the user's terminal, highlighting the latest technological information taking into account the user's excitement.

[1090] Through these steps, users can efficiently find relevant academic papers, quickly understand summaries and trends, and effectively utilize information according to their emotional state.

[1091] The processing flow will be explained below.

[1092] Step 1:

[1093] The user enters a search query using the device's interface. For example, they enter "application of generative models" and click the search button. The device's built-in emotion engine then reads the user's emotions in real time from their facial expressions and voice.

[1094] Step 2:

[1095] The device sends the search query and emotion information entered by the user to the server. Specifically, the device sends the query and emotion information together as an HTTP request to the server.

[1096] Step 3:

[1097] The server analyzes the received query and sentiment information. Based on the query, the server searches academic databases to retrieve a list of relevant academic papers. The search results include metadata such as the paper title, author, publication year, abstract, and number of citations.

[1098] Step 4:

[1099] The server analyzes the list of academic papers it has acquired and passes each paper's metadata and full text to a generative model. The generative model then uses natural language processing technology to generate a summary of each paper. For example, it generates a summary such as, "This paper demonstrates that technology X outperforms conventional technology."

[1100] Step 5:

[1101] The server performs trend analysis based on the generated summary, metadata, and emotion information. For example, if the emotion engine recognizes the user's excitement, it will prioritize analyzing trend information related to the latest technological innovations.

[1102] Step 6:

[1103] As a result of the trend analysis, the server generates a trend report, such as "The application of generative models has developed rapidly in recent years, with technology X attracting particular attention."

[1104] Step 7:

[1105] The server sends the generated summary and trend report to the user's device, formats it in a visually optimized format that takes into account emotional information, and then transmits the data.

[1106] Step 8:

[1107] The terminal displays the received summary and trend report to the user. The user interface displays summary lists and graphs that visualize trend analysis, highlighting important information that corresponds to the user's excitement level. The user can then use this information for further detailed research and discussion.

[1108] As described above, the processing flow of the system combined with the emotion engine enables users to more efficiently find relevant academic papers, quickly understand summaries and trend information, and obtain the most appropriate information according to their emotional state.

[1109] Example 2

[1110] 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."

[1111] Conventional academic resource search systems search academic databases based on user-entered search queries to identify relevant materials, but they have the problem of being unable to provide information that takes into account the user's emotions and interests. This makes it impossible to provide appropriate information or analyze trends according to the user's emotional state, making it difficult to improve user satisfaction and convenience.

[1112] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1113] In this invention, the server includes means for receiving a search query entered by a user, means for searching an academic database based on the search query to identify relevant academic materials, means for using a generative model to generate summaries of the identified academic materials, means for analyzing research trends based on the summaries and associated metadata, means for recognizing user sentiment and adjusting the search query and trend analysis based on that information, and means for displaying the summaries and trend analysis results to the user, thereby providing optimal information according to the user's sentiment, improving the accuracy of search results and trend analysis and user satisfaction.

[1114] A "user" is an entity that uses this system to search for academic materials and obtain information.

[1115] A "search query" is a theme or keyword for academic materials to be searched that a user enters into a terminal.

[1116] An "academic database" is a collection of data that collects and manages academic materials, such as academic papers and research results.

[1117] "Academic materials" refers to documents such as papers, research reports, and presentations stored in academic databases.

[1118] A "generative model" is an algorithm or system that uses artificial intelligence and natural language processing techniques to analyze text data and generate a summary.

[1119] An "abstract" is a document that succinctly summarizes the main points and conclusions of academic material.

[1120] "Metadata" refers to data that indicates information related to academic materials, such as title, author, year of publication, and number of citations.

[1121] A "trend" refers to recent developments or popular trends in a particular academic field or topic.

[1122] "Emotion" refers to the user's psychological state and emotional response, and in this system, it is recognized from facial expressions, voice, input content, etc.

[1123] "Emotion information" is data about the user's emotional state obtained by the emotion engine.

[1124] An "emotion engine" is a program and system that recognizes and analyzes emotions from a user's facial expressions, voice, and input content.

[1125] A "terminal" is a device through which a user enters a search query and receives results, and includes, for example, a personal computer or smartphone.

[1126] The "server" is a central processing unit that processes search queries and sentiment information from users, searches academic databases, generates summaries, and performs trend analysis.

[1127] The present invention combines a system that efficiently finds highly relevant information from a large amount of academic materials and provides summaries and research trends with an emotion engine that recognizes the user's emotions. This makes it possible to provide optimal information tailored to the user's emotions. Specific embodiments of the present invention are described below.

[1128] System configuration

[1129] This system consists of a user terminal, a server, a generative model, an academic database, and an emotion engine.

[1130] 1. User Device:

[1131] It provides an interface for users to input search queries and obtain emotion information. For example, it is a PC or smartphone. The device is equipped with a camera and microphone, and has the ability to analyze the user's facial expressions and voice using an emotion engine.

[1132] 2. Server:

[1133] It is a central device that processes search queries and emotional information sent by users. Specifically, it receives queries and emotional information, searches academic databases, generates abstracts of papers using generative models, and performs trend analysis taking into account the emotional information.

[1134] 3. Generative Model:

[1135] These are algorithms and systems that use natural language processing techniques to generate summaries of academic materials, such as generative AI models using the Transformer architecture.

[1136] 4. Academic databases:

[1137] A collection of data that contains related scholarly materials, such as electronic journals and digital libraries.

[1138] 5. Emotion Engine:

[1139] It is a program and system that recognizes and analyzes emotions from user input, facial expressions, voice, etc. The emotion engine uses a regular camera and microphone to acquire and analyze emotional information in real time.

[1140] Program processing

[1141] When a user enters a search query, the device receives the query and simultaneously acquires the user's emotional information through the emotion engine. The acquired query and emotional information are then sent to the server, which then executes the search query against the academic database.

[1142] The list of academic materials obtained as search results is summarized by a generative model on the server, and trend analysis is then performed based on the generated summaries, metadata, and user sentiment information to uniquely provide the most relevant information.

[1143] Specific examples

[1144] When a user searches for "application of generative modeling," the user enters "application of generative modeling" into the device, and the emotion engine detects the user's excitement. The device then sends the search query and emotion information to the server.

[1145] The server searches academic databases based on the search query "application of generative models" to retrieve relevant materials. The retrieved materials are summarized using a generative model. For example, a summary such as "This research proposes a new application method using generative models" is generated.

[1146] The server then performs trend analysis to prioritize the provision of trend information on the latest technological innovations, taking into account the user's excitement state detected by the emotion engine. The results of the trend analysis and a summary are finally sent to the terminal.

[1147] The terminal receives this information and displays it to the user. The user interface displays summary lists and visualizes the latest technology trends in graphs and charts.

[1148] Prompt Sentence Examples

[1149] "Design a system to search the latest academic literature on the application of generative models and provide optimal summaries and trend information to excited users."

[1150] In this way, the system provides users with the most appropriate information based on their emotional state, allowing them to efficiently search and understand academic materials. It also helps users understand the latest research trends through trend analysis.

[1151] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1152] Step 1:

[1153] User enters search query and acquires emotion information (device)

[1154] A user types "application of generative models" into the search bar of their device and presses the Enter key. Pressing the Enter key triggers the device's built-in emotion engine, which uses the camera and microphone to analyze the user's facial expressions and voice in real time. The input data is the search query "application of generative models" and emotion information (e.g., excited state) acquired by the emotion engine. The emotion engine then outputs the results of the user's facial expression analysis and generates a data packet that is sent to the server along with the search query.

