system

Abstract

Provide a system. A method for automatically collecting academic papers from academic databases; A means for summarizing collected academic papers using natural language processing technology; A method for automatically detecting and explaining technical terms that appear in abstracted academic papers; a means for storing the abstract and commentary in a database; a means for searching the database based on keywords and providing relevant summary information; A system including:

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] Efficient utilization of academic research resources and promotion of collaborative research between universities and companies are extremely important issues in today's regional areas. Specifically, it is difficult for regional companies to access the latest academic research and utilize it in their own research and development. Furthermore, there are limited means for universities to appropriately provide the latest research results to companies and monetize them. Furthermore, securing computing resources for research and development that requires large-scale data processing and simulations is also a major challenge. In these circumstances, there is a need to provide a system that allows for efficient use of the latest research results and achieve regional revitalization and digital transformation (DX) in the region. [Means for solving the problem]

[0005] To solve this problem, the present invention provides the following means.

[0006] First, we will build a system that automatically collects academic papers from academic databases and summarizes their contents using natural language processing technology. This system will also include a function to automatically detect technical terms that appear in the summarized papers and provide explanations for them. This will enable even non-expert users to easily understand the latest academic research findings.

[0007] Second, it provides an intermediary system to promote collaborative research between local companies and universities. In this system, companies input their research needs and issues, and universities describe the requirements for collaborative research. The system analyzes the input data from both parties and makes appropriate matches. Furthermore, by notifying the results of this matching, it is possible to smoothly advance research collaboration.

[0008] Third, by utilizing a high-power GPU data center, we provide computing resources for large-scale data processing and simulations. By accepting requests for computing resources, scheduling them, and returning the results to users, we can improve the speed and efficiency of research and development.

[0009] These measures will enable the introduction of cutting-edge technology to local areas and the efficient utilization of academic research, thereby realizing regional digital transformation and regional revitalization.

[0010] An "academic database" is an online information resource that stores and provides access to academic papers and research results.

[0011] An "academic paper" is a document that reports the results of academic research, usually published in a specialized academic journal.

[0012] "Natural language processing technology" is a general term for technology that uses computers to understand, generate, and analyze human language.

[0013] An "abstract" is a concise summary of the main points or content of a document.

[0014] "Terminology" refers to words and expressions specific to a particular field of expertise.

[0015] A "database" is a system for organizing and storing data, and for efficiently accessing and managing it.

[0016] "Keywords" are important words used to find information in search engines and databases.

[0017] "Matching" is the process of selecting suitable pairs or groups based on specific criteria.

[0018] A "high-power GPU" is a graphics processing unit that has high-performance graphics processing capabilities and enables large-scale parallel calculations.

[0019] "Computational resources" is a general term for hardware and software resources required to perform computational processing.

[0020] "Regional revitalization" is an initiative aimed at revitalizing local economies and communities and achieving sustainable development.

[0021] "Digital transformation (DX)" refers to the innovation of business models and business processes using digital technology. [Brief explanation of the drawings]

[0022] [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

[0023] 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.

[0024] First, the terms used in the following description will be explained.

[0025] 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).

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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."

[0030] [First embodiment]

[0031] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0032] 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.

[0033] 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).

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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.

[0038] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0039] 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.

[0040] 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.

[0041] 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.

[0042] 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."

[0043] This invention is a system for efficiently utilizing academic research resources in rural areas and promoting collaborative research between universities and companies. The system utilizes a high-power GPU data center to support large-scale data processing and simulation. It also provides academic paper collection, summarization, terminology explanations, and search functions, facilitating the application of research results.

[0044] Collecting and summarizing academic papers

[0045] Program operation description:

[0046] Server: Automatically collects academic papers in a specified field from each academic database. For example, to obtain the latest papers on machine learning, a query is sent to the database using APIs or scraping technology and the relevant papers are downloaded.

[0047] Server: Summarizes the collected papers using natural language processing (NLP) techniques. Specifically, after text cleaning, a summary generation algorithm is applied to extract key points.

[0048] Server: Automatically detects technical terms contained in the summary and provides explanations. Refers to a pre-prepared technical terminology dictionary database and adds appropriate explanations to the summary text.

[0049] Examples:

[0050] Server: Collects 100 of the latest papers in the field of machine learning, summarizes their contents, and extracts key points. When the technical term "backpropagation" appears, an explanation of this term is added and included in the summary.

[0051] Providing a search system

[0052] Program operation description:

[0053] User: Enter search keywords from the terminal to search for abstracts of academic papers.

[0054] Server: Based on the entered keywords, the database is searched and relevant summary information is obtained.

[0055] On the device: Display search results to the user, including links to terminology explanations and original papers.

[0056] Examples:

[0057] User: Enter the keywords "environmentally friendly plastic manufacturing techniques."

[0058] Server: Retrieves abstracts of relevant papers from the database and sends them to the user's device.

[0059] On the device: The search results show the user 20 abstracts, each containing a link to an explanation of the term "biodegradable plastic."

[0060] Intermediary for joint research

[0061] Program operation description:

[0062] User (company representative): Enters the company's research needs and issues from the terminal.

[0063] User (university professor): Enters the requirements for collaborative research from a terminal.

[0064] Server: Analyzes input data from companies and universities and makes appropriate matches.

[0065] Server: Notifies companies and universities of the matching results.

[0066] Examples:

[0067] User (company representative): Enter the topic "Research on optimizing the manufacturing process of biodegradable plastics."

[0068] User (university professor): Enters the requirement, "I would like to conduct joint research on materials development."

[0069] Server: Analyzes the data from both parties and matches compatible pairs. Upon notification, both parties contact each other and start collaboration.

[0070] Utilizing GPU data centers

[0071] Program operation description:

[0072] Users (researchers, company personnel): Input requests for computing resources from their terminals.

[0073] Server: Receives requests and schedules computing resources.

[0074] Server: Executes computational tasks in a high-powered GPU data center and returns the results to the user.

[0075] Examples:

[0076] User (company representative): Requests the necessary computing resources to conduct a simulation of the plastics manufacturing process.

[0077] Server: Assigns requests to a powerful GPU and starts the simulation.

[0078] Server: Acquires the simulation results and sends them to the user's device.

[0079] Terminal: Receives the resulting data and analyzes it using visualization tools.

[0080] This will strengthen cooperative relationships between local companies and universities, and by introducing the latest research results to local areas, it will provide a system that will revitalize the local economy and achieve sustainable growth.

[0081] The processing flow will be explained below.

[0082] Collecting and summarizing academic papers

[0083] Step 1:

[0084] Server: Sends queries to the API endpoints of each academic database, including subject areas and publication date ranges.

[0085] Example: To retrieve the latest papers in the field of machine learning, send a specified query to the API of "Academic Database A."

[0086] Step 2:

[0087] Server: Receives article data returned from academic databases and stores metadata (title, author, publication date, etc.) and the text in local storage.

[0088] Example: Parse the paper data returned from the API and save the title "Machine Learning Algorithms", the author "John Doe", and the body of the paper.

[0089] Step 3:

[0090] Server: Performs text cleaning on the received paper text, removing unnecessary parts (headers, footers, tables, etc.).

[0091] Example: Analyzing the text of a paper using regular expressions to remove unnecessary strings and structures.

[0092] Step 4:

[0093] Server: Apply natural language processing (NLP) algorithms to the cleaned-up paper text to generate summaries.

[0094] Example: Using a text summarization algorithm (e.g., Textrank) to extract the main points of an article.

[0095] Step 5:

[0096] Server: Automatically detects technical terms contained in the abstract and retrieves explanations by referencing a pre-prepared technical terminology dictionary database.

[0097] Example: Detect "backpropagation" from a summary and add its explanation "Learning algorithm in neural networks."

[0098] Step 6:

[0099] Server: Stores abstracts and their terminology in a database, and associates them with metadata.

[0100] Example: Save the abstract "This paper demonstrates the efficiency of an algorithm using backpropagation" and its explanation in a database.

[0101] Providing a search system

[0102] Step 1:

[0103] User: Enters search keywords into the device's search interface. The input is sent to the server.

[0104] Example: A user types in "environmentally friendly plastic manufacturing techniques" and presses the search button.

[0105] Step 2:

[0106] Server: Searches for abstracts in the database based on received keywords. Sorts the results based on relevance.

[0107] Example: Obtain multiple summaries that match the search keywords and sort them by relevance.

[0108] Step 3:

[0109] Server: Structures search results and sends them to the user's device, including abstracts, glossary links, and links to original articles.

[0110] Example: Send a summary and its link information to the user's device in JSON format.

[0111] Step 4:

[0112] Terminal: Displays the received search results. When the user clicks on a result, a detailed summary and glossary of terms is displayed.

[0113] Example: Search results are displayed in a list of 20 abstracts, and when the user clicks on the link for "biodegradable plastic," a pop-up explanation appears.

[0114] Intermediary for joint research

[0115] Step 1:

[0116] User (company representative): Enters the company's research needs and issues from the terminal and sends them to the server.

[0117] Example: A company enters "Research on optimizing the manufacturing process of biodegradable plastics."

[0118] Step 2:

[0119] User (university professor): Enters university research requirements from the terminal and sends them to the server.

[0120] Example: A university professor enters, "I would like to conduct joint research on materials development."

[0121] Step 3:

[0122] Server: Analyzes input data from companies and universities, calculates the degree of compatibility between research needs and requirements, and matches compatible pairs.

[0123] Example: Applying machine learning algorithms to analyze input data and calculate a matching score.

[0124] Step 4:

[0125] Server: Notifies companies and universities of match results, including contact information and next step instructions.

[0126] Example: Notify the company and university of the match results and provide contact information for both parties.

[0127] Utilizing GPU data centers

[0128] Step 1:

[0129] Users (researchers, company personnel): Enter requests for computing resources from their terminals and send them to the server.

[0130] Example: Requesting specific computational resources (number of GPUs, time, etc.) for a simulation of a plastics manufacturing process.

[0131] Step 2:

[0132] Server: Analyzes received requests and schedules them by allocating available computing resources.

[0133] Example: Check the resource status of the GPU data center and allocate an appropriate GPU.

[0134] Step 3:

[0135] Server: Executes computational tasks and initiates data processing and simulations using high-powered GPUs.

[0136] Example: Loading required datasets and launching scripts to perform computational tasks.

[0137] Step 4:

[0138] Server: Receives the calculation results and sends them to the user's device. The results are formatted in a format that can be analyzed.

[0139] Example: Convert simulation results into JSON or CSV format and send them to the user's device.

[0140] Step 5:

[0141] Terminal: Displays the received calculation results and helps the user analyze them.

[0142] Example: Visualizing calculation results in graphs and charts and presenting them to the user.

[0143] By using the above specific processing steps, the system of the present invention can realize efficient utilization of academic research, promotion of collaborative research, and large-scale data processing.

[0144] Example 1

[0145] 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."

[0146] The lack of effective methods for utilizing academic research resources in rural areas, the lack of promotion of collaborative research between universities and companies, and the lack of efficient processing of large amounts of data and simulation environments are issues that need to be addressed. Furthermore, there is a need to improve the situation where academic literature collection, summaries, explanations of technical terms, and effective search functions are not adequately provided.

[0147] 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.

[0148] In this invention, the server includes means for automatically collecting academic literature from an academic information repository, means for summarizing the collected academic literature using natural language processing technology, means for automatically detecting and defining technical terms that appear in the summarized academic literature, means for storing the summaries and definitions in the information repository, means for searching the information repository based on keywords and providing related summary information, means for displaying related information extracted from the information repository on a user terminal, means for performing large-volume data processing and simulations using a high-performance computing unit, and means for accepting and scheduling requests for computing resources, thereby enabling the effective collection and use of academic information, the promotion of collaborative research, and the efficient processing of large volumes of data.

[0149] An "academic information repository" is a database for storing, organizing, and accessing academic literature and research data.

[0150] "Academic literature" refers to documents such as papers, articles, and reports published as the results of academic research.

[0151] "Automatic collection means" refers to technology that automatically retrieves relevant literature from academic information repositories based on specific conditions or keywords.

[0152] "Natural language processing technology" is a computational technology for understanding and generating human language, and is a method for summarizing, translating, and analyzing documents.

[0153] "Summarizing" is the technique of summarizing the main points and conclusions of academic literature in a short form.

[0154] "Jargon" refers to specialized words and phrases used in a particular field or discipline.

[0155] "Means for automatic detection and definition assignment" refers to technology that identifies technical terms in text and automatically adds their meanings and explanations.

[0156] "Means for storing information in a repository" refers to the technology for storing collected literature, its summaries, and definitions in a database.

[0157] "Means for searching information repositories based on keywords" refers to a technology for searching documents and abstracts in information repositories based on keywords entered by the user.

[0158] The "means for providing related summary information" is a technique for providing the user with a summary obtained as a search result.

[0159] "Means for displaying on the user's terminal" refers to technology that displays search results, summary information, definitions of technical terms, etc. on the user's device.

[0160] A "high-performance computing unit" is high-performance hardware for efficiently processing large amounts of data and performing complex calculations.

[0161] "Means for accepting requests for computing resources and scheduling" refers to a technique in which users request the computing resources they need and determine the execution order of computing tasks based on that request.

[0162] This invention is a system for efficiently utilizing academic research resources in local areas and promoting collaborative research between academic institutions and companies. The system utilizes high-performance computing units to support large-scale data processing and simulation. It also provides academic literature collection, summarization, glossary, and search functions, facilitating the application of research results.

[0163] The main components of the system include an academic information repository, a server, terminals, and a high-performance computing unit. The specific processes and technologies used for each component are described below.

[0164] Collecting and summarizing academic literature

[0165] The server does the following:

[0166] Automatically collect academic literature in a specified field from each academic information repository. For example, by sending queries to the database using APIs or scraping technology and downloading the relevant literature. For example, by using a Python library to automate the collection process.

[0167] The collected literature is summarized using natural language processing techniques. Specifically, text cleaning is performed using Python libraries such as NLTK and Spacy, and summaries are generated by applying generative AI models such as BERT and GPT-3 (registered trademark).