[1155] Step 2:

[1156] Send query and emotion information to the server (device)

[1157] The device sends the search query "Application of generative model" and the acquired emotion information together in a single data packet to the server. The transmission is triggered when the user presses the Enter key. The data packet contains the search keyword and the user's emotion (e.g., excitement) and is transferred to the server using a transmission protocol. The server is then ready to analyze the received data.

[1158] Step 3:

[1159] Receive query and emotion information (server)

[1160] The server receives the data packet sent from the terminal. The received data packet is decoded by the server's receiving module, and the search query "Application of generative model" and emotional information (e.g., excitement state) are separated. The server then passes the received search query to the analysis module, and the emotional information is passed to the emotion analysis module.

[1161] Step 4:

[1162] Search academic databases (server)

[1163] The server's query engine queries the received search query "application of generative models" to the academic database. The input data is the search query, and the database query execution engine executes the query. As a result, relevant paper data (metadata such as title, author, publication year, abstract, and number of citations) from the academic database is output as search results.

[1164] Step 5:

[1165] Generate a summary based on the search results (server)

[1166] The server passes search results obtained from the academic database to the generative model, which uses natural language processing techniques to generate summaries of the papers. The input data is a list of related papers and their metadata, which the generative model analyzes to create a concise summary. The output is a summary such as, "This research proposes a new application method using the generative model."

[1167] Step 6:

[1168] Conduct trend analysis taking into account emotional information (server)

[1169] The server performs trend analysis using the generated paper summaries and related metadata, as well as user sentiment information. The input data are the abstracts, metadata, and sentiment information. The sentiment analysis engine takes into account the user's state of excitement and prioritizes the extraction of trend information related to the latest technological innovations. The output is a trend report optimized for user sentiment.

[1170] Step 7:

[1171] Send the final result to the user terminal (server)

[1172] The server then combines the generated abstracts and the adjusted trend report into a single data packet and sends it to the user's device. The transmission protocol is HTTP or WebSocket. The data packet contains a list of abstracts and a trend report, allowing the user to receive the latest research trends and abstracts.

[1173] Step 8:

[1174] Display the results to the user (terminal)

[1175] The terminal receives the data packets sent from the server, analyzes the received summary statements and trend reports, and displays them. The user interface displays the summary list in an easy-to-read format and visualizes the latest technology trend information in graphs and charts. This is a step that allows users to efficiently obtain information and deepen their understanding.

[1176] (Application example 2)

[1177] 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."

[1178] Currently, security measures rely mainly on pre-defined rules or manual user intervention, making it difficult to implement appropriate measures that reflect the user's psychological state in real time. While existing systems identify relevant materials from large volumes of academic documents and generate summaries, they do not present search results or countermeasures that take the user's emotional state into account. As a result, it is difficult for users to obtain optimal information and countermeasures in situations where they feel stressed or tense.

[1179] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving a search query entered by a user, means for searching an academic database based on the search query and identifying relevant academic materials, means for using a generative model to generate a summary of the identified academic materials, means for analyzing research trends based on the generated summary and related information, means for displaying the summary and trend analysis results to the user, and means for recognizing the user's emotional state using an emotion engine and proposing and implementing search results and security measures based on that. This makes it possible to provide optimal information and security measures in real time according to the user's psychological state.

[1180] A "search query" is a keyword or phrase that a user enters into a search system.

[1181] An "academic database" is a collection of information that contains academic materials.

[1182] "Academic materials" are documents containing academic information, such as research and academic papers.

[1183] A "generative model" is an algorithm that uses natural language processing techniques to automatically perform specific tasks.

[1184] "Natural language processing technology" is an artificial intelligence technology that has the ability to understand and generate human language.

[1185] An "abstract" is a sentence that succinctly summarizes the main points and conclusions of a document.

[1186] "Related information" refers to the metadata and contextual information that accompanies scholarly documents and abstracts.

[1187] "Trend analysis" is the process of identifying and analyzing patterns and trends based on data.

[1188] The "emotion engine" is a program that recognizes the user's emotional state from their facial expressions, voice, and input content.

[1189] An "emotional state" is the psychological state that a user is feeling at that moment.

[1190] "Search Results" are lists of relevant information returned by the system based on a search query.

[1191] "Security measures" are specific measures and procedures to ensure user safety.

[1192] "Suggestion and execution" means presenting the optimal option to the user and automatically implementing that option if necessary.

[1193] This invention is a system that recognizes a user's emotional state in real time and then proposes and implements relevant information and security measures based on that information. Below, we will explain in detail the program and processing of the system that realizes this application example.

[1194] System Configuration

[1195] This system consists of a user terminal, a server, a generative model, an academic database, and an emotion engine.

[1196] 1. User Device

[1197] It provides an interface for users to enter search queries and obtain emotion information.

[1198] Hardware used: Smartphone, smart glasses

[1199] Software used: React Native and ARKit as front-end technologies

[1200] 2. Server

[1201] It processes query and sentiment information to search, summarize, and analyze trends in academic materials.

[1202] Software used: Node.js, Express.js, MongoDB, PostgreSQL

[1203] 3. Generative Model

[1204] Generate summaries of academic materials using natural language processing techniques.

[1205] Software used: Google Cloud Natural Language API, OpenAI GPT-3

[1206] 4. Academic Databases

[1207] A collection of data containing related academic materials.

[1208] Software used: MongoDB, PostgreSQL

[1209] 5. Emotion Engine

[1210] A program that recognizes and analyzes emotions from the user's facial expressions, voice, and input content.

[1211] Software used: IBM Watson, Microsoft Azure Cognitive Services

[1212] A natural language description of the process

[1213] Acquiring emotion data

[1214] The user device captures the user's facial expressions using a camera on a smartphone or smart glasses, and records their voice using a microphone. This allows for real-time emotional data acquisition, and the emotional data from the facial and vocal signals is analyzed using an emotion engine (IBM Watson or Microsoft Azure Cognitive Services).

[1215] Determining the risk level

[1216] The server evaluates the emotional data obtained from the emotion engine in real time, and if negative emotions such as tension or anxiety are strong, it judges the risk level to be high.

[1217] Proposing and implementing measures

[1218] If the risk level is determined to be high, the server will send a push notification to the user suggesting specific countermeasures. For example, it may suggest security measures such as "locking the front door." If the user allows "auto-execution," the specified security measures will be automatically executed. At this time, the server performs operations through an API that controls smart home devices (Google Home, Amazon Alexa, etc.).

[1219] Specific examples

[1220] When the user returns home late at night, the user device detects a state of tension. The server determines the security risk level to be high and notifies the user as follows: "You are feeling tense. Do you want to lock the front door?" and "Do you want to allow automatic locking?"

[1221] Prompt Sentence Examples

[1222] "If users are nervous, suggest security measures."

[1223] "Please judge the risk level based on the following emotional data. Emotional data: [Facial expression: 'Tension', Audio: 'Anxiety']"

[1224] This makes it possible to provide optimal information and security measures in real time according to the user's psychological state.

[1225] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1226] Step 1:

[1227] The user enters a search query.

[1228] Input: The user enters a "search query" into the device.

[1229] How it works: The device receives a user's search query and prepares to process it.

[1230] Output: Search query data from the user

[1231] Step 2:

[1232] The emotion engine reads the user's emotions in real time.

[1233] Input: Video and audio data captured from the camera and microphone on the user's device.

[1234] How it works: The device sends video and audio data to the emotion engine, which then analyzes it.

[1235] Output: Emotional state data (e.g., tension, anxiety)

[1236] Step 3:

[1237] The server receives the search query and the emotional state.

[1238] Input: User search query data and emotional state data.

[1239] Operation: The device sends this data to the server.

[1240] Output: Search query data and emotional state data are passed to the server.

[1241] Step 4:

[1242] The server performs searches against academic databases.

[1243] Input: Search query data.

[1244] How it works: The server searches academic databases to identify relevant academic materials.

[1245] Output: List of related academic materials

[1246] Step 5:

[1247] A generative model generates summaries of relevant academic material.

[1248] Input: List of relevant academic resources.

[1249] How it works: The server passes a list of materials to the generative model, which then generates a summary using natural language processing techniques.

[1250] Output: Generated summary

[1251] Step 6:

[1252] The server analyzes research trends based on the generated abstracts and associated metadata.

[1253] Input: Generated summary sentences and associated metadata, emotional state data.

[1254] How it works: The server aggregates this data and uses trend analysis algorithms to analyze trends.

[1255] Output: Trend analysis results

[1256] Step 7:

[1257] Determine security risk levels based on emotional states.

[1258] Input: Emotional state data.

[1259] Operation: The server evaluates the emotional state data and determines the risk level.

[1260] Output: Risk level data (e.g. high, medium, low)

[1261] Step 8:

[1262] If the security risk level is determined to be high, countermeasures will be proposed.

[1263] Input: Risk level data.

[1264] What it does: If the server determines a high risk, it will send a notification to the user suggesting specific security measures.

[1265] Output: Specific action suggestions (e.g., "Would you like to lock your front door?")

[1266] Step 9:

[1267] If the user allows automatic execution, the specified security measures are implemented.

[1268] Input: User authorization data.

[1269] How it works: The server receives the user's permission and executes the measures through the control API of the smart home device.

[1270] Output: Security measures taken (e.g., front door automatically locks)

[1271] Through the above steps, optimal information provision and security measures are automatically implemented in real time according to the user's emotions.

[1272] 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.

[1273] 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.

[1274] 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.

[1275] [Fourth embodiment]

[1276] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1277] 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.

[1278] 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).

[1279] 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.

[1280] 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.

[1281] 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).

[1282] 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.

[1283] 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.

[1284] 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.

[1285] 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.

[1286] 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.

[1287] 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.

[1288] 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."

[1289] The present invention relates to an AI system that efficiently finds highly relevant academic papers from a large number of papers and provides their summaries and research trends. Specific embodiments for carrying out the present invention will be described below.

[1290] System configuration

[1291] This system consists of a user terminal, a server, a generative model, and an academic database.

[1292] 1. User terminal: Provides an interface for users to input topics and keywords.

[1293] 2. Server: Processes input queries, searches academic papers, generates summaries, and performs trend analysis.

[1294] 3. Generative model: Generate summaries of academic papers using natural language processing techniques.

[1295] 4. Academic database: A collection of data containing relevant academic papers.

[1296] Program processing

[1297] 1. User enters query (terminal):

[1298] A user inputs a specific topic or keyword, such as "application of generative modeling," through the interface of their device. When the user clicks the search button, the input query is sent to the server.

[1299] 2. The server receives the query:

[1300] The server receives and interprets queries submitted by users, and based on these queries, the server searches academic databases to find relevant academic papers.

[1301] 3. Searching academic databases:

[1302] The server queries academic databases to retrieve a list of relevant academic papers, returning metadata such as the paper's title, author, publication year, abstract, and number of citations.

[1303] 4. Summary generation using generative models:

[1304] The server then passes the retrieved list of papers to a generative model, which then uses natural language processing techniques to generate a summary of each paper, including the paper's main points and conclusions.

[1305] 5. Trend Analysis:

[1306] The server analyzes research trends based on the generated abstracts and associated metadata, taking into account the year of publication, number of citations, and key keywords of the papers. For example, a trend report can be generated showing that "Technology X" is attracting particular attention in recent research.

[1307] 6. Displaying the results:

[1308] The server formats the generated summary and trend analysis results and sends them back to the user's terminal, where the user's terminal displays the received information in a format that the user can visually confirm, such as by displaying a graph that visualizes the trend analysis along with the summary list.

[1309] Specific examples

[1310] A user searches for "applications of generative models."

[1311] 1. Device: The user enters "application of generative models" in the search bar and clicks the search button.

[1312] 2. Server: Receives the query and searches academic databases to retrieve a list of relevant papers on "applications of technology X."

[1313] 3. Summary generation: The server uses the generative model to generate a summary of the paper, creating a summary statement such as, "This paper shows that technology X outperforms conventional technology."

[1314] 4. Trend analysis: The server generates a trend report based on the publication year, number of citations, and major keywords, stating that "the application of generative models has developed rapidly in recent years."

[1315] 5. Results display: A summary and trend report will be displayed on the user's terminal, allowing the user to conduct further research and discussion based on the information.

[1316] As described above, the system of the present invention efficiently searches for related academic papers and provides their summaries and trend information, thereby enabling researchers and students to quickly obtain the information they need.

[1317] The processing flow will be explained below.

[1318] Step 1:

[1319] The user inputs a search query using the interface of their device. For example, the user inputs "application of generative model" as the search query and clicks the search button.

[1320] Step 2:

[1321] The terminal sends the query entered by the user to the server. Specifically, the terminal sends the query as an HTTP request to the server.

[1322] Step 3:

[1323] The server analyzes the received query and executes a search query against the appropriate academic database, which retrieves a list of relevant papers.

[1324] Step 4:

[1325] The server analyzes the list of academic papers it has obtained and extracts the metadata for each paper (title, author, year of publication, abstract, number of citations, etc.).

[1326] Step 5:

[1327] The server passes the extracted metadata and full text of the papers to a generative model, which uses natural language processing techniques to generate a summary of each academic paper.

[1328] Step 6:

[1329] The generative model returns a summary of each paper to the server, including its main points and conclusions. For example, it generates a summary such as, "The effectiveness of new generative model technology X has been demonstrated."

[1330] Step 7:

[1331] The server performs trend analysis based on the generated abstracts and metadata, taking into account the year of publication, number of citations, and main keywords of the papers to identify research trends.

[1332] Step 8:

[1333] Based on the results of the trend analysis, the server generates a trend report, such as "The application of generative models has developed rapidly in recent years, with technology X attracting particular attention."

[1334] Step 9:

[1335] The server sends the generated summary and trend report to the user's terminal, formatted in a user-friendly format.

[1336] Step 10:

[1337] The terminal displays the received summary and trend report to the user. The user interface displays a summary list and visualized trend graphs. The user can then conduct further research based on this information.

[1338] Through the above steps, users can efficiently find relevant academic papers, quickly understand summaries and trends, and effectively utilize information.

[1339] Example 1

[1340] 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."

[1341] Conventional academic paper search systems make it difficult to efficiently find relevant papers, and it is also difficult to quickly understand the content of each paper. Furthermore, they lack the functionality to analyze and provide research trends, making it difficult for researchers and students to quickly obtain the information they need. For this reason, there was a need to develop a system that could efficiently extract necessary knowledge from vast amounts of information and provide it in an easy-to-understand manner.

[1342] 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.

[1343] In this invention, the server includes a means for a user to input topics and keywords, a means for sending a query to the server, a means for receiving and analyzing the query, a means for searching academic databases and retrieving related academic papers, a means for generating summaries of the papers using a generative model, a means for analyzing research trends based on the generated summaries and related metadata, and a means for formatting the summaries and trend analysis results and transmitting them to a user terminal. This makes it possible to efficiently search for highly relevant academic papers and quickly provide their summaries and trend information.