[0168] Automatically detects technical terms contained in the summary and assigns definitions to them. Refers to a pre-prepared technical terminology dictionary database and adds appropriate definitions to the summary.

[0169] Examples:

[0170] The server collects 100 of the latest articles in the field of machine learning, summarizes their contents, and extracts key points. When the technical term "backpropagation" appears, it adds a definition of the term and includes it in the summary.

[0171] Example prompt sentence:

[0172] "Collect the latest papers in the field of machine learning and generate summaries, including an explanation of the technical term 'backpropagation'."

[0173] Providing a search system

[0174] The server and the device perform the following operations:

[0175] Users enter search keywords into their terminals to search for abstracts of academic literature.

[0176] The server searches the information repository based on the input keywords and retrieves relevant summary information. The search process is performed using databases such as ElasticSearch (registered trademark) and MySQL (registered trademark).

[0177] The terminal displays the search results to the user, including links to explanations of technical terms and original literature.

[0178] Examples:

[0179] The user inputs the keywords "environmentally friendly plastic manufacturing technology."

[0180] The server retrieves abstracts of relevant documents from the information repository and sends them to the user's terminal.

[0181] The device displays 20 search results and summaries to the user, each of which includes a link to an explanation of the technical term "biodegradable plastic."

[0182] Example prompt sentence:

[0183] "Displays abstracts of research papers for the search keyword 'environmentally friendly plastic manufacturing technology,' and includes links to explanations of the technical term 'biodegradable plastic.'"

[0184] Intermediary for joint research

[0185] The server does the following:

[0186] Users (company personnel) input their company's research needs and issues from the terminal.

[0187] Users (academic institutions) input the requirements for collaborative research from their terminals.

[0188] The server analyzes input information from companies and academic institutions to make appropriate matches, using machine learning algorithms to compare research needs and skill sets.

[0189] The server notifies the companies and academic institutions of the matching results.

[0190] Examples:

[0191] The user (company representative) inputs the topic of "Research on optimizing the manufacturing process of biodegradable plastics."

[0192] The user (academic institution) enters the requirement that "I would like to conduct joint research on materials development."

[0193] The server analyzes the information of both parties and matches suitable pairs. Upon notification, both parties can contact each other and begin collaborative research.

[0194] Example prompt sentence:

[0195] "Matching the corporate research needs of 'Optimizing the manufacturing process of biodegradable plastics' with the academic institution's requirements of 'Joint research on materials development'"

[0196] Utilizing high-performance computing units

[0197] The server and the device perform the following operations:

[0198] Users (researchers, company personnel) input requests for computing resources from their terminals.

[0199] The server receives requests and schedules computing resources, using Slurm or Kubernetes to efficiently manage computing tasks.

[0200] The computational task is performed on a high-performance computing unit (e.g., a high-performance GPU cluster) and the results are returned to the user.

[0201] The terminal receives the resulting data and analyzes it using a visualization tool (e.g., Tableau or Matplotlib).

[0202] Examples:

[0203] A user (company representative) requests the necessary computing resources to perform a simulation of a plastic manufacturing process.

[0204] The server assigns the request to a high-performance GPU and starts the simulation.

[0205] The server obtains the simulation results and sends them to the user's terminal.

[0206] The terminal receives the resulting data and analyzes it using visualization tools.

[0207] Example prompt sentence:

[0208] "To simulate the plastics manufacturing process, the necessary computational resources are requested from a high-performance computing unit, and the results are analyzed using visualization tools."

[0209] In this way, local businesses and academic institutions can strengthen their collaborative relationships, introduce the latest research results to local areas, and provide a system for revitalizing the local economy and achieving sustainable growth.

[0210] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0211] Collecting and summarizing academic literature

[0212] Step 1: Collecting academic literature

[0213] Server: Automatically collects academic literature in specified fields from each academic information repository.

[0214] Input: Access information for academic repository and search query

[0215] What it does: It sends a query to the API of Elsevier or arXiv to get a list of relevant documents, from which the server downloads the documents in PDF or text format.

[0216] Output: Collected academic literature

[0217] Step 2: Generate summaries of academic literature

[0218] Server: Summarizes the collected literature using natural language processing technology.

[0219] Input: Collected academic literature

[0220] How it works: It uses Python's NLTK and Spacy to preprocess text data (cleaning, tokenization, stop word removal), and then generates summaries using generative AI models such as BERT and GPT-3.

[0221] Output: Summary

[0222] Step 3: Adding explanations of technical terms

[0223] Server: Automatically detects technical terms contained in the abstract and assigns definitions.

[0224] Input: Abstract, technical dictionary database

[0225] Specific operation: Analyze the summary sentence, refer to the terminology dictionary database, and add definitions corresponding to the detected terminology to the summary sentence.

[0226] Output: Summary with definition

[0227] Providing a search system

[0228] Step 1: Enter search keywords

[0229] User: Enter search keywords from the terminal and search for abstracts of academic literature.

[0230] Input: Search keyword

[0231] Specific operation: The user enters a keyword into the device's search bar and presses the "Search" button.

[0232] Output: The search keywords are sent to the server.

[0233] Step 2: Database search

[0234] Server: Based on the entered keywords, the information repository is searched and relevant summary information is obtained.

[0235] Input: Search keyword

[0236] How it works: The server uses databases such as Elasticsearch or MySQL to perform full-text searches for keywords, and prioritizes retrieving relevant summaries.

[0237] Output: List of relevant summary sentences

[0238] Step 3: Viewing search results

[0239] Terminal: Display search results to the user.

[0240] Input: List of relevant abstract sentences

[0241] Specific operation: The terminal displays a list of abstracts, allowing the user to check their contents. The abstracts also include links to explanations of technical terms and original documents.

[0242] Output: Display of search results

[0243] Intermediary for joint research

[0244] Step 1: Enter your research needs and requirements

[0245] User (company representative): Enters the company's research needs and issues from the terminal.

[0246] Input: Details of your research needs or problems

[0247] Specific operation: The company representative enters the issue into the input form and presses the submit button.

[0248] Output: The input research needs are sent to the server.

[0249] User (academic institution): Enters the requirements for collaborative research from a terminal.

[0250] Input: Details of collaboration requirements

[0251] Specific operations: The person in charge at the academic institution enters the requirements for the collaborative research into the input form and presses the submit button.

[0252] Output: The entered collaborative research requirements are sent to the server.

[0253] Step 2: Analyze the input data

[0254] Server: Analyzes the information entered by companies and academic institutions and makes appropriate matches.

[0255] Input: Corporate research needs, academic institutions' collaborative research requirements

[0256] What it does: It uses machine learning algorithms to compare research needs with available skill sets and select the best match.

[0257] Output: Matching results

[0258] Step 3: Notification of match results

[0259] Server: Notifies companies and academic institutions of the matching results.

[0260] Input: Matching results

[0261] Specific actions: Companies and academic institutions will be notified of the matching results via email and in-system notifications.

[0262] Output: Matching notification

[0263] Utilizing high-performance computing units

[0264] Step 1: Submit a request for computing resources

[0265] Users (researchers, company personnel): Input requests for computing resources from their terminals.

[0266] Input: Details of computing resources (number of GPUs, amount of memory, etc.)

[0267] Specific operation: A researcher or company representative inputs the required computing resources and presses the request submit button.

[0268] Output: A request for computational resources is sent to the server.

[0269] Step 2: Scheduling resources

[0270] Server: Receives requests and schedules computing resources.

[0271] Input: Computational resource request

[0272] How it works: Servers use resource management tools such as Slurm and Kubernetes to efficiently allocate and schedule computational tasks.

[0273] Output: Scheduling information

[0274] Step 3: Execute the computational task

[0275] Server: Executes computational tasks on high performance computing units and produces results.

[0276] Input: Scheduling information

[0277] Specific operation: The computation task is executed on a high-performance computing unit (e.g., a high-performance GPU cluster) and the computation is monitored until completion.

[0278] Output: calculation result

[0279] Step 4: Return and display results

[0280] Server: Returns the calculation results to the user.

[0281] Input: Calculation result

[0282] Specific operation: Send the calculation result to the user's device.

[0283] Output: The calculation result is sent to the terminal.

[0284] Terminal: Receives the resulting data and analyzes it using visualization tools.

[0285] Input: Submitted calculation result

[0286] Specific operation: Download the result data and perform data analysis and visualization using visualization tools such as Tableau or Matplotlib.

[0287] Output: Visualized result data

[0288] The above steps will concretely implement the collection of academic literature, the generation of summaries, the provision of a search system, the mediation of collaborative research, and the effective utilization of high-performance computing units.

[0289] (Application example 1)

[0290] 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."

[0291] Modern manufacturing requires the rapid adoption of new technologies and research results, and the optimization of manufacturing processes. However, there has been insufficient promotion of efficient use of academic research resources and collaborative research between universities and companies in regional areas. As a result, regional manufacturing companies are unable to quickly utilize the latest research results, which could lead to a decline in their competitiveness. Another issue is the difficulty of securing the high-power computing resources required for manufacturing process simulations and data calculations.

[0292] 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.

[0293] In this invention, the server includes means for automatically collecting academic papers from academic databases, means for summarizing the collected academic papers using natural language processing technology, means for automatically detecting technical terms appearing in the summarized academic papers and providing explanations, means for storing the summaries and explanations in a database, means for searching the database based on keywords and providing related summary information, means for displaying search results on a user interface, means for utilizing a group of high-performance computers to support simulations and data calculations, means for requesting computing resources, scheduling and executing tasks, and means for returning calculation results to the user. This enables the utilization of the latest academic papers, the improvement of manufacturers' technologies, and the efficient use of high-performance computing resources.

[0294] An "academic database" is a database that collects and stores academic papers and research results, and allows them to be searched and viewed.

[0295] An "academic paper" is a written document that summarizes the results of academic research and is published in academic journals, conference proceedings, etc.

[0296] "Natural language processing technology" is a technology that allows computers to understand human language and analyze and generate it.

[0297] A "summary" is a short, concise summary of the most important parts of an original text.

[0298] "Terminology" refers to terms or words used professionally in a particular field.

[0299] "Explanation" means explaining technical terms and complex content in an easy-to-understand manner.

[0300] A "database" is a computer device that systematically organizes and stores large amounts of information, making it easy to search and extract.

[0301] "Keywords" are typical words used when searching and classifying data.

[0302] A "user interface" is a screen or operating means for interaction between a user and a system.

[0303] "Simulation" is a method of virtually reproducing real-world events and processes for analysis and prediction.

[0304] A "high-power computing group" is a group of multiple high-performance computing devices that can simultaneously execute large-scale computational processes.

[0305] "Computational resources" refers to the hardware and software required to execute computational processes.

[0306] "Task scheduling" refers to efficiently managing the use of computing resources and determining the execution order and allocation of each task.

[0307] The purpose of this invention is to provide a system that promotes the efficient use of academic research resources and collaborative research between universities and companies in order to contribute to industrial applications. This system includes academic paper collection, abstract generation, terminology explanation, search functionality, and the use of a group of powerful computers. The specific processing and details of each part of the system are explained below.

[0308] The system consists of a server and client terminals, and the server has the following main functions:

[0309] 1. Collection of academic papers

[0310] The server automatically collects academic papers from academic databases using API calls and web scraping techniques. Specifically, it periodically collects papers in a specified field and stores them in a database.

[0311] 2. Abstract generation for papers

[0312] The server then uses natural language processing (NLP) techniques to summarize the collected academic papers. This process uses generative AI models such as BERT and GPT. After cleaning the text of the papers, it extracts key points and generates concise summaries.

[0313] 3. Adding technical term explanations

[0314] The server automatically detects technical terms contained in the abstract and provides explanations by referencing a pre-prepared technical terminology dictionary database.

[0315] 4. Providing a search system

[0316] The server provides a system that allows users to search for abstracts of academic papers by entering keywords, using Elasticsearch and other search algorithms to provide fast and accurate search results.

[0317] 5. Intermediary for collaborative research

[0318] Users can input their company's research needs and issues, and universities can input their collaborative research requirements. The server analyzes this data, matches and notifies compatible pairs.

[0319] 6. Utilizing high-power computers

[0320] The server uses high-performance computing resources to process large amounts of data and perform simulations. For example, we use NVIDIA GPU Cloud. The system schedules computational tasks and returns the results to the user after execution.

[0321] Examples:

[0322] If a user enters the keyword "optimization of an automated storage system using an AI robot," the server will collect abstracts of related papers and provide search results with explanations of the technical terms "backpropagation" and "reinforcement learning." The specific prompt for the generative AI model is as follows:

[0323] Example prompt sentence:

[0324] "Collect the latest academic papers on manufacturing technology and summarize the content related to 'Automated Storage System Using AI Robots'. Also, provide explanations of technical terms such as 'backpropagation' and 'reinforcement learning'."

[0325] As described above, this system will enable the rapid introduction of new technologies in the manufacturing industry and the efficient utilization of research results.

[0326] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0327] Step 1:

[0328] The server collects academic papers from academic databases using API calls and web scraping technology. The input is query information in a specified field, and the output is the collected academic papers. This allows for efficient collection of highly relevant academic papers.

[0329] Step 2:

[0330] The server performs text cleaning on the collected academic papers. Specific examples of cleaning include removing HTML tags and unnecessary spaces. The input is the raw paper text collected in step 1, and the output is the cleaned paper text. This creates a foundation for subsequent processing to be performed accurately.

[0331] Step 3:

[0332] The server summarizes the paper using natural language processing technology. Specifically, it applies a generative AI model (e.g., BERT or GPT). The input is the paper text cleaned in step 2, and the output is a summary that extracts the important points. This allows users to grasp the gist of a lengthy paper in a short amount of time.

[0333] Step 4:

[0334] The server automatically detects technical terms that appear in the summarized paper and provides explanations. This process refers to a technical dictionary. The input is the summary generated in step 3, and the output is the summary with explanations. This makes it easier for users with insufficient technical knowledge to understand the content.