[1344] - "Query" refers to the string of characters or keywords that a user enters to search for specific information.

[1345] "Server" refers to a computer system that receives queries from users, processes them, and provides appropriate search results, summaries, and trend analysis results.

[1346] "User Terminal" means the computer or mobile device used by a User to access the System, enter queries, and view results.

[1347] An "academic database" refers to a collection of data that contains and is used to search academic papers and related information.

[1348] A "generative model" refers to an algorithm or software that uses natural language processing techniques to automatically generate summaries of academic papers.

[1349] An "abstract" is a piece of text that concisely summarizes the main points and conclusions of an academic paper.

[1350] "Trend analysis" refers to the process of identifying recent trends and hot topics in a research field based on the metadata of retrieved academic papers.

[1351] "Metadata" refers to information about the data itself, such as the paper's title, author, publication year, abstract, and number of citations.

[1352] "HTTP request" refers to the protocol used by a user terminal to request data or services from a server.

[1353] "Natural language processing technology" refers to the techniques and methods that allow computers to understand, generate, and analyze human language.

[1354] This system efficiently finds related academic papers by allowing users to input topics and keywords, and provides summaries and research trends. Specifically, it consists of a user terminal, a server, a generative model, and an academic database.

[1355] User terminal

[1356] The user device provides an interface for users to input queries. This is implemented as a web or mobile application and is designed to allow users to easily input topics and keywords. For example, a user might input the query "application of generative AI model."

[1357] server

[1358] The server receives queries sent from user devices and analyzes their contents. Based on these queries, the server searches academic databases to find relevant academic papers. Academic databases often use SQL Server or Elasticsearch.

[1359] The obtained list of papers is passed to a generative model, which generates a summary of each paper. The generative model uses natural language processing techniques such as the Transformer or BERT model. The generated summaries are then stored on the server.

[1360] The server then analyzes the abstracts and associated metadata to identify research trends, taking into account factors such as the year of publication, number of citations, and key keywords of the papers. For example, it can generate a trend report that states, "The application of generative AI models has rapidly increased in recent years."

[1361] Trend analysis

[1362] The server formats the generated summary and trend analysis results and sends them to the user's terminal, where the user can visually check the results, such as a summary list or a graph visualizing the trend analysis.

[1363] Specific examples

[1364] The following is the series of processes that a user goes through when searching for "applications of generative models."

[1365] 1. User: The user enters "application of generative models" in the search bar and clicks the search button.

[1366] 2. Server: Receives the query and searches academic databases to obtain a list of relevant papers on "applications of generative models."

[1367] 3. Generative model: The server uses the generative model to generate a summary of the paper, creating a summary statement such as, "This paper demonstrates that generative models outperform conventional techniques."

[1368] 4. Trend analysis: The server generates a trend report based on the publication year, number of citations, and major keywords, stating that "the application of generative models has developed rapidly in recent years."

[1369] 5. Display of results: A summary and trend report is displayed on the user's terminal, allowing the user to conduct further research or discussion based on the information.

[1370] Example prompts to input to a generative AI model:

[1371] "Please search for the latest academic papers on applications of generative models and provide summaries and research trends."

[1372] This system makes it possible to efficiently search for relevant academic papers and quickly provide their summaries and trend information, enabling researchers and students to quickly and accurately obtain the information they need.

[1373] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1374] Step 1:

[1375] Query input (user)

[1376] A user enters a specific topic or keyword into the search field of a web application that serves as the interface for their device, for example, "applications of generative AI models," and clicks the search button. This action stacks the query locally on the device, ready to be sent.

[1377] Input: User-entered topics or keywords (e.g., "applications of generative AI models")

[1378] Output: Query data sent by the device to the server

[1379] Step 2:

[1380] Query sending (terminal)

[1381] The device generates an HTTP request containing the query entered by the user and sends it to the server's API endpoint, along with the user's session information and a timestamp.

[1382] Input: User-entered query, session information, timestamp

[1383] Output: HTTP request sent to the server

[1384] Step 3:

[1385] Query reception and interpretation (server)

[1386] The server analyzes the received HTTP request and extracts the query portion. It obtains the query "application of generative AI model" from the received request and normalizes the query using natural language processing. This analysis is performed using a text analysis library or similar.

[1387] Input: Query data in the HTTP request

[1388] Output: Normalized query

[1389] Step 4:

[1390] Academic database search (server)

[1391] The server searches academic databases using normalized queries, executing SQL or Elasticsearch queries to filter and retrieve relevant academic papers.

[1392] Input: Normalized query

[1393] Output: A search result list of relevant academic papers

[1394] Step 5:

[1395] Obtaining a list of papers (server)

[1396] The server creates a list of relevant papers based on the search results obtained from academic databases. Specifically, it collects metadata such as the paper's title, author, publication year, abstract, and number of citations, and compiles it into a list.

[1397] Input: Search results from academic databases

[1398] Output: List of article metadata

[1399] Step 6:

[1400] Paper summary generation (server / generative model)

[1401] The server generates summaries of each paper using generative models, such as the Transformer and BERT models, which use natural language processing techniques, to generate summaries that include the main points and conclusions of each paper.

[1402] Input: Article metadata

[1403] Output: Generated paper abstract

[1404] Step 7:

[1405] Trend analysis (server)

[1406] The server analyzes research trends based on the generated abstracts and related metadata, taking into account the year of publication, number of citations, and key keywords of the paper, and generates a trend report such as, "The application of generative AI models has rapidly increased in recent years."

[1407] Input: Generated paper abstracts and metadata

[1408] Output: Trend report

[1409] Step 8:

[1410] Result formatting and sending (server)

[1411] The server formats the generated summary and trend analysis results and sends them to the user's device. The formatted information is encoded in, for example, JSON format and included in the HTTP response.

[1412] Input: Summary and trend analysis results

[1413] Output: HTTP response sent to the user device

[1414] Step 9:

[1415] Result display (terminal)

[1416] The terminal analyzes the results received from the server and displays them visually to the user, who can view summary lists and graphs visualizing trend analysis.

[1417] Input: HTTP response from the server

[1418] Output: On-screen summary list and trend analysis results

[1419] (Application example 1)

[1420] 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."

[1421] Currently, it takes a great deal of time and effort for researchers to efficiently find relevant papers from a large number of academic papers and analyze their summaries and research trends. Furthermore, there are only a limited number of systems that automatically generate paper summaries and display them visually, and convenience is particularly required on mobile devices such as smartphones. Furthermore, there is a lack of trend analysis to track research progress. To solve these problems, a system is needed that can quickly identify relevant academic papers and efficiently provide their summaries and trends.

[1422] 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.

[1423] In this invention, the server includes means for receiving a search query entered by a user, means for searching an academic database based on the search query to identify relevant academic papers, means for using a generative model to generate summaries of the identified academic papers, means for analyzing research trends based on the generated summaries and related data, means for visually displaying the summaries and trend analysis results to the user, means for passing the searched paper list to the generative model to generate summaries of each paper, and means for analyzing the number of citations and major keywords of each paper to detect research trends. This makes it possible to quickly identify relevant academic papers and efficiently provide their summaries and trends.

[1424] "Means for receiving a user-entered search query" refers to a device or software that provides an interface for a user to enter a topic or keyword they wish to search for and that is capable of receiving such input.

[1425] "Means for searching academic databases based on the search query and identifying relevant academic papers" refers to a device or software that searches academic databases based on a search query entered by a user and finds relevant academic papers.

[1426] "Means for using a generative model to generate summaries of the identified academic literature" refers to a device or software that uses AI technology known as a generative model to create summaries in order to automatically summarize the contents of the identified academic literature.

[1427] "Means for analyzing research trends based on the generated abstracts and associated data" refers to devices or software that use the generated abstracts and associated metadata to analyze movements and trends in a particular research field.

[1428] "Means for visually displaying the summary and trend analysis results to the user" refers to a device or software that displays the generated summary and trend analysis results on a screen in a format that is easy for the user to understand.

[1429] "Means for passing the searched paper list to a generative model and generating a summary of each paper" refers to a device or software that inputs the list of multiple papers obtained as search results into a generative model and generates a summary of each paper.

[1430] "Means for detecting research trends by analyzing the number of citations and major keywords of each document" refers to a device or software that analyzes the number of citations and major keywords of each paper and detects research trends based on that information.

[1431] The present invention relates to an AI system that efficiently finds highly relevant academic papers from a large number of papers and provides summaries and research trends. Specific embodiments for carrying out the present invention are described below.

[1432] System configuration

[1433] This system consists of a user terminal, a server, a generative model, and an academic database.

[1434] 1. User device: Provides an interface for users to input topics and keywords. Users can enter search queries through smartphone applications and visually check the results.

[1435] 2. Server: Processes input queries, searches academic papers, generates summaries, and performs trend analysis. The server is implemented using Node.js and Express.

[1436] 3. Generative model: Generate summaries of academic papers using natural language processing techniques. OpenAI's GPT-4 is used to generate summary sentences.

[1437] 4. Academic database: A collection of data containing related academic papers. Academic data can be obtained using APIs such as PubMed and arXiv.

[1438] System Operation

[1439] A user enters a specific keyword into the search bar of a smartphone app, for example, "application of generative models," and clicks the search button. This query is sent to the server.

[1440] The server calls the API of an academic database based on the received query to obtain a list of related papers, which it then passes to a generative model (GPT-4) to generate summaries of each paper.

[1441] The generated summaries are returned to the server, where a trend analysis is performed based on the metadata of each paper (number of citations, year of publication, main keywords, etc.). The progress of research and popular technologies are analyzed, and the results are generated as a trend report.

[1442] Finally, the server sends the summary and trend report to the user's terminal, where the user can visually check this information. For example, a summary list is displayed, and below it, visualized information such as trend graphs and keyword frequency rates can be viewed.

[1443] Example sentences

[1444] Example prompt sentence:

[1445] Summarize the following academic paper:

[1446] Title: Generative Models and Their Applications

[1447] Abstract: This paper explores the various applications of generative models, including their usage in image generation, natural language processing, and other fields...

[1448] This system allows researchers to quickly identify relevant academic papers and efficiently obtain their summaries and trends. For example, if a user searches for "applications of generative models," the system will display the latest related research papers along with their summaries. Furthermore, based on trend analysis, recent research trends can be visually confirmed, allowing researchers to efficiently grasp the progress of research.

[1449] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1450] Step 1:

[1451] The user enters a query

[1452] A user enters a topic or keyword into the search bar of a smartphone app, for example, "application of generative models." When the user clicks the search button, the query is sent from the device to the server.

[1453] Input: Search keywords entered by the user (e.g., "application of generative models")

[1454] Output: The search query sent to the server

[1455] Step 2:

[1456] The server receives the query and searches the academic database.

[1457] The server receives the search query sent from the device and calls the API of an academic database based on the query to search for related academic papers. The server uses the API of PubMed or arXiv, for example.

[1458] Input: The received search query

[1459] Data processing: Calling APIs of academic databases based on queries

[1460] Output: A list of academic papers obtained as search results

[1461] Step 3:

[1462] Pass the obtained list of academic papers to the generative model

[1463] The server passes each paper in the list of academic papers to a generative model (GPT-4) to generate a summary. The generative model analyzes the content of each paper and generates a summary containing the main points.

[1464] Input: List of academic papers

[1465] Data Computation: Generating summaries using generative models

[1466] Output: Abstract of each paper

[1467] Step 4:

[1468] Conduct trend analysis based on the generated summaries and metadata

[1469] The server analyzes research trends based on the generated abstracts and each paper's metadata (year of publication, number of citations, major keywords, etc.), for example, by analyzing the frequency of keyword appearances and the year of publication of the paper.

[1470] Input: Abstract and article metadata

[1471] Data Calculation: Trend Analysis through Metadata Analysis

[1472] Output: Trend report

[1473] Step 5:

[1474] Sends summary and trend reports to user terminals for display

[1475] The server formats the generated summary and trend report and sends them to the user's terminal, which receives and visually displays them, for example, visualizing the summary list along with trend graphs and keyword frequency rates.

[1476] Input: Abstract and trend report

[1477] Data processing: formatting summaries and trend reports

[1478] Output: Information displayed on the user's terminal (summary list, trend graph, etc.)

[1479] 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.

[1480] The present invention combines an emotion engine with a system that efficiently finds highly relevant academic papers from a large number of papers and provides summaries and research trends. This makes it possible to recognize the user's emotions and provide optimal information based on those emotions. Specific embodiments of the present invention are described below.

[1481] System configuration

[1482] This system consists of a user terminal, a server, a generative model, an academic database, and an emotion engine.

[1483] 1. User terminal: Provides an interface where users can input topics and keywords and obtain emotional information.

[1484] 2. Server: Processes queries and sentiment information, searches academic papers, generates summaries, and performs trend analysis.

[1485] 3. Generative model: Generate summaries of academic papers using natural language processing techniques.

[1486] 4. Academic database: A collection of data containing relevant academic papers.

[1487] 5. Emotion engine: A program that recognizes and analyzes emotions from user input, facial expressions, voice, etc.

[1488] Program processing

[1489] 1. User enters a query and sentiment is read (on device):

[1490] The user enters a search query such as "application of generative model" on the device. The device is also equipped with an emotion engine that reads the user's emotions in real time from their facial expressions, voice, and input content. The read emotion information is then sent to the server along with the query.

[1491] 2. The server receives the query and emotion information:

[1492] The server receives the search query and sentiment information sent by the user, and then executes the search query against the appropriate academic database based on the received query and sentiment information.

[1493] 3. Searching academic databases:

[1494] The server queries academic databases to retrieve a list of relevant academic papers, returning metadata such as the paper's title, author, publication year, abstract, and number of citations.

[1495] 4. Summary generation using generative models:

[1496] The server then passes the retrieved list of papers to a generative model, which uses natural language processing techniques to generate a summary of each paper, including the main points and conclusions of the paper.

[1497] 5. Trend analysis taking into account sentiment information:

[1498] The server analyzes research trends by taking into account emotional information along with the generated summaries and associated metadata. For example, the content of academic papers is filtered to match the user's interests and state. If the emotional engine recognizes the user's state of excitement, it will prioritize providing trend information on the latest technological innovations.

[1499] 6. Displaying the results:

[1500] The server then sends the generated summary and a trend report adjusted based on the sentiment information to the user's device. The device then displays the received information in a format that is easy for the user to understand, such as a graph visualizing the trend analysis along with the summary list.

[1501] Specific examples

[1502] The user searches for "applications of generative models." The user's emotions are also recognized.

[1503] 1. Device: The user enters "Application of generative model" and clicks the search button. At the same time, the emotion engine detects the user's excitement state.