[0335] Step 5:

[0336] The server stores the summary and commentary in a database. The input is the summary with commentary obtained in step 4, and the output is the data stored in the database. This makes it easy to search and reference later.

[0337] Step 6:

[0338] Users can search for abstracts of academic papers by entering keywords from a client terminal. The input is the keywords entered by the user, and the output is a list of abstracts of the search results. This allows users to quickly obtain the information they need.

[0339] Step 7:

[0340] The server searches the database using a search algorithm (e.g., Elasticsearch). The input is the keyword entered by the user in step 6, and the output is a list of related summary sentences. This returns highly accurate search results.

[0341] Step 8:

[0342] The server displays the search results on a user interface. The input is the list of abstracts obtained in step 7, and the output is a search result screen displayed on the user's device. This allows the user to obtain information in an easy-to-view format.

[0343] Step 9:

[0344] Corporate users input their research needs and issues. The input is information about the company's research topics and needs, and the output is that information saved in a database. This provides the basic information for collaborative research.

[0345] Step 10:

[0346] University users input the requirements for collaborative research. The input is the university's research requirements and desired items, and the output is that information saved in a database. This enables matching with researchers.

[0347] Step 11:

[0348] The server analyzes the input data from companies and universities and matches suitable pairs. The input is the data obtained in steps 9 and 10, and the output is the matching results. This determines the optimal collaborative research partner.

[0349] Step 12:

[0350] The server notifies the company and the university of the matching results. The input is the matching result obtained in step 11, and the output is the state in which the notification has been sent. This allows the company and the university to communicate with each other and begin collaborative research.

[0351] Step 13:

[0352] The server uses a group of powerful computers to perform simulations and data processing. The input is the user request, and the output is the calculation result. This allows for efficient processing of large amounts of data.

[0353] Step 14:

[0354] The server schedules computing resources. The input is a request for computing resources, and the output is a scheduled state. This allows for optimal allocation of computing resources.

[0355] Step 15:

[0356] The server returns the calculation results to the user. The input is the calculation result obtained in step 13, and the output is the calculation result sent to the user's terminal. This allows the user to analyze and utilize the results.

[0357] Step 16:

[0358] The user receives the calculation results and analyzes them using a visualization tool. The input is the calculation results obtained in step 15, and the output is visualized data. This makes it easier for the user to intuitively understand the data.

[0359] 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.

[0360] This system aims to efficiently utilize academic research resources in rural areas and promote collaborative research between universities and companies. It also aims to improve the user experience by incorporating an emotion engine that recognizes user emotions. The system utilizes a high-power GPU data center to support large-scale data processing and simulation. It also provides academic paper collection, summarization, terminology explanations, and search functions, as well as an interface that takes the user's psychological state into account through the emotion engine.

[0361] Collecting and summarizing academic papers

[0362] Program operation description:

[0363] Server: Automatically collects academic papers in specified fields from each academic database. It sends queries to the database using APIs and scraping technology and downloads the relevant papers.

[0364] Server: Generates summaries of collected papers using natural language processing (NLP) techniques, performs text cleaning, and applies summary generation algorithms to extract key points.

[0365] Server: Automatically detects technical terms contained in the abstract and retrieves explanations by referencing a pre-prepared technical terminology dictionary database.

[0366] Server: Stores abstracts and their terminology in a database, and associates them with metadata.

[0367] Examples:

[0368] Server: Collects 100 of the latest papers in the field of machine learning, summarizes their contents, and extracts key points. For example, it adds an explanation of the term "backpropagation" that appears in the paper to the summary.

[0369] Providing a search system

[0370] Program operation description:

[0371] User: Enter search keywords from the terminal to search for abstracts of academic papers.

[0372] Server: Searches the database based on the entered keywords and retrieves relevant abstracts.

[0373] On the device: Display search results to the user, including links to terminology explanations and original papers.

[0374] Examples:

[0375] User: Enter the keywords "environmentally friendly plastic manufacturing techniques."

[0376] Server: Retrieves abstracts of relevant papers from the database and sends them to the user's device.

[0377] On the device: The search results show the user 20 abstracts, each containing a link to an explanation of the term "biodegradable plastic."

[0378] Intermediary for joint research

[0379] Program operation description:

[0380] User (company representative): Enters the company's research needs and issues from the terminal and sends them to the server.

[0381] User (university professor): Enters university research requirements from the terminal and sends them to the server.

[0382] Server: Analyzes input data from companies and universities, calculates the degree of compatibility between research needs and requirements, and matches compatible pairs.

[0383] Server: Notifies companies and universities of the matching results.

[0384] Examples:

[0385] User (company representative): Enter the topic "Research on optimizing the manufacturing process of biodegradable plastics."

[0386] User (university professor): Enters the requirement, "I would like to conduct joint research on materials development."

[0387] Server: Analyzes the data from both parties and matches compatible pairs. Upon notification, both parties contact each other and start collaboration.

[0388] Utilizing GPU data centers

[0389] Program operation description:

[0390] Users (researchers, company personnel): Enter requests for computing resources from their terminals and send them to the server.

[0391] Server: Analyzes received requests and schedules them by allocating available computing resources.

[0392] Server: Executes computational tasks and initiates data processing and simulations using high-powered GPUs.

[0393] Server: Receives the calculation results and sends them to the user's device. The results are formatted in a format that can be analyzed.

[0394] Terminal: Displays the received calculation results and helps the user analyze them.

[0395] Examples:

[0396] User (company representative): Requests the necessary computing resources to conduct a simulation of the plastics manufacturing process.

[0397] Server: Assigns requests to a powerful GPU and starts the simulation.

[0398] Server: Acquires the simulation results and sends them to the user's device.

[0399] Terminal: Receives the resulting data and analyzes it using visualization tools.

[0400] Combining Emotion Engines

[0401] Program operation description:

[0402] User: When searching for academic papers or entering collaborative research on the device, the emotion recognition system analyzes the user's facial expressions, voice, and text input.

[0403] Server: The emotion engine analyzes the user's emotions in real time and optimizes the user experience based on the results.

[0404] Example: Adjusting the relevance or presentation of search results when users express dissatisfaction or confusion with the results.

[0405] Terminal: Presents information optimized by the emotion engine to the user.

[0406] Examples:

[0407] User: If you express frustration while searching for "eco-friendly plastic manufacturing techniques," the emotion engine will detect that.

[0408] Server: Re-analyzes the search results and presents more relevant results.

[0409] On your device: Show new search results and give you quick access to the information you want.

[0410] This will provide a system that takes user emotions into consideration, enable the introduction of cutting-edge technology to local areas, and make efficient use of academic research, thereby realizing regional digital transformation and regional revitalization.

[0411] The processing flow will be explained below.

[0412] Collecting and summarizing academic papers

[0413] Step 1:

[0414] Server: Sends queries to the API endpoints of each academic database, including subject areas and publication date ranges.

[0415] Example: To retrieve the latest papers in the field of machine learning, send the query "machine learning AND 2023" to the API of "Academic Database A."

[0416] Step 2:

[0417] Server: Receives article data returned from academic databases and stores metadata (title, author, publication date, etc.) and the text in local storage.

[0418] Example: Analyze paper data returned from an API and save the title "Machine Learning Algorithms", author "John Doe", and text to local storage.

[0419] Step 3:

[0420] Server: Performs text cleaning on the received paper text, removing unnecessary parts (headers, footers, tables, etc.).

[0421] Example: Using regular expressions to remove unnecessary tags and special characters from the body of a paper.

[0422] Step 4:

[0423] Server: Apply natural language processing (NLP) algorithms to the cleaned-up paper text to generate summaries.

[0424] Example: Extracting key sentences using the Textrank algorithm for text summarization.

[0425] Step 5:

[0426] Server: Automatically detects technical terms contained in the abstract and retrieves explanations by referencing a pre-prepared technical terminology dictionary database.

[0427] Example: Detect "backpropagation" from a summary and add an explanation "neural network learning algorithm."

[0428] Step 6:

[0429] Server: Stores the abstracts and their terminology in a database, and associates them with metadata.

[0430] Example: Save the abstract "This paper discusses the performance of an algorithm using backpropagation" in a database.

[0431] Providing a search system

[0432] Step 1:

[0433] User: Enter search keywords into the device's search interface and press the search button.

[0434] Example: Enter "environmentally friendly plastic manufacturing techniques" and click the search button.

[0435] Step 2:

[0436] Server: Searches for abstracts in the database based on received keywords. Sorts the results based on relevance.

[0437] Example: Search the database using the keyword "environmentally friendly plastic manufacturing technology," retrieve relevant abstracts, and sort them by relevance.

[0438] Step 3:

[0439] Server: Structures the search results and sends them to the user's device, including any necessary glossary links.

[0440] Example: Send a summary and link information in JSON format.

[0441] Step 4:

[0442] Terminal: Displays the received search results. When the user clicks on a result, a detailed summary and glossary is displayed.

[0443] Example: Search results are displayed in a list of 20 abstracts, and when the user clicks on the link for "biodegradable plastic," a pop-up explanation appears.

[0444] Intermediary for joint research

[0445] Step 1:

[0446] User (company representative): Enters the company's research needs and issues into the input form on the device.

[0447] Example: A company writes "Research into optimizing the manufacturing process of biodegradable plastics."

[0448] Step 2:

[0449] User (university professor): Enter the requirements for the collaborative research into the input form on the terminal.

[0450] Example: A university professor enters, "I would like to conduct joint research on materials development."

[0451] Step 3:

[0452] Server: Analyzes input data from companies and universities, calculates the degree of compatibility between research needs and requirements, and matches compatible pairs.

[0453] Example: Machine learning algorithms analyze input data and match companies and universities with good matches.

[0454] Step 4:

[0455] Server: Notifies companies and universities of match results, including contact information and next step instructions.

[0456] Example: Notify the match results by email and provide contact information for the company and university.

[0457] Utilizing GPU data centers

[0458] Step 1:

[0459] Users (researchers, company personnel): Enter requests for computing resources using the request form on their terminal.

[0460] Example: Requesting specific computational resources (number of GPUs, time, etc.) for a simulation of a plastics manufacturing process.

[0461] Step 2:

[0462] Server: Analyzes received requests and schedules them by allocating available computing resources.

[0463] Example: Check the resource status of the GPU data center and allocate an appropriate GPU.

[0464] Step 3:

[0465] Server: Executes computational tasks and initiates data processing and simulations using high-powered GPUs.

[0466] Example: Loading required datasets and launching scripts to perform computational tasks.

[0467] Step 4:

[0468] Server: Receives the calculation results and sends them to the user's device. The results are formatted in a format that can be analyzed.

[0469] Example: Convert simulation results into JSON or CSV format and send them to the user's device.

[0470] Step 5:

[0471] Terminal: Displays the received calculation results and helps the user analyze them.

[0472] Example: Visualizing calculation results in graphs and charts and presenting them to the user.

[0473] Combining Emotion Engines

[0474] Step 1:

[0475] User: When searching for academic papers or entering collaborative research on the device, the emotion recognition system analyzes the user's facial expressions, voice, and text input.

[0476] Example: A user provides facial expression data via a webcam and also provides voice input.

[0477] Step 2:

[0478] Server: The emotion engine analyzes the user's emotions in real time and optimizes the user experience based on the results.

[0479] Example: An emotion engine detects frustration or anxiety and adaptively modifies the interface.

[0480] Step 3:

[0481] Server: Re-analyzes search results and adjusts how information is presented based on the user's sentiment.

[0482] Example: If a user expresses frustration, reevaluate the relevance of search results and present more relevant results.

[0483] Step 4:

[0484] Terminal: Presents information optimized by the emotion engine to the user.

[0485] Example: Visually highlighting new search results and related information to help users quickly access the information they need.

[0486] This will provide a system that takes user emotions into consideration, enable the introduction of cutting-edge technology to local areas, and make efficient use of academic research, thereby realizing regional digital transformation and regional revitalization.

[0487] Example 2

[0488] 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."

[0489] Modern academic research is highly specialized, making it difficult for researchers to efficiently collect, analyze, and search large amounts of academic information. It is also challenging to facilitate collaborative research between companies and educational institutions, and to efficiently manage data processing and simulations that require high-performance computing resources. Furthermore, there is a need to improve the user experience by providing an interface that takes user emotions into consideration.

[0490] 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.

[0491] In this invention, the server includes: means for automatically collecting academic papers from an academic information database; means for summarizing the collected academic papers using natural language processing technology; means for automatically detecting technical terms appearing in the summarized academic papers and providing explanations; means for storing the summaries and explanations in an information storage device; means for searching the information storage device based on search keywords and providing relevant summary information; and means for analyzing a user's emotional state using emotion recognition technology and adjusting the information provided. This allows researchers and corporate personnel to efficiently collect, summarize, and search large amounts of academic information, and further enables them to provide appropriate information taking user emotions into consideration. The system can also be used as an intermediary system to facilitate collaborative research between companies and educational institutions, and efficiently execute data processing and simulations using high-performance computing devices.

[0492] An "academic information database" is a collection of information that stores academic papers and research results.

[0493] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate human language.

[0494] An "abstract" is a concise summary of the key points extracted from an academic paper.

[0495] "Terminology" is the specific terminology used in a particular field or area.

[0496] "Explanation" refers to documents or information that provide detailed explanations of technical terms, etc.

[0497] An "information storage device" is a physical or virtual device for storing data or information.

[0498] "Search keywords" are specific words or phrases used when searching for information.

[0499] "Emotion recognition technology" is a technology that determines emotions by analyzing a user's facial expressions, voice, text, etc.

[0500] "Collaborative research" refers to research activities conducted jointly by multiple researchers or organizations.

[0501] The "intermediary system" is a system that supports matching collaborative research between companies and educational institutions.

[0502] A "high performance computing device" is a device equipped with computing resources that can process large amounts of data and perform calculations at high speed.

[0503] MODE FOR CARRYING OUT THE INVENTION

[0504] The present invention is a system for efficiently collecting, analyzing, and searching academic information and mediating collaborative research. This system is composed of the following multiple modules:

[0505] Collecting and summarizing academic papers

[0506] Server: Sends queries to academic information databases (e.g., IEEE Xplore and PubMed) using APIs and scraping technologies (e.g., BeautifulSoup, Scrapy) to automatically collect academic papers in the specified field. This process allows researchers to collect the information they need without any effort.

[0507] Server: Summarizes the collected papers using natural language processing techniques (e.g., BERT, GPT-3). First, text cleaning is performed to remove unnecessary parts, and then a summary generation algorithm is used to extract the key points of the paper.

[0508] Server: Automatically detects technical terms contained in the abstract and provides explanations by referencing a pre-prepared technical terminology dictionary database. This method makes it possible to provide information that is easy to understand even for non-expert users.

[0509] Server: Stores the generated abstracts and their descriptions in an information storage device (e.g., a database), and also associates and stores the metadata of the collected papers (author names, paper titles, publication dates, etc.).

[0510] Example: A server collects 100 of the latest papers in the field of machine learning, summarizes their contents using the BERT model, adds an explanation to the technical term "backpropagation" in the paper, and saves it in an information storage device.

[0511] Providing a search system

[0512] User: Enter keywords to search for abstracts of academic papers from a terminal.

[0513] Server: Searches the information storage device based on the entered keywords and retrieves related abstracts.

[0514] Terminal: The search results are displayed to the user, including links to terminology explanations and original papers.

[0515] Example: A user enters the search keyword "environmentally friendly plastic manufacturing technology," and the server retrieves the corresponding abstracts and sends them to the terminal. The terminal displays 20 abstracts as search results, including a link to an explanation of "biodegradable plastic."

[0516] Intermediary for joint research

[0517] User (company representative): Enters the company's research needs and issues from the terminal and sends them to the server.

[0518] User (university professor): Enters university research requirements from the terminal and sends them to the server.

[0519] Server: Analyzes input data from companies and educational institutions, calculates the degree of compatibility between research needs and requirements, matches compatible pairs, and notifies both parties.

[0520] Example: A company representative inputs "Research into optimizing the manufacturing process of biodegradable plastics," and a university professor inputs "I would like to conduct joint research on materials development." The server analyzes the data of both parties and matches and notifies compatible pairs based on common research themes.

[0521] Utilizing high-performance computing devices

[0522] Users (researchers, company personnel): Enter requests for computing resources from their terminals and send them to the server.

[0523] Server: Analyzes received requests and schedules them by allocating available computing resources.

[0524] Server: Executes computational tasks and initiates data processing and simulations using high performance computing devices (e.g., NVIDIA Tesla).

[0525] Server: Receives the calculation results and sends them to the user's device. The results are formatted in a format that can be analyzed.

[0526] Terminal: Displays the received calculation results and helps the user analyze them.

[0527] Example: A company representative requests computing resources to run a simulation of a plastics manufacturing process. The server assigns the task to a high-performance computing device, runs the simulation, and sends the results to the terminal, which displays the results using a visualization tool.

[0528] Combining Emotion Engines

[0529] User: When searching for academic papers or entering collaborative research on the device, the emotion recognition system analyzes the user's facial expressions, voice, and text input.

[0530] Server: The emotion engine analyzes user emotions in real time and optimizes the user experience based on the results. For example, if a user expresses dissatisfaction or confusion about a search result, it adjusts the relevance and presentation of the search result.

[0531] Terminal: Presents information optimized by the emotion engine to the user.

[0532] Example: If a user expresses frustration while searching for "eco-friendly plastic manufacturing techniques," the emotion engine detects this and the server re-analyzes the search results to present more relevant results. The device then displays new search results, helping the user quickly access the information they are looking for.

[0533] This will provide a system that takes user emotions into consideration, enable the introduction of cutting-edge technology to local areas, and make efficient use of academic research, thereby realizing regional digital transformation (DX) and regional revitalization.

[0534] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0535] Collecting and summarizing academic papers

[0536] Step 1: Collecting academic papers

[0537] Server: Sends queries to each academic database (e.g., IEEE Xplore or PubMed) using APIs or scraping technologies (e.g., BeautifulSoup, Scrapy) to automatically collect academic papers in the specified field.

[0538] Input: URL of academic database API or web page

[0539] Output: Collected academic paper data (text format)

[0540] Specific operation: The server sends a query about "autonomous driving technology" to the IEEE Xplore API and receives data on relevant papers.

[0541] Step 2: Text Cleaning

[0542] Server: Cleans the collected text, removing unnecessary tags and characters, and formatting it into a parseable format.

[0543] Input: Collected academic paper data

[0544] Output: Cleaned text data

[0545] Specific operation: The server removes HTML tags from the paper text and standardizes line break codes.

[0546] Step 3: Summary generation

[0547] Server: Generates a summary of the paper using natural language processing (NLP) techniques (e.g., BERT, GPT-3), extracting key points and summarizing them concisely.

[0548] Input: Cleaned text data

[0549] Output: Summary

[0550] What it does: The server uses the BERT model to generate a summary about "sensor fusion techniques for self-driving cars" from the cleaned text.

[0551] Step 4: Detect and explain terminology

[0552] Server: Automatically detects technical terms contained in the abstract and provides explanations by referencing a technical term dictionary database.

[0553] Input: Abstract, technical dictionary database

[0554] Output: Summary with glossary

[0555] Specific operation: The server detects the technical term "sensor fusion" in the summary text and retrieves its definition from the dictionary database.

[0556] Step 5: Store in the database

[0557] Server: Stores the abstracts and their terminology in an information storage device. Metadata of collected papers (author names, paper titles, publication dates, etc.) is also associated and saved.

[0558] Input: Abstract, glossary, metadata

[0559] Output: Abstract and commentary stored in a database

[0560] Specific operation: The server stores the abstract and terminology in an information storage device and correctly associates the respective metadata.

[0561] Providing a search system

[0562] Step 1: Enter search keywords

[0563] User: Enter search keywords from the device.

[0564] Input: Search keyword

[0565] Output: Search request to the server

[0566] What happens: A user types "eco-friendly plastic manufacturing techniques" into their browser's search bar.

[0567] Step 2: Database search

[0568] Server: Searches the information storage device based on the entered keywords and retrieves related abstracts.

[0569] Input: Search keyword

[0570] Output: Related summary sentence data

[0571] Specific operation: The server queries the database based on the keyword "environmentally friendly plastic manufacturing technology" and retrieves related abstracts.

[0572] Step 3: Submit and view search results

[0573] Terminal: The search results are displayed to the user, including links to terminology explanations and original papers.

[0574] Input: relevant summary sentence data

[0575] Output: Search results displayed to the user

[0576] Specific behavior: The device displays the search results in a list format, including a technical link for "biodegradable plastic."

[0577] Intermediary for joint research

[0578] Step 1: Input your research needs and problems

[0579] User (company representative): Enters the company's research needs and issues from the terminal and sends them to the server.

[0580] Input: Corporate research needs and challenges

[0581] Output: Research needs data to server

[0582] Specific actions: A company representative enters "Research on optimizing the manufacturing process of biodegradable plastics" into a web form and presses the submit button.

[0583] Step 2: Enter your university research requirements

[0584] User (university professor): Enters university research requirements from the terminal and sends them to the server.

[0585] Input: University study requirements

[0586] Output: Research requirements data to the server

[0587] Specific operation: A university professor enters "I would like to conduct joint research on materials development" into a web form and presses the submit button.

[0588] Step 3: Data analysis and goodness-of-fit calculation

[0589] Server: Analyzes input data from companies and educational institutions, calculates the degree of compatibility between research needs and requirements, and matches compatible pairs.

[0590] Input: Corporate research needs data, University research requirements data

[0591] Output: Matching result data

[0592] How it works: The server compares the company's needs with the university's requirements and calculates the degree of fit based on common research themes.

[0593] Step 4: Notification of match results

[0594] Server: Notifies companies and universities of the matching results.

[0595] Input: Matching result data

[0596] Output: Send notification

[0597] Specific operation: The server sends the matching results to the company representative and university professor using email and notification functions.

[0598] Utilizing high-performance computing devices

[0599] Step 1: Request Computational Resources

[0600] Users (researchers, company personnel): Enter requests for computing resources from their terminals and send them to the server.

[0601] Input: A request for compute resources

[0602] Output: Request data to the server

[0603] Specific operation: A user requests computational resources to perform a "simulation of a plastics manufacturing process."

[0604] Step 2: Scheduling and resource allocation

[0605] Server: Analyzes received requests and schedules them by allocating available computing resources.

[0606] Input: Request data

[0607] Output: Resource allocation data

[0608] What it does: The server makes a powerful GPU (e.g., NVIDIA Tesla) available and assigns it to the requested task.

[0609] Step 3: Execute the computational task

[0610] Server: Executes computational tasks and initiates data processing and simulations using high-performance computing devices.

[0611] Input: Resource allocation data

[0612] Output: Calculation result data

[0613] What it does: The server runs simulations on a GPU cluster and monitors the execution status.

[0614] Step 4: Obtaining and sending the calculation results

[0615] Server: Receives the calculation results and sends them to the user's device. The results are formatted in a format that can be analyzed.

[0616] Input: Calculation result data

[0617] Output: Formatted calculation result data

[0618] Specific operation: The server collects the simulation results and sends them to the user in CSV format.

[0619] Step 5: Displaying results and supporting analysis

[0620] Terminal: Displays the received calculation results and helps the user analyze them.

[0621] Input: Formatted calculation result data

[0622] Output: Visualized result data

[0623] Specific operation: The terminal imports the CSV file and visualizes the data using a tool for displaying graphs (e.g., Matplotlib).

[0624] Combining Emotion Engines

[0625] Step 1: Obtaining emotion data

[0626] User: When searching for academic papers or entering collaborative research on the device, the emotion recognition system analyzes the user's facial expressions, voice, and text input.

[0627] Input: User's facial expression data, voice data, text data

[0628] Output: Emotion data to the server

[0629] How it works: The user performs a search using the camera and microphone, and the facial expressions and voice recordings are sent to the emotion engine in real time.

[0630] Step 2: Sentiment analysis and feedback

[0631] Server: The emotion engine analyzes user emotions in real time and optimizes the user experience based on the results. Example: If a user expresses dissatisfaction or confusion with a search result, the relevance and presentation of the search result can be adjusted.

[0632] Input: Emotion data

[0633] Output: Optimized search result data

[0634] What it does: The server uses Emotion AI to detect "irritation" from facial expressions and re-runs the search algorithm to optimize the results.

[0635] Step 3: Viewing optimized information

[0636] Terminal: Presents information optimized by the emotion engine to the user.

[0637] Input: Optimized search result data

[0638] Output: Optimized display of information to the user

[0639] What it does: Your device displays reparsed search results, allowing you to quickly access the information you want.

[0640] This will provide a system that takes user emotions into consideration, enable the introduction of cutting-edge technology to local areas, and make efficient use of academic research, thereby realizing regional digital transformation (DX) and regional revitalization.

[0641] (Application example 2)

[0642] 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."

[0643] In the field of modern research and development, it is important to efficiently utilize local academic research resources and promote collaborative research between universities and companies. However, collecting and understanding related academic information is difficult, and there is a lack of efficient means to search and use it. Furthermore, there is a lack of interfaces that take the user's psychological state into consideration, and there is a need to improve the user experience. No systems have been proposed to address these issues.

[0644] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0645] In this invention, the server includes: means for automatically collecting academic papers from an academic database; means for summarizing the collected academic data using natural language processing technology; means for automatically detecting technical terms that appear in the summarized academic data and providing explanations; means for storing the summaries and explanations in a database; means for searching the database based on keywords and providing related summary information; and means including an emotion recognition engine that recognizes the user's emotional state and optimizes search results and the user interface. This enables efficient collection and utilization of academic information and an improved user experience.

[0646] An "academic database" is a collection of data containing academic papers and research materials that researchers can access to obtain information.

[0647] An "academic paper" is a document in which a researcher describes in detail the results of his or her research, and is usually published in a specialist journal or at an academic conference.

[0648] "Collection means" refers to a method or device for systematically collecting specific information, and in this system it mainly refers to APIs and scraping technology.

[0649] "Natural language processing technology" is a technology for understanding and processing human language, and is used for generating summaries and text analysis.

[0650] A "summarization means" is a method or device for summarizing lengthy information in a short and concise manner.

[0651] "Terminology" is a word with a specific meaning used in a particular technical area or field.

[0652] "Explanatory means" refers to a method or device that explains specialized information in an easy-to-understand manner.

[0653] A database is a collection of information that efficiently stores and manages data and allows necessary information to be quickly searched.

[0654] A "search means" is a method or device for searching for information in a database based on specific keywords or conditions.

[0655] "Related information" is information that is determined to be semantically related to a particular theme or keyword.

[0656] An "emotion recognition engine" is a system that analyzes and recognizes emotions from a user's facial expressions, voice, text input, etc.

[0657] "User interface optimization" refers to improving the system's display and operation methods to enhance the user experience.

[0658] This invention is a system for promoting efficient utilization of academic research resources in local areas and collaborative research between universities and companies, and also aims to improve the user experience by combining it with an emotion engine that recognizes the user's emotions.

[0659] System Configuration

[0660] The system includes the following main components:

[0661] Server: Includes a means for automatically collecting academic papers from academic databases and summarizing the collected academic data using natural language processing technology.

[0662] Server: Includes a means to automatically detect and explain technical terms that appear in summarized academic data.

[0663] Server: Contains a means for storing abstracts and descriptions in a database, searching the database based on keywords, and providing relevant abstract information.

[0664] Server: Includes an emotion recognition engine that recognizes the user's emotional state and optimizes search results and the user interface.

[0665] Terminal: Includes a means for users to input research needs and issues of companies and for universities to input requirements for collaborative research.

[0666] Server: Includes the means to analyze the input data of companies and universities and match suitable pairs.

[0667] Server: Includes a means to notify matching results.

[0668] Server: Includes means to use high-powered GPUs to process large amounts of data and perform simulations.

[0669] Hardware and software used

[0670] The following hardware and software are used to realize this system.