[1504] 2. Server: Receives the query along with the emotion information and searches the academic database to obtain a list of relevant papers on "applications of technology X."

[1505] 3. Summary generation: The server uses the generative model to generate a summary of the paper, such as "This paper demonstrates that technology X outperforms conventional technologies."

[1506] 4. Sentiment Analysis: Taking into account the excitement level of users, trend analysis prioritizes trend information on the latest technological innovations.

[1507] 5. Result display: A summary and trend report is displayed on the user's terminal, highlighting the latest technological information taking into account the user's excitement.

[1508] Through these steps, users can efficiently find relevant academic papers, quickly understand summaries and trends, and effectively utilize information according to their emotional state.

[1509] The processing flow will be explained below.

[1510] Step 1:

[1511] The user enters a search query using the device's interface. For example, they enter "application of generative models" and click the search button. The device's built-in emotion engine then reads the user's emotions in real time from their facial expressions and voice.

[1512] Step 2:

[1513] The device sends the search query and emotion information entered by the user to the server. Specifically, the device sends the query and emotion information together as an HTTP request to the server.

[1514] Step 3:

[1515] The server analyzes the received query and sentiment information. Based on the query, the server searches academic databases to retrieve a list of relevant academic papers. The search results include metadata such as the paper title, author, publication year, abstract, and number of citations.

[1516] Step 4:

[1517] The server analyzes the list of academic papers it has acquired and passes each paper's metadata and full text to a generative model. The generative model then uses natural language processing technology to generate a summary of each paper. For example, it generates a summary such as, "This paper demonstrates that technology X outperforms conventional technology."

[1518] Step 5:

[1519] The server performs trend analysis based on the generated summary, metadata, and emotion information. For example, if the emotion engine recognizes the user's excitement, it will prioritize analyzing trend information related to the latest technological innovations.

[1520] Step 6:

[1521] As a result of the trend analysis, the server generates a trend report, such as "The application of generative models has developed rapidly in recent years, with technology X attracting particular attention."

[1522] Step 7:

[1523] The server sends the generated summary and trend report to the user's device, formats it in a visually optimized format that takes into account emotional information, and then transmits the data.

[1524] Step 8:

[1525] The terminal displays the received summary and trend report to the user. The user interface displays summary lists and graphs that visualize trend analysis, highlighting important information that corresponds to the user's excitement level. The user can then use this information for further detailed research and discussion.

[1526] As described above, the processing flow of the system combined with the emotion engine enables users to more efficiently find relevant academic papers, quickly understand summaries and trend information, and obtain the most appropriate information according to their emotional state.

[1527] Example 2

[1528] 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."

[1529] Conventional academic resource search systems search academic databases based on user-entered search queries to identify relevant materials, but they have the problem of being unable to provide information that takes into account the user's emotions and interests. This makes it impossible to provide appropriate information or analyze trends according to the user's emotional state, making it difficult to improve user satisfaction and convenience.

[1530] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1531] In this invention, the server includes means for receiving a search query entered by a user, means for searching an academic database based on the search query to identify relevant academic materials, means for using a generative model to generate summaries of the identified academic materials, means for analyzing research trends based on the summaries and associated metadata, means for recognizing user sentiment and adjusting the search query and trend analysis based on that information, and means for displaying the summaries and trend analysis results to the user, thereby providing optimal information according to the user's sentiment, improving the accuracy of search results and trend analysis and user satisfaction.

[1532] A "user" is an entity that uses this system to search for academic materials and obtain information.

[1533] A "search query" is a theme or keyword for academic materials to be searched that a user enters into a terminal.

[1534] An "academic database" is a collection of data that collects and manages academic materials, such as academic papers and research results.

[1535] "Academic materials" refers to documents such as papers, research reports, and presentations stored in academic databases.

[1536] A "generative model" is an algorithm or system that uses artificial intelligence and natural language processing techniques to analyze text data and generate a summary.

[1537] An "abstract" is a document that succinctly summarizes the main points and conclusions of academic material.

[1538] "Metadata" refers to data that indicates information related to academic materials, such as title, author, year of publication, and number of citations.

[1539] A "trend" refers to recent developments or popular trends in a particular academic field or topic.

[1540] "Emotion" refers to the user's psychological state and emotional response, and in this system, it is recognized from facial expressions, voice, input content, etc.

[1541] "Emotion information" is data about the user's emotional state obtained by the emotion engine.

[1542] An "emotion engine" is a program and system that recognizes and analyzes emotions from a user's facial expressions, voice, and input content.

[1543] A "terminal" is a device through which a user enters a search query and receives results, and includes, for example, a personal computer or smartphone.

[1544] The "server" is a central processing unit that processes search queries and sentiment information from users, searches academic databases, generates summaries, and performs trend analysis.

[1545] The present invention combines a system that efficiently finds highly relevant information from a large amount of academic materials and provides summaries and research trends with an emotion engine that recognizes the user's emotions. This makes it possible to provide optimal information tailored to the user's emotions. Specific embodiments of the present invention are described below.

[1546] System configuration

[1547] This system consists of a user terminal, a server, a generative model, an academic database, and an emotion engine.

[1548] 1. User Device:

[1549] It provides an interface for users to input search queries and obtain emotion information. For example, it is a PC or smartphone. The device is equipped with a camera and microphone, and has the ability to analyze the user's facial expressions and voice using an emotion engine.

[1550] 2. Server:

[1551] It is a central device that processes search queries and emotional information sent by users. Specifically, it receives queries and emotional information, searches academic databases, generates abstracts of papers using generative models, and performs trend analysis taking into account the emotional information.

[1552] 3. Generative Model:

[1553] These are algorithms and systems that use natural language processing techniques to generate summaries of academic materials, such as generative AI models using the Transformer architecture.

[1554] 4. Academic databases:

[1555] A collection of data that contains related scholarly materials, such as electronic journals and digital libraries.

[1556] 5. Emotion Engine:

[1557] It is a program and system that recognizes and analyzes emotions from user input, facial expressions, voice, etc. The emotion engine uses a regular camera and microphone to acquire and analyze emotional information in real time.

[1558] Program processing

[1559] When a user enters a search query, the device receives the query and simultaneously acquires the user's emotional information through the emotion engine. The acquired query and emotional information are then sent to the server, which then executes the search query against the academic database.

[1560] The list of academic materials obtained as search results is summarized by a generative model on the server, and trend analysis is then performed based on the generated summaries, metadata, and user sentiment information to uniquely provide the most relevant information.

[1561] Specific examples

[1562] When a user searches for "application of generative modeling," the user enters "application of generative modeling" into the device, and the emotion engine detects the user's excitement. The device then sends the search query and emotion information to the server.

[1563] The server searches academic databases based on the search query "application of generative models" to retrieve relevant materials. The retrieved materials are summarized using a generative model. For example, a summary such as "This research proposes a new application method using generative models" is generated.

[1564] The server then performs trend analysis to prioritize the provision of trend information on the latest technological innovations, taking into account the user's excitement state detected by the emotion engine. The results of the trend analysis and a summary are finally sent to the terminal.

[1565] The terminal receives this information and displays it to the user. The user interface displays summary lists and visualizes the latest technology trends in graphs and charts.

[1566] Prompt Sentence Examples

[1567] "Design a system to search the latest academic literature on the application of generative models and provide optimal summaries and trend information to excited users."

[1568] In this way, the system provides users with the most appropriate information based on their emotional state, allowing them to efficiently search and understand academic materials. It also helps users understand the latest research trends through trend analysis.