[0671] Hardware: General purpose personal computer, webcam, powerful GPU

[0672] Software: Python, requests, BeautifulSoup, Transformers, OpenCV, Tkinter

[0673] Specific processing

[0674] 1. The server collects academic papers in a specified field from each academic database using APIs and scraping technology. For example, it collects papers in the field of machine learning and generates summaries.

[0675] 2. The server uses natural language processing (NLP) technology to clean the collected academic data, generate summaries, and automatically detect technical terms and provide explanations.

[0676] 3. The server stores the generated abstracts and descriptions in a database, where they are associated with metadata and can be efficiently searched.

[0677] 4. The user inputs keywords through the terminal, and the server searches the database based on the keywords and returns relevant summary information. For example, it presents search results related to "environmentally friendly plastic manufacturing technology."

[0678] 5. The server analyzes the user's facial expressions in real time using OpenCV to recognize the user's emotional state. The emotion recognition engine analyzes the user's emotions and, if they indicate dissatisfaction or confusion, optimizes the search results.

[0679] 6. The server and terminals provide a means for companies to input their research needs and issues, and for universities to input their collaborative research requirements. The data is analyzed and suitable pairs are matched. Notifications are sent, enabling companies and universities to conduct collaborative research.

[0680] 7. The server uses a powerful GPU to accept requests for computing resources, perform large-scale data processing and simulations, and return the results to the user's device. For example, it provides the results of a simulation of a plastic manufacturing process.

[0681] Examples of concrete examples and prompts

[0682] Examples:

[0683] User (factory researcher): "I want to know about the latest research on manufacturing processes for new materials," and enters the search keywords "new materials manufacturing process."

[0684] System: Collects relevant recent papers from arXiv and generates summaries for each.

[0685] User: Review the generated summary and refer to the glossary of technical terms.

[0686] Emotion engine: Analyzes the user's facial expressions and re-suggests search results if they indicate dissatisfaction or confusion.

[0687] Example prompt for a generative AI model:

[0688] "Generate a paper summary: 'Please generate a summary of the following abstract. Keep it concise and highlight the key points.'"

[0689] "Emotion recognition: 'User appears to be frustrated with the current search results. Suggest alternative keywords or improve the relevance of the displayed results.'"

[0690] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0691] Step 1:

[0692] The server collects academic papers in a specified field from academic databases using APIs or scraping technology. The input of this step is "academic papers in the specified field," and the output is "collected academic papers." Specifically, the server sends queries to academic databases such as arXiv and downloads relevant papers in HTML format.

[0693] Step 2:

[0694] The server uses natural language processing (NLP) technology to clean the collected academic data and generate summaries. The input for this step is the collected academic papers, and the output is a summarized academic paper. Specifically, the server uses Python's Transformers library to run a generative AI model to summarize the paper text.

[0695] Step 3:

[0696] The server automatically detects technical terms that appear in the summarized academic data and provides explanations. The input for this step is the "summarized academic paper," and the output is the "summary with explanations of the technical terms." Specifically, the server uses NLP technology to extract technical terms and obtain explanations from a pre-prepared technical terminology dictionary database.

[0697] Step 4:

[0698] The server stores the generated abstract and commentary in a database. The input to this step is the "abstract with terminology commentary" and the output is the "abstract and commentary stored in the database." The server associates this data with metadata and saves it in the database.

[0699] Step 5:

[0700] The user inputs keywords into the terminal, and the server searches the database based on the keywords and returns relevant summary information. The input for this step is the "keyword entered by the user," and the output is the "related summary information." Specifically, the user inputs keywords on a smartphone or PC, and the server queries the database using the keywords and sends the results to the terminal.

[0701] Step 6:

[0702] The server analyzes the user's facial expressions in real time using OpenCV to recognize the user's emotional state. The input of this step is "user's facial expression data" and the output is "emotion recognition results." Specifically, a webcam connected to the device captures the user's face, and the server receives and analyzes the video data.

[0703] Step 7:

[0704] The device optimizes search results and the user interface based on the results of the emotion recognition engine. The inputs to this step are the "emotion recognition results" and "search results," and the output is "optimized search results and user interface." Specifically, if the user expresses dissatisfaction with the search results, the server reanalyzes them and presents more relevant results.

[0705] Step 8:

[0706] Users input their company's research needs and issues through their terminals, and universities input the requirements for collaborative research. The input for this step is the "company's research needs and university research requirements," and the output is the "input research needs and requirements data." Users input the data into their respective terminals, and the server receives the data.

[0707] Step 9:

[0708] The server analyzes the input data of companies and universities and matches compatible pairs. The inputs of this step are "company research needs" and "university research requirements," and the output is "matching results." The server calculates the degree of compatibility and associates the research needs and requirements of companies and universities.

[0709] Step 10:

[0710] The server notifies the matching results. The input of this step is the "matching result" and the output is "notification to companies and universities." The server sends emails and pop-up notifications to companies and universities that have been successfully matched.

[0711] Step 11:

[0712] The server uses a high-power GPU to accept requests for computational resources and execute large-volume data processing or simulations. The input to this step is a "request for computational resources" and the output is the "computational results." The server schedules the high-power GPU based on the user's request to execute data processing or simulations.

[0713] Step 12:

[0714] The server returns the result to the user's terminal. The input of this step is the "computation result" and the output is the "computation result returned to the user's terminal." The server formats the computation result and sends it to the terminal in a format that the user can analyze.

[0715] 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.

[0716] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0717] 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.

[0718] [Second embodiment]

[0719] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0720] 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.

[0721] 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).

[0722] 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.

[0723] 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.

[0724] 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).

[0725] 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.

[0726] 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.

[0727] 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.

[0728] 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.

[0729] 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.

[0730] 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."

[0731] This invention is a system for efficiently utilizing academic research resources in rural areas and promoting collaborative research between universities and companies. The system utilizes a high-power GPU data center to support large-scale data processing and simulation. It also provides academic paper collection, summarization, terminology explanations, and search functions, facilitating the application of research results.

[0732] Collecting and summarizing academic papers

[0733] Program operation description:

[0734] Server: Automatically collects academic papers in a specified field from each academic database. For example, to obtain the latest papers on machine learning, a query is sent to the database using APIs or scraping technology and the relevant papers are downloaded.

[0735] Server: Summarizes the collected papers using natural language processing (NLP) techniques. Specifically, after text cleaning, a summary generation algorithm is applied to extract key points.

[0736] Server: Automatically detects technical terms contained in the summary and provides explanations. Refers to a pre-prepared technical terminology dictionary database and adds appropriate explanations to the summary text.

[0737] Examples:

[0738] Server: Collects 100 of the latest papers in the field of machine learning, summarizes their contents, and extracts key points. When the technical term "backpropagation" appears, an explanation of this term is added and included in the summary.

[0739] Providing a search system

[0740] Program operation description:

[0741] User: Enter search keywords from the terminal to search for abstracts of academic papers.

[0742] Server: Based on the entered keywords, the database is searched and relevant summary information is obtained.

[0743] On the device: Display search results to the user, including links to terminology explanations and original papers.

[0744] Examples:

[0745] User: Enter the keywords "environmentally friendly plastic manufacturing techniques."

[0746] Server: Retrieves abstracts of relevant papers from the database and sends them to the user's device.

[0747] On the device: The search results show the user 20 abstracts, each containing a link to an explanation of the term "biodegradable plastic."

[0748] Intermediary for joint research

[0749] Program operation description:

[0750] User (company representative): Enters the company's research needs and issues from the terminal.

[0751] User (university professor): Enters the requirements for collaborative research from a terminal.

[0752] Server: Analyzes input data from companies and universities and makes appropriate matches.

[0753] Server: Notifies companies and universities of the matching results.

[0754] Examples:

[0755] User (company representative): Enter the topic "Research on optimizing the manufacturing process of biodegradable plastics."

[0756] User (university professor): Enters the requirement, "I would like to conduct joint research on materials development."

[0757] Server: Analyzes the data from both parties and matches compatible pairs. Upon notification, both parties contact each other and start collaboration.

[0758] Utilizing GPU data centers

[0759] Program operation description:

[0760] Users (researchers, company personnel): Input requests for computing resources from their terminals.

[0761] Server: Receives requests and schedules computing resources.

[0762] Server: Executes computational tasks in a high-powered GPU data center and returns the results to the user.

[0763] Examples:

[0764] User (company representative): Requests the necessary computing resources to conduct a simulation of the plastics manufacturing process.

[0765] Server: Assigns requests to a powerful GPU and starts the simulation.

[0766] Server: Acquires the simulation results and sends them to the user's device.

[0767] Terminal: Receives the resulting data and analyzes it using visualization tools.

[0768] This will strengthen cooperative relationships between local companies and universities, and by introducing the latest research results to local areas, it will provide a system that will revitalize the local economy and achieve sustainable growth.

[0769] The processing flow will be explained below.

[0770] Collecting and summarizing academic papers

[0771] Step 1:

[0772] Server: Sends queries to the API endpoints of each academic database, including subject areas and publication date ranges.

[0773] Example: To retrieve the latest papers in the field of machine learning, send a specified query to the API of "Academic Database A."

[0774] Step 2:

[0775] Server: Receives article data returned from academic databases and stores metadata (title, author, publication date, etc.) and the text in local storage.

[0776] Example: Parse the paper data returned from the API and save the title "Machine Learning Algorithms", the author "John Doe", and the body of the paper.

[0777] Step 3:

[0778] Server: Performs text cleaning on the received paper text, removing unnecessary parts (headers, footers, tables, etc.).

[0779] Example: Analyzing the text of a paper using regular expressions to remove unnecessary strings and structures.

[0780] Step 4:

[0781] Server: Apply natural language processing (NLP) algorithms to the cleaned-up paper text to generate summaries.

[0782] Example: Using a text summarization algorithm (e.g., Textrank) to extract the main points of an article.

[0783] Step 5:

[0784] Server: Automatically detects technical terms contained in the abstract and retrieves explanations by referencing a pre-prepared technical terminology dictionary database.

[0785] Example: Detect "backpropagation" from a summary and add its explanation "Learning algorithm in neural networks."

[0786] Step 6:

[0787] Server: Stores abstracts and their terminology in a database, and associates them with metadata.

[0788] Example: Save the abstract "This paper demonstrates the efficiency of an algorithm using backpropagation" and its explanation in a database.

[0789] Providing a search system

[0790] Step 1:

[0791] User: Enters search keywords into the device's search interface. The input is sent to the server.

[0792] Example: A user types in "environmentally friendly plastic manufacturing techniques" and presses the search button.

[0793] Step 2:

[0794] Server: Searches for abstracts in the database based on received keywords. Sorts the results based on relevance.

[0795] Example: Obtain multiple summaries that match the search keywords and sort them by relevance.

[0796] Step 3:

[0797] Server: Structures search results and sends them to the user's device, including abstracts, glossary links, and links to original articles.

[0798] Example: Send a summary and its link information to the user's device in JSON format.

[0799] Step 4:

[0800] Terminal: Displays the received search results. When the user clicks on a result, a detailed summary and glossary of terms is displayed.

[0801] Example: Search results are displayed in a list of 20 abstracts, and when the user clicks on the link for "biodegradable plastic," a pop-up explanation appears.

[0802] Intermediary for joint research

[0803] Step 1:

[0804] User (company representative): Enters the company's research needs and issues from the terminal and sends them to the server.

[0805] Example: A company enters "Research on optimizing the manufacturing process of biodegradable plastics."

[0806] Step 2:

[0807] User (university professor): Enters university research requirements from the terminal and sends them to the server.

[0808] Example: A university professor enters, "I would like to conduct joint research on materials development."

[0809] Step 3:

[0810] Server: Analyzes input data from companies and universities, calculates the degree of compatibility between research needs and requirements, and matches compatible pairs.

[0811] Example: Applying machine learning algorithms to analyze input data and calculate a matching score.

[0812] Step 4:

[0813] Server: Notifies companies and universities of match results, including contact information and next step instructions.

[0814] Example: Notify the company and university of the match results and provide contact information for both parties.

[0815] Utilizing GPU data centers

[0816] Step 1:

[0817] Users (researchers, company personnel): Enter requests for computing resources from their terminals and send them to the server.

[0818] Example: Requesting specific computational resources (number of GPUs, time, etc.) for a simulation of a plastics manufacturing process.

[0819] Step 2:

[0820] Server: Analyzes received requests and schedules them by allocating available computing resources.

[0821] Example: Check the resource status of the GPU data center and allocate an appropriate GPU.

[0822] Step 3:

[0823] Server: Executes computational tasks and initiates data processing and simulations using high-powered GPUs.

[0824] Example: Loading required datasets and launching scripts to perform computational tasks.

[0825] Step 4:

[0826] Server: Receives the calculation results and sends them to the user's device. The results are formatted in a format that can be analyzed.

[0827] Example: Convert simulation results into JSON or CSV format and send them to the user's device.

[0828] Step 5:

[0829] Terminal: Displays the received calculation results and helps the user analyze them.

[0830] Example: Visualizing calculation results in graphs and charts and presenting them to the user.

[0831] By using the above specific processing steps, the system of the present invention can realize efficient utilization of academic research, promotion of collaborative research, and large-scale data processing.

[0832] Example 1

[0833] 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."

[0834] The lack of effective methods for utilizing academic research resources in rural areas, the lack of promotion of collaborative research between universities and companies, and the lack of efficient processing of large amounts of data and simulation environments are issues that need to be addressed. Furthermore, there is a need to improve the situation where academic literature collection, summaries, explanations of technical terms, and effective search functions are not adequately provided.

[0835] 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.

[0836] In this invention, the server includes means for automatically collecting academic literature from an academic information repository, means for summarizing the collected academic literature using natural language processing technology, means for automatically detecting and defining technical terms that appear in the summarized academic literature, means for storing the summaries and definitions in the information repository, means for searching the information repository based on keywords and providing related summary information, means for displaying related information extracted from the information repository on a user terminal, means for performing large-volume data processing and simulations using a high-performance computing unit, and means for accepting and scheduling requests for computing resources, thereby enabling the effective collection and use of academic information, the promotion of collaborative research, and the efficient processing of large volumes of data.

[0837] An "academic information repository" is a database for storing, organizing, and accessing academic literature and research data.

[0838] "Academic literature" refers to documents such as papers, articles, and reports published as the results of academic research.

[0839] "Automatic collection means" refers to technology that automatically retrieves relevant literature from academic information repositories based on specific conditions or keywords.

[0840] "Natural language processing technology" is a computational technology for understanding and generating human language, and is a method for summarizing, translating, and analyzing documents.

[0841] "Summarizing" is the technique of summarizing the main points and conclusions of academic literature in a short form.

[0842] "Jargon" refers to specialized words and phrases used in a particular field or discipline.

[0843] "Means for automatic detection and definition assignment" refers to technology that identifies technical terms in text and automatically adds their meanings and explanations.

[0844] "Means for storing information in a repository" refers to the technology for storing collected literature, its summaries, and definitions in a database.

[0845] "Means for searching information repositories based on keywords" refers to a technology for searching documents and abstracts in information repositories based on keywords entered by the user.

[0846] The "means for providing related summary information" is a technique for providing the user with a summary obtained as a search result.

[0847] "Means for displaying on the user's terminal" refers to technology that displays search results, summary information, definitions of technical terms, etc. on the user's device.

[0848] A "high-performance computing unit" is high-performance hardware for efficiently processing large amounts of data and performing complex calculations.

[0849] "Means for accepting requests for computing resources and scheduling" refers to a technique in which users request the computing resources they need and determine the execution order of computing tasks based on that request.

[0850] This invention is a system for efficiently utilizing academic research resources in local areas and promoting collaborative research between academic institutions and companies. The system utilizes high-performance computing units to support large-scale data processing and simulation. It also provides academic literature collection, summarization, glossary, and search functions, facilitating the application of research results.

[0851] The main components of the system include an academic information repository, a server, terminals, and a high-performance computing unit. The specific processes and technologies used for each component are described below.

[0852] Collecting and summarizing academic literature

[0853] The server does the following:

[0854] Automatically collect academic literature in a specified field from each academic information repository. For example, by sending queries to the database using APIs or scraping technology and downloading the relevant literature. For example, by using a Python library to automate the collection process.

[0855] The collected literature is summarized using natural language processing techniques. Specifically, text cleaning is performed using Python libraries such as NLTK and Spacy, and summaries are generated by applying generative AI models such as BERT and GPT-3.

[0856] Automatically detects technical terms contained in the summary and assigns definitions to them. Refers to a pre-prepared technical terminology dictionary database and adds appropriate definitions to the summary.

[0857] Examples:

[0858] The server collects 100 of the latest articles in the field of machine learning, summarizes their contents, and extracts key points. When the technical term "backpropagation" appears, it adds a definition of the term and includes it in the summary.

[0859] Example prompt sentence:

[0860] "Collect the latest papers in the field of machine learning and generate summaries, including an explanation of the technical term 'backpropagation'."

[0861] Providing a search system

[0862] The server and the device perform the following operations:

[0863] Users enter search keywords into their terminals to search for abstracts of academic literature.

[0864] The server searches the information repository based on the input keywords and retrieves relevant summary information. The search process is performed using a database such as Elasticsearch or MySQL.

[0865] The terminal displays the search results to the user, including links to explanations of technical terms and original literature.

[0866] Examples:

[0867] The user inputs the keywords "environmentally friendly plastic manufacturing technology."

[0868] The server retrieves abstracts of relevant documents from the information repository and sends them to the user's terminal.

[0869] The device displays 20 search results and summaries to the user, each of which includes a link to an explanation of the technical term "biodegradable plastic."

[0870] Example prompt sentence:

[0871] "Displays abstracts of research papers for the search keyword 'environmentally friendly plastic manufacturing technology,' and includes links to explanations of the technical term 'biodegradable plastic.'"

[0872] Intermediary for joint research

[0873] The server does the following:

[0874] Users (company personnel) input their company's research needs and issues from the terminal.

[0875] Users (academic institutions) input the requirements for collaborative research from their terminals.

[0876] The server analyzes input information from companies and academic institutions to make appropriate matches, using machine learning algorithms to compare research needs and skill sets.

[0877] The server notifies the companies and academic institutions of the matching results.

[0878] Examples:

[0879] The user (company representative) inputs the topic of "Research on optimizing the manufacturing process of biodegradable plastics."

[0880] The user (academic institution) enters the requirement that "I would like to conduct joint research on materials development."

[0881] The server analyzes the information of both parties and matches suitable pairs. Upon notification, both parties can contact each other and begin collaborative research.

[0882] Example prompt sentence:

[0883] "Matching the corporate research needs of 'Optimizing the manufacturing process of biodegradable plastics' with the academic institution's requirements of 'Joint research on materials development'"

[0884] Utilizing high-performance computing units

[0885] The server and the device perform the following operations:

[0886] Users (researchers, company personnel) input requests for computing resources from their terminals.

[0887] The server receives requests and schedules computing resources, using Slurm or Kubernetes to efficiently manage computing tasks.

[0888] The computational task is performed on a high-performance computing unit (e.g., a high-performance GPU cluster) and the results are returned to the user.

[0889] The terminal receives the resulting data and analyzes it using a visualization tool (e.g., Tableau or Matplotlib).

[0890] Examples:

[0891] A user (company representative) requests the necessary computing resources to perform a simulation of a plastic manufacturing process.

[0892] The server assigns the request to a high-performance GPU and starts the simulation.

[0893] The server obtains the simulation results and sends them to the user's terminal.

[0894] The terminal receives the resulting data and analyzes it using visualization tools.

[0895] Example prompt sentence:

[0896] "To simulate the plastics manufacturing process, the necessary computational resources are requested from a high-performance computing unit, and the results are analyzed using visualization tools."

[0897] In this way, local businesses and academic institutions can strengthen their collaborative relationships, introduce the latest research results to local areas, and provide a system for revitalizing the local economy and achieving sustainable growth.

[0898] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0899] Collecting and summarizing academic literature

[0900] Step 1: Collecting academic literature

[0901] Server: Automatically collects academic literature in specified fields from each academic information repository.

[0902] Input: Access information for academic repository and search query

[0903] What it does: It sends a query to the API of Elsevier or arXiv to get a list of relevant documents, from which the server downloads the documents in PDF or text format.

[0904] Output: Collected academic literature

[0905] Step 2: Generate summaries of academic literature

[0906] Server: Summarizes the collected literature using natural language processing technology.

[0907] Input: Collected academic literature

[0908] How it works: It uses Python's NLTK and Spacy to preprocess text data (cleaning, tokenization, stop word removal), and then generates summaries using generative AI models such as BERT and GPT-3.

[0909] Output: Summary

[0910] Step 3: Adding explanations of technical terms

[0911] Server: Automatically detects technical terms contained in the abstract and assigns definitions.

[0912] Input: Abstract, technical dictionary database

[0913] Specific operation: Analyze the summary sentence, refer to the terminology dictionary database, and add definitions corresponding to the detected terminology to the summary sentence.

[0914] Output: Summary with definition

[0915] Providing a search system

[0916] Step 1: Enter search keywords

[0917] User: Enter search keywords from the terminal and search for abstracts of academic literature.

[0918] Input: Search keyword

[0919] Specific operation: The user enters a keyword into the device's search bar and presses the "Search" button.

[0920] Output: The search keywords are sent to the server.

[0921] Step 2: Database search

[0922] Server: Based on the entered keywords, the information repository is searched and relevant summary information is obtained.

[0923] Input: Search keyword

[0924] How it works: The server uses databases such as Elasticsearch or MySQL to perform full-text searches for keywords, and prioritizes retrieving relevant summaries.

[0925] Output: List of relevant summary sentences

[0926] Step 3: Viewing search results

[0927] Terminal: Display search results to the user.

[0928] Input: List of relevant abstract sentences

[0929] Specific operation: The terminal displays a list of abstracts, allowing the user to check their contents. The abstracts also include links to explanations of technical terms and original documents.

[0930] Output: Display of search results

[0931] Intermediary for joint research

[0932] Step 1: Enter your research needs and requirements

[0933] User (company representative): Enters the company's research needs and issues from the terminal.

[0934] Input: Details of your research needs or problems

[0935] Specific operation: The company representative enters the issue into the input form and presses the submit button.

[0936] Output: The input research needs are sent to the server.

[0937] User (academic institution): Enters the requirements for collaborative research from a terminal.

[0938] Input: Details of collaboration requirements

[0939] Specific operations: The person in charge at the academic institution enters the requirements for the collaborative research into the input form and presses the submit button.

[0940] Output: The entered collaborative research requirements are sent to the server.

[0941] Step 2: Analyze the input data

[0942] Server: Analyzes the information entered by companies and academic institutions and makes appropriate matches.

[0943] Input: Corporate research needs, academic institutions' collaborative research requirements

[0944] What it does: It uses machine learning algorithms to compare research needs with available skill sets and select the best match.

[0945] Output: Matching results

[0946] Step 3: Notification of match results

[0947] Server: Notifies companies and academic institutions of the matching results.

[0948] Input: Matching results

[0949] Specific actions: Companies and academic institutions will be notified of the matching results via email and in-system notifications.

[0950] Output: Matching notification

[0951] Utilizing high-performance computing units

[0952] Step 1: Submit a request for computing resources

[0953] Users (researchers, company personnel): Input requests for computing resources from their terminals.

[0954] Input: Details of computing resources (number of GPUs, amount of memory, etc.)

[0955] Specific operation: A researcher or company representative inputs the required computing resources and presses the request submit button.

[0956] Output: A request for computational resources is sent to the server.

[0957] Step 2: Scheduling resources

[0958] Server: Receives requests and schedules computing resources.

[0959] Input: Computational resource request

[0960] How it works: Servers use resource management tools such as Slurm and Kubernetes to efficiently allocate and schedule computational tasks.

[0961] Output: Scheduling information

[0962] Step 3: Execute the computational task

[0963] Server: Executes computational tasks on high performance computing units and produces results.

[0964] Input: Scheduling information

[0965] Specific operation: The computation task is executed on a high-performance computing unit (e.g., a high-performance GPU cluster) and the computation is monitored until completion.

[0966] Output: calculation result

[0967] Step 4: Return and display results

[0968] Server: Returns the calculation results to the user.

[0969] Input: Calculation result

[0970] Specific operation: Send the calculation result to the user's device.

[0971] Output: The calculation result is sent to the terminal.

[0972] Terminal: Receives the resulting data and analyzes it using visualization tools.

[0973] Input: Submitted calculation result

[0974] Specific operation: Download the result data and perform data analysis and visualization using visualization tools such as Tableau or Matplotlib.

[0975] Output: Visualized result data

[0976] The above steps will concretely implement the collection of academic literature, the generation of summaries, the provision of a search system, the mediation of collaborative research, and the effective utilization of high-performance computing units.

[0977] (Application example 1)

[0978] 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."

[0979] Modern manufacturing requires the rapid adoption of new technologies and research results, and the optimization of manufacturing processes. However, there has been insufficient promotion of efficient use of academic research resources and collaborative research between universities and companies in regional areas. As a result, regional manufacturing companies are unable to quickly utilize the latest research results, which could lead to a decline in their competitiveness. Another issue is the difficulty of securing the high-power computing resources required for manufacturing process simulations and data calculations.

[0980] 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.

[0981] In this invention, the server includes means for automatically collecting academic papers from academic databases, means for summarizing the collected academic papers using natural language processing technology, means for automatically detecting technical terms appearing in the summarized academic papers and providing explanations, means for storing the summaries and explanations in a database, means for searching the database based on keywords and providing related summary information, means for displaying search results on a user interface, means for utilizing a group of high-performance computers to support simulations and data calculations, means for requesting computing resources, scheduling and executing tasks, and means for returning calculation results to the user. This enables the utilization of the latest academic papers, the improvement of manufacturers' technologies, and the efficient use of high-performance computing resources.

[0982] An "academic database" is a database that collects and stores academic papers and research results, and allows them to be searched and viewed.

[0983] An "academic paper" is a written document that summarizes the results of academic research and is published in academic journals, conference proceedings, etc.

[0984] "Natural language processing technology" is a technology that allows computers to understand human language and analyze and generate it.

[0985] A "summary" is a short, concise summary of the most important parts of an original text.

[0986] "Terminology" refers to terms or words used professionally in a particular field.

[0987] "Explanation" means explaining technical terms and complex content in an easy-to-understand manner.

[0988] A "database" is a computer device that systematically organizes and stores large amounts of information, making it easy to search and extract.

[0989] "Keywords" are typical words used when searching and classifying data.

[0990] A "user interface" is a screen or operating means for interaction between a user and a system.

[0991] "Simulation" is a method of virtually reproducing real-world events and processes for analysis and prediction.

[0992] A "high-power computing group" is a group of multiple high-performance computing devices that can simultaneously execute large-scale computational processes.

[0993] "Computational resources" refers to the hardware and software required to execute computational processes.

[0994] "Task scheduling" refers to efficiently managing the use of computing resources and determining the execution order and allocation of each task.

[0995] The purpose of this invention is to provide a system that promotes the efficient use of academic research resources and collaborative research between universities and companies in order to contribute to industrial applications. This system includes academic paper collection, abstract generation, terminology explanation, search functionality, and the use of a group of powerful computers. The specific processing and details of each part of the system are explained below.

[0996] The system consists of a server and client terminals, and the server has the following main functions:

[0997] 1. Collection of academic papers

[0998] The server automatically collects academic papers from academic databases using API calls and web scraping techniques. Specifically, it periodically collects papers in a specified field and stores them in a database.

[0999] 2. Abstract generation for papers

[1000] The server then uses natural language processing (NLP) techniques to summarize the collected academic papers. This process uses generative AI models such as BERT and GPT. After cleaning the text of the papers, it extracts key points and generates concise summaries.

[1001] 3. Adding technical term explanations

[1002] The server automatically detects technical terms contained in the abstract and provides explanations by referencing a pre-prepared technical terminology dictionary database.