[1569] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1570] Step 1:

[1571] User enters search query and acquires emotion information (device)

[1572] A user types "application of generative models" into the search bar of their device and presses the Enter key. Pressing the Enter key triggers the device's built-in emotion engine, which uses the camera and microphone to analyze the user's facial expressions and voice in real time. The input data is the search query "application of generative models" and emotion information (e.g., excited state) acquired by the emotion engine. The emotion engine then outputs the results of the user's facial expression analysis and generates a data packet that is sent to the server along with the search query.

[1573] Step 2:

[1574] Send query and emotion information to the server (device)

[1575] The device sends the search query "Application of generative model" and the acquired emotion information together in a single data packet to the server. The transmission is triggered when the user presses the Enter key. The data packet contains the search keyword and the user's emotion (e.g., excitement) and is transferred to the server using a transmission protocol. The server is then ready to analyze the received data.

[1576] Step 3:

[1577] Receive query and emotion information (server)

[1578] The server receives the data packet sent from the terminal. The received data packet is decoded by the server's receiving module, and the search query "Application of generative model" and emotional information (e.g., excitement state) are separated. The server then passes the received search query to the analysis module, and the emotional information is passed to the emotion analysis module.

[1579] Step 4:

[1580] Search academic databases (server)

[1581] The server's query engine queries the received search query "application of generative models" to the academic database. The input data is the search query, and the database query execution engine executes the query. As a result, relevant paper data (metadata such as title, author, publication year, abstract, and number of citations) from the academic database is output as search results.

[1582] Step 5:

[1583] Generate a summary based on the search results (server)

[1584] The server passes search results obtained from the academic database to the generative model, which uses natural language processing techniques to generate summaries of the papers. The input data is a list of related papers and their metadata, which the generative model analyzes to create a concise summary. The output is a summary such as, "This research proposes a new application method using the generative model."

[1585] Step 6:

[1586] Conduct trend analysis taking into account emotional information (server)

[1587] The server performs trend analysis using the generated paper summaries and related metadata, as well as user sentiment information. The input data are the abstracts, metadata, and sentiment information. The sentiment analysis engine takes into account the user's state of excitement and prioritizes the extraction of trend information related to the latest technological innovations. The output is a trend report optimized for user sentiment.

[1588] Step 7:

[1589] Send the final result to the user terminal (server)

[1590] The server then combines the generated abstracts and the adjusted trend report into a single data packet and sends it to the user's device. The transmission protocol is HTTP or WebSocket. The data packet contains a list of abstracts and a trend report, allowing the user to receive the latest research trends and abstracts.

[1591] Step 8:

[1592] Display the results to the user (terminal)

[1593] The terminal receives the data packets sent from the server, analyzes the received summary statements and trend reports, and displays them. The user interface displays the summary list in an easy-to-read format and visualizes the latest technology trend information in graphs and charts. This is a step that allows users to efficiently obtain information and deepen their understanding.

[1594] (Application example 2)

[1595] 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."

[1596] Currently, security measures rely mainly on pre-defined rules or manual user intervention, making it difficult to implement appropriate measures that reflect the user's psychological state in real time. While existing systems identify relevant materials from large volumes of academic documents and generate summaries, they do not present search results or countermeasures that take the user's emotional state into account. As a result, it is difficult for users to obtain optimal information and countermeasures in situations where they feel stressed or tense.

[1597] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving a search query entered by a user, means for searching an academic database based on the search query and identifying relevant academic materials, means for using a generative model to generate a summary of the identified academic materials, means for analyzing research trends based on the generated summary and related information, means for displaying the summary and trend analysis results to the user, and means for recognizing the user's emotional state using an emotion engine and proposing and implementing search results and security measures based on that. This makes it possible to provide optimal information and security measures in real time according to the user's psychological state.

[1598] A "search query" is a keyword or phrase that a user enters into a search system.

[1599] An "academic database" is a collection of information that contains academic materials.

[1600] "Academic materials" are documents containing academic information, such as research and academic papers.

[1601] A "generative model" is an algorithm that uses natural language processing techniques to automatically perform specific tasks.

[1602] "Natural language processing technology" is an artificial intelligence technology that has the ability to understand and generate human language.

[1603] An "abstract" is a sentence that succinctly summarizes the main points and conclusions of a document.

[1604] "Related information" refers to the metadata and contextual information that accompanies scholarly documents and abstracts.

[1605] "Trend analysis" is the process of identifying and analyzing patterns and trends based on data.

[1606] The "emotion engine" is a program that recognizes the user's emotional state from their facial expressions, voice, and input content.

[1607] An "emotional state" is the psychological state that a user is feeling at that moment.

[1608] "Search Results" are lists of relevant information returned by the system based on a search query.

[1609] "Security measures" are specific measures and procedures to ensure user safety.

[1610] "Suggestion and execution" means presenting the optimal option to the user and automatically implementing that option if necessary.

[1611] This invention is a system that recognizes a user's emotional state in real time and then proposes and implements relevant information and security measures based on that information. Below, we will explain in detail the program and processing of the system that realizes this application example.

[1612] System Configuration

[1613] This system consists of a user terminal, a server, a generative model, an academic database, and an emotion engine.

[1614] 1. User Device

[1615] It provides an interface for users to enter search queries and obtain emotion information.

[1616] Hardware used: Smartphone, smart glasses

[1617] Software used: React Native and ARKit as front-end technologies

[1618] 2. Server

[1619] It processes query and sentiment information to search, summarize, and analyze trends in academic materials.

[1620] Software used: Node.js, Express.js, MongoDB, PostgreSQL

[1621] 3. Generative Model

[1622] Generate summaries of academic materials using natural language processing techniques.

[1623] Software used: Google Cloud Natural Language API, OpenAI GPT-3

[1624] 4. Academic Databases

[1625] A collection of data containing related academic materials.

[1626] Software used: MongoDB, PostgreSQL

[1627] 5. Emotion Engine

[1628] A program that recognizes and analyzes emotions from the user's facial expressions, voice, and input content.

[1629] Software used: IBM Watson, Microsoft Azure Cognitive Services

[1630] A natural language description of the process

[1631] Acquiring emotion data

[1632] The user device captures the user's facial expressions using a camera on a smartphone or smart glasses, and records their voice using a microphone. This allows for real-time emotional data acquisition, and the emotional data from the facial and vocal signals is analyzed using an emotion engine (IBM Watson or Microsoft Azure Cognitive Services).

[1633] Determining the risk level

[1634] The server evaluates the emotional data obtained from the emotion engine in real time, and if negative emotions such as tension or anxiety are strong, it judges the risk level to be high.

[1635] Proposing and implementing measures

[1636] If the risk level is determined to be high, the server will send a push notification to the user suggesting specific countermeasures. For example, it may suggest security measures such as "locking the front door." If the user allows "auto-execution," the specified security measures will be automatically executed. At this time, the server performs operations through an API that controls smart home devices (Google Home, Amazon Alexa, etc.).

[1637] Specific examples

[1638] When the user returns home late at night, the user device detects a state of tension. The server determines the security risk level to be high and notifies the user as follows: "You are feeling tense. Do you want to lock the front door?" and "Do you want to allow automatic locking?"

[1639] Prompt Sentence Examples

[1640] "If users are nervous, suggest security measures."

[1641] "Please judge the risk level based on the following emotional data. Emotional data: [Facial expression: 'Tension', Audio: 'Anxiety']"

[1642] This makes it possible to provide optimal information and security measures in real time according to the user's psychological state.