[1003] 4. Providing a search system

[1004] The server provides a system that allows users to search for abstracts of academic papers by entering keywords, using Elasticsearch and other search algorithms to provide fast and accurate search results.

[1005] 5. Intermediary for collaborative research

[1006] Users can input their company's research needs and issues, and universities can input their collaborative research requirements. The server analyzes this data, matches and notifies compatible pairs.

[1007] 6. Utilizing high-power computers

[1008] The server uses high-performance computing resources to process large amounts of data and perform simulations. For example, we use NVIDIA GPU Cloud. The system schedules computational tasks and returns the results to the user after execution.

[1009] Examples:

[1010] If a user enters the keyword "optimization of an automated storage system using an AI robot," the server will collect abstracts of related papers and provide search results with explanations of the technical terms "backpropagation" and "reinforcement learning." The specific prompt for the generative AI model is as follows:

[1011] Example prompt sentence:

[1012] "Collect the latest academic papers on manufacturing technology and summarize the content related to 'Automated Storage System Using AI Robots'. Also, provide explanations of technical terms such as 'backpropagation' and 'reinforcement learning'."

[1013] As described above, this system will enable the rapid introduction of new technologies in the manufacturing industry and the efficient utilization of research results.

[1014] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1015] Step 1:

[1016] The server collects academic papers from academic databases using API calls and web scraping technology. The input is query information in a specified field, and the output is the collected academic papers. This allows for efficient collection of highly relevant academic papers.

[1017] Step 2:

[1018] The server performs text cleaning on the collected academic papers. Specific examples of cleaning include removing HTML tags and unnecessary spaces. The input is the raw paper text collected in step 1, and the output is the cleaned paper text. This creates a foundation for subsequent processing to be performed accurately.

[1019] Step 3:

[1020] The server summarizes the paper using natural language processing technology. Specifically, it applies a generative AI model (e.g., BERT or GPT). The input is the paper text cleaned in step 2, and the output is a summary that extracts the important points. This allows users to grasp the gist of a lengthy paper in a short amount of time.

[1021] Step 4:

[1022] The server automatically detects technical terms that appear in the summarized paper and provides explanations. This process refers to a technical dictionary. The input is the summary generated in step 3, and the output is the summary with explanations. This makes it easier for users with insufficient technical knowledge to understand the content.

[1023] Step 5:

[1024] The server stores the summary and commentary in a database. The input is the summary with commentary obtained in step 4, and the output is the data stored in the database. This makes it easy to search and reference later.

[1025] Step 6:

[1026] Users can search for abstracts of academic papers by entering keywords from a client terminal. The input is the keywords entered by the user, and the output is a list of abstracts of the search results. This allows users to quickly obtain the information they need.

[1027] Step 7:

[1028] The server searches the database using a search algorithm (e.g., Elasticsearch). The input is the keyword entered by the user in step 6, and the output is a list of related summary sentences. This returns highly accurate search results.

[1029] Step 8:

[1030] The server displays the search results on a user interface. The input is the list of abstracts obtained in step 7, and the output is a search result screen displayed on the user's device. This allows the user to obtain information in an easy-to-view format.

[1031] Step 9:

[1032] Corporate users input their research needs and issues. The input is information about the company's research topics and needs, and the output is that information saved in a database. This provides the basic information for collaborative research.

[1033] Step 10:

[1034] University users input the requirements for collaborative research. The input is the university's research requirements and desired items, and the output is that information saved in a database. This enables matching with researchers.

[1035] Step 11:

[1036] The server analyzes the input data from companies and universities and matches suitable pairs. The input is the data obtained in steps 9 and 10, and the output is the matching results. This determines the optimal collaborative research partner.

[1037] Step 12:

[1038] The server notifies the company and the university of the matching results. The input is the matching result obtained in step 11, and the output is the state in which the notification has been sent. This allows the company and the university to communicate with each other and begin collaborative research.

[1039] Step 13:

[1040] The server uses a group of powerful computers to perform simulations and data processing. The input is the user request, and the output is the calculation result. This allows for efficient processing of large amounts of data.

[1041] Step 14:

[1042] The server schedules computing resources. The input is a request for computing resources, and the output is a scheduled state. This allows for optimal allocation of computing resources.

[1043] Step 15:

[1044] The server returns the calculation results to the user. The input is the calculation result obtained in step 13, and the output is the calculation result sent to the user's terminal. This allows the user to analyze and utilize the results.

[1045] Step 16:

[1046] The user receives the calculation results and analyzes them using a visualization tool. The input is the calculation results obtained in step 15, and the output is visualized data. This makes it easier for the user to intuitively understand the data.

[1047] 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.

[1048] This system aims to efficiently utilize academic research resources in rural areas and promote collaborative research between universities and companies. It also aims to improve the user experience by incorporating an emotion engine that recognizes user emotions. The system utilizes a high-power GPU data center to support large-scale data processing and simulation. It also provides academic paper collection, summarization, terminology explanations, and search functions, as well as an interface that takes the user's psychological state into account through the emotion engine.

[1049] Collecting and summarizing academic papers

[1050] Program operation description:

[1051] Server: Automatically collects academic papers in specified fields from each academic database. It sends queries to the database using APIs and scraping technology and downloads the relevant papers.

[1052] Server: Generates summaries of collected papers using natural language processing (NLP) techniques, performs text cleaning, and applies summary generation algorithms to extract key points.

[1053] Server: Automatically detects technical terms contained in the abstract and retrieves explanations by referencing a pre-prepared technical terminology dictionary database.

[1054] Server: Stores abstracts and their terminology in a database, and associates them with metadata.

[1055] Examples:

[1056] Server: Collects 100 of the latest papers in the field of machine learning, summarizes their contents, and extracts key points. For example, it adds an explanation of the term "backpropagation" that appears in the paper to the summary.

[1057] Providing a search system

[1058] Program operation description:

[1059] User: Enter search keywords from the terminal to search for abstracts of academic papers.

[1060] Server: Searches the database based on the entered keywords and retrieves relevant abstracts.

[1061] On the device: Display search results to the user, including links to terminology explanations and original papers.

[1062] Examples:

[1063] User: Enter the keywords "environmentally friendly plastic manufacturing techniques."

[1064] Server: Retrieves abstracts of relevant papers from the database and sends them to the user's device.

[1065] On the device: The search results show the user 20 abstracts, each containing a link to an explanation of the term "biodegradable plastic."

[1066] Intermediary for joint research

[1067] Program operation description:

[1068] User (company representative): Enters the company's research needs and issues from the terminal and sends them to the server.

[1069] User (university professor): Enters university research requirements from the terminal and sends them to the server.

[1070] Server: Analyzes input data from companies and universities, calculates the degree of compatibility between research needs and requirements, and matches compatible pairs.

[1071] Server: Notifies companies and universities of the matching results.

[1072] Examples:

[1073] User (company representative): Enter the topic "Research on optimizing the manufacturing process of biodegradable plastics."

[1074] User (university professor): Enters the requirement, "I would like to conduct joint research on materials development."

[1075] Server: Analyzes the data from both parties and matches compatible pairs. Upon notification, both parties contact each other and start collaboration.

[1076] Utilizing GPU data centers

[1077] Program operation description:

[1078] Users (researchers, company personnel): Enter requests for computing resources from their terminals and send them to the server.

[1079] Server: Analyzes received requests and schedules them by allocating available computing resources.

[1080] Server: Executes computational tasks and initiates data processing and simulations using high-powered GPUs.

[1081] Server: Receives the calculation results and sends them to the user's device. The results are formatted in a format that can be analyzed.

[1082] Terminal: Displays the received calculation results and helps the user analyze them.

[1083] Examples:

[1084] User (company representative): Requests the necessary computing resources to conduct a simulation of the plastics manufacturing process.

[1085] Server: Assigns requests to a powerful GPU and starts the simulation.

[1086] Server: Acquires the simulation results and sends them to the user's device.

[1087] Terminal: Receives the resulting data and analyzes it using visualization tools.

[1088] Combining Emotion Engines

[1089] Program operation description:

[1090] User: When searching for academic papers or entering collaborative research on the device, the emotion recognition system analyzes the user's facial expressions, voice, and text input.

[1091] Server: The emotion engine analyzes the user's emotions in real time and optimizes the user experience based on the results.

[1092] Example: Adjusting the relevance or presentation of search results when users express dissatisfaction or confusion with the results.

[1093] Terminal: Presents information optimized by the emotion engine to the user.

[1094] Examples:

[1095] User: If you express frustration while searching for "eco-friendly plastic manufacturing techniques," the emotion engine will detect that.

[1096] Server: Re-analyzes the search results and presents more relevant results.

[1097] On your device: Show new search results and give you quick access to the information you want.

[1098] This will provide a system that takes user emotions into consideration, enable the introduction of cutting-edge technology to local areas, and make efficient use of academic research, thereby realizing regional digital transformation and regional revitalization.

[1099] The processing flow will be explained below.

[1100] Collecting and summarizing academic papers

[1101] Step 1:

[1102] Server: Sends queries to the API endpoints of each academic database, including subject areas and publication date ranges.

[1103] Example: To retrieve the latest papers in the field of machine learning, send the query "machine learning AND 2023" to the API of "Academic Database A."

[1104] Step 2:

[1105] Server: Receives article data returned from academic databases and stores metadata (title, author, publication date, etc.) and the text in local storage.

[1106] Example: Analyze paper data returned from an API and save the title "Machine Learning Algorithms", author "John Doe", and text to local storage.

[1107] Step 3:

[1108] Server: Performs text cleaning on the received paper text, removing unnecessary parts (headers, footers, tables, etc.).

[1109] Example: Using regular expressions to remove unnecessary tags and special characters from the body of a paper.

[1110] Step 4:

[1111] Server: Apply natural language processing (NLP) algorithms to the cleaned-up paper text to generate summaries.

[1112] Example: Extracting key sentences using the Textrank algorithm for text summarization.

[1113] Step 5:

[1114] Server: Automatically detects technical terms contained in the abstract and retrieves explanations by referencing a pre-prepared technical terminology dictionary database.

[1115] Example: Detect "backpropagation" from a summary and add an explanation "neural network learning algorithm."

[1116] Step 6:

[1117] Server: Stores the abstracts and their terminology in a database, and associates them with metadata.

[1118] Example: Save the abstract "This paper discusses the performance of an algorithm using backpropagation" in a database.

[1119] Providing a search system

[1120] Step 1:

[1121] User: Enter search keywords into the device's search interface and press the search button.

[1122] Example: Enter "environmentally friendly plastic manufacturing techniques" and click the search button.

[1123] Step 2:

[1124] Server: Searches for abstracts in the database based on received keywords. Sorts the results based on relevance.

[1125] Example: Search the database using the keyword "environmentally friendly plastic manufacturing technology," retrieve relevant abstracts, and sort them by relevance.

[1126] Step 3:

[1127] Server: Structures the search results and sends them to the user's device, including any necessary glossary links.

[1128] Example: Send a summary and link information in JSON format.

[1129] Step 4:

[1130] Terminal: Displays the received search results. When the user clicks on a result, a detailed summary and glossary is displayed.

[1131] Example: Search results are displayed in a list of 20 abstracts, and when the user clicks on the link for "biodegradable plastic," a pop-up explanation appears.

[1132] Intermediary for joint research

[1133] Step 1:

[1134] User (company representative): Enters the company's research needs and issues into the input form on the device.

[1135] Example: A company writes "Research into optimizing the manufacturing process of biodegradable plastics."

[1136] Step 2:

[1137] User (university professor): Enter the requirements for the collaborative research into the input form on the terminal.

[1138] Example: A university professor enters, "I would like to conduct joint research on materials development."

[1139] Step 3:

[1140] Server: Analyzes input data from companies and universities, calculates the degree of compatibility between research needs and requirements, and matches compatible pairs.

[1141] Example: Machine learning algorithms analyze input data and match companies and universities with good matches.

[1142] Step 4:

[1143] Server: Notifies companies and universities of match results, including contact information and next step instructions.

[1144] Example: Notify the match results by email and provide contact information for the company and university.

[1145] Utilizing GPU data centers

[1146] Step 1:

[1147] Users (researchers, company personnel): Enter requests for computing resources using the request form on their terminal.

[1148] Example: Requesting specific computational resources (number of GPUs, time, etc.) for a simulation of a plastics manufacturing process.

[1149] Step 2:

[1150] Server: Analyzes received requests and schedules them by allocating available computing resources.

[1151] Example: Check the resource status of the GPU data center and allocate an appropriate GPU.

[1152] Step 3:

[1153] Server: Executes computational tasks and initiates data processing and simulations using high-powered GPUs.

[1154] Example: Loading required datasets and launching scripts to perform computational tasks.

[1155] Step 4:

[1156] Server: Receives the calculation results and sends them to the user's device. The results are formatted in a format that can be analyzed.

[1157] Example: Convert simulation results into JSON or CSV format and send them to the user's device.

[1158] Step 5:

[1159] Terminal: Displays the received calculation results and helps the user analyze them.

[1160] Example: Visualizing calculation results in graphs and charts and presenting them to the user.

[1161] Combining Emotion Engines

[1162] Step 1:

[1163] User: When searching for academic papers or entering collaborative research on the device, the emotion recognition system analyzes the user's facial expressions, voice, and text input.

[1164] Example: A user provides facial expression data via a webcam and also provides voice input.

[1165] Step 2:

[1166] Server: The emotion engine analyzes the user's emotions in real time and optimizes the user experience based on the results.

[1167] Example: An emotion engine detects frustration or anxiety and adaptively modifies the interface.

[1168] Step 3:

[1169] Server: Re-analyzes search results and adjusts how information is presented based on the user's sentiment.

[1170] Example: If a user expresses frustration, reevaluate the relevance of search results and present more relevant results.

[1171] Step 4:

[1172] Terminal: Presents information optimized by the emotion engine to the user.

[1173] Example: Visually highlighting new search results and related information to help users quickly access the information they need.