[1643] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1644] Step 1:

[1645] The user enters a search query.

[1646] Input: The user enters a "search query" into the device.

[1647] How it works: The device receives a user's search query and prepares to process it.

[1648] Output: Search query data from the user

[1649] Step 2:

[1650] The emotion engine reads the user's emotions in real time.

[1651] Input: Video and audio data captured from the camera and microphone on the user's device.

[1652] How it works: The device sends video and audio data to the emotion engine, which then analyzes it.

[1653] Output: Emotional state data (e.g., tension, anxiety)

[1654] Step 3:

[1655] The server receives the search query and the emotional state.

[1656] Input: User search query data and emotional state data.

[1657] Operation: The device sends this data to the server.

[1658] Output: Search query data and emotional state data are passed to the server.

[1659] Step 4:

[1660] The server performs searches against academic databases.

[1661] Input: Search query data.

[1662] How it works: The server searches academic databases to identify relevant academic materials.

[1663] Output: List of related academic materials

[1664] Step 5:

[1665] A generative model generates summaries of relevant academic material.

[1666] Input: List of relevant academic resources.

[1667] How it works: The server passes a list of materials to the generative model, which then generates a summary using natural language processing techniques.

[1668] Output: Generated summary

[1669] Step 6:

[1670] The server analyzes research trends based on the generated abstracts and associated metadata.

[1671] Input: Generated summary sentences and associated metadata, emotional state data.

[1672] How it works: The server aggregates this data and uses trend analysis algorithms to analyze trends.

[1673] Output: Trend analysis results

[1674] Step 7:

[1675] Determine security risk levels based on emotional states.

[1676] Input: Emotional state data.

[1677] Operation: The server evaluates the emotional state data and determines the risk level.

[1678] Output: Risk level data (e.g. high, medium, low)

[1679] Step 8:

[1680] If the security risk level is determined to be high, countermeasures will be proposed.

[1681] Input: Risk level data.

[1682] What it does: If the server determines a high risk, it will send a notification to the user suggesting specific security measures.

[1683] Output: Specific action suggestions (e.g., "Would you like to lock your front door?")

[1684] Step 9:

[1685] If the user allows automatic execution, the specified security measures are implemented.

[1686] Input: User authorization data.

[1687] How it works: The server receives the user's permission and executes the measures through the control API of the smart home device.

[1688] Output: Security measures taken (e.g., front door automatically locks)

[1689] Through the above steps, optimal information provision and security measures are automatically implemented in real time according to the user's emotions.

[1690] 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.

[1691] 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.

[1692] 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.

[1693] 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.

[1694] 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.

[1695] 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.

[1696] 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).

[1697] 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.

[1698] 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."

[1699] 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.

[1700] 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).

[1701] 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.

[1702] 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.

[1703] 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.

[1704] 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.

[1705] 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.

[1706] 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.

[1707] 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.

[1708] 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.

[1709] 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.

[1710] 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.

[1711] The following is further disclosed regarding the above embodiment.

[1712] (Claim 1)

[1713] means for receiving a search query entered by a user;

[1714] means for searching academic databases based on the search query to identify relevant academic papers;

[1715] means for using the generative model to generate summaries of the identified academic papers;

[1716] means for analyzing research trends based on the generated summaries and associated metadata;

[1717] means for displaying the summary and trend analysis results to a user;

[1718] A system including:

[1719] (Claim 2)

[1720] 2. The system according to claim 1, wherein the generative model generates summaries of academic papers using natural language processing techniques.

[1721] (Claim 3)

[1722] 2. The system of claim 1, wherein the means for analyzing research trends identifies trends based on the year of publication and the number of citations of papers.

[1723] "Example 1"

[1724] (Claim 1)

[1725] a means for users to input topics or keywords;

[1726] means for transmitting the input query to a server;

[1727] means for receiving and analyzing said query;

[1728] A means of searching academic databases and retrieving relevant academic papers;

[1729] means for generating summaries of the acquired academic papers using a generative model;

[1730] means for analyzing research trends based on summaries and associated metadata generated by the generative model;

[1731] means for formatting and transmitting the summary and trend analysis results to a user terminal;

[1732] means for displaying the transmitted summary and trend analysis results on a terminal;

[1733] A system including:

[1734] (Claim 2)

[1735] 2. The system according to claim 1, wherein the generative model generates summaries of academic papers using natural language processing techniques.

[1736] (Claim 3)

[1737] 2. The system of claim 1, wherein the means for analyzing research trends identifies trends based on the year of publication and the number of citations of papers.

[1738] "Application Example 1"

[1739] (Claim 1)

[1740] means for receiving a search query entered by a user;

[1741] means for searching academic databases based on the search query to identify relevant academic papers;

[1742] means for using a generative model to generate summaries of the identified academic literature;

[1743] means for analyzing research trends based on the generated summaries and associated data;

[1744] means for visually displaying the summary and trend analysis results to a user;

[1745] A means for passing the retrieved paper list to a generative model to generate a summary of each paper;

[1746] A means of detecting research trends by analyzing the number of citations and main keywords of each document;

[1747] A system including:

[1748] (Claim 2)

[1749] 2. The system of claim 1, wherein the generative model generates summaries of academic literature using natural language processing techniques.

[1750] (Claim 3)

[1751] 2. The system of claim 1, wherein the means for analyzing research trends identifies trends based on the publication year and number of citations of a document.

[1752] "Example 2: Combining Emotion Engines"

[1753] (Claim 1)

[1754] means for receiving a search query entered by a user;

[1755] means for searching academic databases based on the search query to identify relevant academic materials;

[1756] means for using the generative model to generate summaries of the identified academic material;

[1757] means for analyzing research trends based on said summaries and associated metadata;

[1758] a means for recognizing user sentiment and adjusting search queries and trend analysis based on that sentiment;

[1759] means for displaying the summary and trend analysis results to a user;

[1760] A system including:

[1761] (Claim 2)

[1762] 2. The system of claim 1, wherein the generative model generates summaries of academic materials using natural language processing techniques.

[1763] (Claim 3)

[1764] The system of claim 1, wherein the means for analyzing research trends identifies trends based on the year of publication and the number of citations of a paper, and further adjusts the trend information taking into account the user's emotional information.

[1765] "Application example 2 when combining emotion engines"

[1766] (Claim 1)

[1767] means for receiving a search query entered by a user;

[1768] means for searching academic databases based on the search query to identify relevant academic materials;

[1769] means for using the generative model to generate summaries of the identified academic material;

[1770] means for analyzing research trends based on the generated summaries and related information;

[1771] means for displaying the summary and trend analysis results to a user;

[1772] A means of recognizing the user's emotional state using an emotion engine and proposing and implementing search results and security measures based on that state;

[1773] A system including:

[1774] (Claim 2)

[1775] 2. The system of claim 1, wherein the generative model generates summaries of academic materials using natural language processing techniques.

[1776] (Claim 3)

[1777] 2. The system of claim 1, wherein the means for analyzing research trends identifies trends based on the publication year and number of citations of a document. [Explanation of symbols]

[1778] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving a search query entered by a user; means for searching academic databases based on the search query to identify relevant academic papers; means for using the generative model to generate summaries of the identified academic papers; means for analyzing research trends based on the generated summaries and associated metadata; means for displaying the summary and trend analysis results to a user; A system including:

2. The system of claim 1 , wherein the generative model generates summaries of academic papers using natural language processing techniques.

3. 2. The system of claim 1, wherein the means for analyzing research trends identifies trends based on the year of publication and the number of citations of papers.

Images