[1174] This will provide a system that takes user emotions into consideration, enable the introduction of cutting-edge technology to local areas, and make efficient use of academic research, thereby realizing regional digital transformation and regional revitalization.

[1175] Example 2

[1176] 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."

[1177] Modern academic research is highly specialized, making it difficult for researchers to efficiently collect, analyze, and search large amounts of academic information. It is also challenging to facilitate collaborative research between companies and educational institutions, and to efficiently manage data processing and simulations that require high-performance computing resources. Furthermore, there is a need to improve the user experience by providing an interface that takes user emotions into consideration.

[1178] 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.

[1179] In this invention, the server includes: means for automatically collecting academic papers from an academic information database; means for summarizing the collected academic papers using natural language processing technology; means for automatically detecting technical terms appearing in the summarized academic papers and providing explanations; means for storing the summaries and explanations in an information storage device; means for searching the information storage device based on search keywords and providing relevant summary information; and means for analyzing a user's emotional state using emotion recognition technology and adjusting the information provided. This allows researchers and corporate personnel to efficiently collect, summarize, and search large amounts of academic information, and further enables them to provide appropriate information taking user emotions into consideration. The system can also be used as an intermediary system to facilitate collaborative research between companies and educational institutions, and efficiently execute data processing and simulations using high-performance computing devices.

[1180] An "academic information database" is a collection of information that stores academic papers and research results.

[1181] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate human language.

[1182] An "abstract" is a concise summary of the key points extracted from an academic paper.

[1183] "Terminology" is the specific terminology used in a particular field or area.

[1184] "Explanation" refers to documents or information that provide detailed explanations of technical terms, etc.

[1185] An "information storage device" is a physical or virtual device for storing data or information.

[1186] "Search keywords" are specific words or phrases used when searching for information.

[1187] "Emotion recognition technology" is a technology that determines emotions by analyzing a user's facial expressions, voice, text, etc.

[1188] "Collaborative research" refers to research activities conducted jointly by multiple researchers or organizations.

[1189] The "intermediary system" is a system that supports matching collaborative research between companies and educational institutions.

[1190] A "high performance computing device" is a device equipped with computing resources that can process large amounts of data and perform calculations at high speed.

[1191] MODE FOR CARRYING OUT THE INVENTION

[1192] The present invention is a system for efficiently collecting, analyzing, and searching academic information and mediating collaborative research. This system is composed of the following multiple modules:

[1193] Collecting and summarizing academic papers

[1194] Server: Sends queries to academic information databases (e.g., IEEE Xplore and PubMed) using APIs and scraping technologies (e.g., BeautifulSoup, Scrapy) to automatically collect academic papers in the specified field. This process allows researchers to collect the information they need without any effort.

[1195] Server: Summarizes the collected papers using natural language processing techniques (e.g., BERT, GPT-3). First, text cleaning is performed to remove unnecessary parts, and then a summary generation algorithm is used to extract the key points of the paper.

[1196] Server: Automatically detects technical terms contained in the abstract and provides explanations by referencing a pre-prepared technical terminology dictionary database. This method makes it possible to provide information that is easy to understand even for non-expert users.

[1197] Server: Stores the generated abstracts and their descriptions in an information storage device (e.g., a database), and also associates and stores the metadata of the collected papers (author names, paper titles, publication dates, etc.).

[1198] Example: A server collects 100 of the latest papers in the field of machine learning, summarizes their contents using the BERT model, adds an explanation to the technical term "backpropagation" in the paper, and saves it in an information storage device.

[1199] Providing a search system

[1200] User: Enter keywords to search for abstracts of academic papers from a terminal.

[1201] Server: Searches the information storage device based on the entered keywords and retrieves related abstracts.

[1202] Terminal: The search results are displayed to the user, including links to terminology explanations and original papers.

[1203] Example: A user enters the search keyword "environmentally friendly plastic manufacturing technology," and the server retrieves the corresponding abstracts and sends them to the terminal. The terminal displays 20 abstracts as search results, including a link to an explanation of "biodegradable plastic."

[1204] Intermediary for joint research

[1205] User (company representative): Enters the company's research needs and issues from the terminal and sends them to the server.

[1206] User (university professor): Enters university research requirements from the terminal and sends them to the server.

[1207] Server: Analyzes input data from companies and educational institutions, calculates the degree of compatibility between research needs and requirements, matches compatible pairs, and notifies both parties.

[1208] Example: A company representative inputs "Research into optimizing the manufacturing process of biodegradable plastics," and a university professor inputs "I would like to conduct joint research on materials development." The server analyzes the data of both parties and matches and notifies compatible pairs based on common research themes.

[1209] Utilizing high-performance computing devices

[1210] Users (researchers, company personnel): Enter requests for computing resources from their terminals and send them to the server.

[1211] Server: Analyzes received requests and schedules them by allocating available computing resources.

[1212] Server: Executes computational tasks and initiates data processing and simulations using high performance computing devices (e.g., NVIDIA Tesla).

[1213] Server: Receives the calculation results and sends them to the user's device. The results are formatted in a format that can be analyzed.

[1214] Terminal: Displays the received calculation results and helps the user analyze them.

[1215] Example: A company representative requests computing resources to run a simulation of a plastics manufacturing process. The server assigns the task to a high-performance computing device, runs the simulation, and sends the results to the terminal, which displays the results using a visualization tool.

[1216] Combining Emotion Engines

[1217] User: When searching for academic papers or entering collaborative research on the device, the emotion recognition system analyzes the user's facial expressions, voice, and text input.

[1218] Server: The emotion engine analyzes user emotions in real time and optimizes the user experience based on the results. For example, if a user expresses dissatisfaction or confusion about a search result, it adjusts the relevance and presentation of the search result.

[1219] Terminal: Presents information optimized by the emotion engine to the user.

[1220] Example: If a user expresses frustration while searching for "eco-friendly plastic manufacturing techniques," the emotion engine detects this and the server re-analyzes the search results to present more relevant results. The device then displays new search results, helping the user quickly access the information they are looking for.

[1221] This will provide a system that takes user emotions into consideration, enable the introduction of cutting-edge technology to local areas, and make efficient use of academic research, thereby realizing regional digital transformation (DX) and regional revitalization.

[1222] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1223] Collecting and summarizing academic papers

[1224] Step 1: Collecting academic papers

[1225] Server: Sends queries to each academic database (e.g., IEEE Xplore or PubMed) using APIs or scraping technologies (e.g., BeautifulSoup, Scrapy) to automatically collect academic papers in the specified field.

[1226] Input: URL of academic database API or web page

[1227] Output: Collected academic paper data (text format)

[1228] Specific operation: The server sends a query about "autonomous driving technology" to the IEEE Xplore API and receives data on relevant papers.

[1229] Step 2: Text Cleaning

[1230] Server: Cleans the collected text, removing unnecessary tags and characters, and formatting it into a parseable format.

[1231] Input: Collected academic paper data

[1232] Output: Cleaned text data

[1233] Specific operation: The server removes HTML tags from the paper text and standardizes line break codes.

[1234] Step 3: Summary generation

[1235] Server: Generates a summary of the paper using natural language processing (NLP) techniques (e.g., BERT, GPT-3), extracting key points and summarizing them concisely.

[1236] Input: Cleaned text data

[1237] Output: Summary

[1238] What it does: The server uses the BERT model to generate a summary about "sensor fusion techniques for self-driving cars" from the cleaned text.

[1239] Step 4: Detect and explain terminology

[1240] Server: Automatically detects technical terms contained in the abstract and provides explanations by referencing a technical term dictionary database.

[1241] Input: Abstract, technical dictionary database

[1242] Output: Summary with glossary

[1243] Specific operation: The server detects the technical term "sensor fusion" in the summary text and retrieves its definition from the dictionary database.

[1244] Step 5: Store in the database

[1245] Server: Stores the abstracts and their terminology in an information storage device. Metadata of collected papers (author names, paper titles, publication dates, etc.) is also associated and saved.

[1246] Input: Abstract, glossary, metadata

[1247] Output: Abstract and commentary stored in a database

[1248] Specific operation: The server stores the abstract and terminology in an information storage device and correctly associates the respective metadata.

[1249] Providing a search system

[1250] Step 1: Enter search keywords

[1251] User: Enter search keywords from the device.

[1252] Input: Search keyword

[1253] Output: Search request to the server

[1254] What happens: A user types "eco-friendly plastic manufacturing techniques" into their browser's search bar.

[1255] Step 2: Database search

[1256] Server: Searches the information storage device based on the entered keywords and retrieves related abstracts.

[1257] Input: Search keyword

[1258] Output: Related summary sentence data

[1259] Specific operation: The server queries the database based on the keyword "environmentally friendly plastic manufacturing technology" and retrieves related abstracts.

[1260] Step 3: Submit and view search results

[1261] Terminal: The search results are displayed to the user, including links to terminology explanations and original papers.

[1262] Input: relevant summary sentence data

[1263] Output: Search results displayed to the user

[1264] Specific behavior: The device displays the search results in a list format, including a technical link for "biodegradable plastic."

[1265] Intermediary for joint research

[1266] Step 1: Input your research needs and problems

[1267] User (company representative): Enters the company's research needs and issues from the terminal and sends them to the server.

[1268] Input: Corporate research needs and challenges

[1269] Output: Research needs data to server

[1270] Specific actions: A company representative enters "Research on optimizing the manufacturing process of biodegradable plastics" into a web form and presses the submit button.

[1271] Step 2: Enter your university research requirements

[1272] User (university professor): Enters university research requirements from the terminal and sends them to the server.

[1273] Input: University study requirements

[1274] Output: Research requirements data to the server

[1275] Specific operation: A university professor enters "I would like to conduct joint research on materials development" into a web form and presses the submit button.

[1276] Step 3: Data analysis and goodness-of-fit calculation

[1277] Server: Analyzes input data from companies and educational institutions, calculates the degree of compatibility between research needs and requirements, and matches compatible pairs.

[1278] Input: Corporate research needs data, University research requirements data

[1279] Output: Matching result data

[1280] How it works: The server compares the company's needs with the university's requirements and calculates the degree of fit based on common research themes.

[1281] Step 4: Notification of match results

[1282] Server: Notifies companies and universities of the matching results.

[1283] Input: Matching result data

[1284] Output: Send notification

[1285] Specific operation: The server sends the matching results to the company representative and university professor using email and notification functions.

[1286] Utilizing high-performance computing devices

[1287] Step 1: Request Computational Resources

[1288] Users (researchers, company personnel): Enter requests for computing resources from their terminals and send them to the server.

[1289] Input: A request for compute resources

[1290] Output: Request data to the server

[1291] Specific operation: A user requests computational resources to perform a "simulation of a plastics manufacturing process."

[1292] Step 2: Scheduling and resource allocation

[1293] Server: Analyzes received requests and schedules them by allocating available computing resources.

[1294] Input: Request data

[1295] Output: Resource allocation data

[1296] What it does: The server makes a powerful GPU (e.g., NVIDIA Tesla) available and assigns it to the requested task.

[1297] Step 3: Execute the computational task

[1298] Server: Executes computational tasks and initiates data processing and simulations using high-performance computing devices.

[1299] Input: Resource allocation data

[1300] Output: Calculation result data

[1301] What it does: The server runs simulations on a GPU cluster and monitors the execution status.

[1302] Step 4: Obtaining and sending the calculation results

[1303] Server: Receives the calculation results and sends them to the user's device. The results are formatted in a format that can be analyzed.

[1304] Input: Calculation result data

[1305] Output: Formatted calculation result data

[1306] Specific operation: The server collects the simulation results and sends them to the user in CSV format.

[1307] Step 5: Displaying results and supporting analysis

[1308] Terminal: Displays the received calculation results and helps the user analyze them.

[1309] Input: Formatted calculation result data

[1310] Output: Visualized result data

[1311] Specific operation: The terminal imports the CSV file and visualizes the data using a tool for displaying graphs (e.g., Matplotlib).

[1312] Combining Emotion Engines

[1313] Step 1: Obtaining emotion data

[1314] User: When searching for academic papers or entering collaborative research on the device, the emotion recognition system analyzes the user's facial expressions, voice, and text input.

[1315] Input: User's facial expression data, voice data, text data

[1316] Output: Emotion data to the server

[1317] How it works: The user performs a search using the camera and microphone, and the facial expressions and voice recordings are sent to the emotion engine in real time.

[1318] Step 2: Sentiment analysis and feedback

[1319] Server: The emotion engine analyzes user emotions in real time and optimizes the user experience based on the results. Example: If a user expresses dissatisfaction or confusion with a search result, the relevance and presentation of the search result can be adjusted.

[1320] Input: Emotion data

[1321] Output: Optimized search result data

[1322] What it does: The server uses Emotion AI to detect "irritation" from facial expressions and re-runs the search algorithm to optimize the results.

[1323] Step 3: Viewing optimized information

[1324] Terminal: Presents information optimized by the emotion engine to the user.

[1325] Input: Optimized search result data

[1326] Output: Optimized display of information to the user

[1327] What it does: Your device displays reparsed search results, allowing you to quickly access the information you want.

[1328] This will provide a system that takes user emotions into consideration, enable the introduction of cutting-edge technology to local areas, and make efficient use of academic research, thereby realizing regional digital transformation (DX) and regional revitalization.

[1329] (Application example 2)

[1330] 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."

[1331] In the field of modern research and development, it is important to efficiently utilize local academic research resources and promote collaborative research between universities and companies. However, collecting and understanding related academic information is difficult, and there is a lack of efficient means to search and use it. Furthermore, there is a lack of interfaces that take the user's psychological state into consideration, and there is a need to improve the user experience. No systems have been proposed to address these issues.

[1332] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1333] In this invention, the server includes: means for automatically collecting academic papers from an academic database; means for summarizing the collected academic data using natural language processing technology; means for automatically detecting technical terms that appear in the summarized academic data and providing explanations; means for storing the summaries and explanations in a database; means for searching the database based on keywords and providing related summary information; and means including an emotion recognition engine that recognizes the user's emotional state and optimizes search results and the user interface. This enables efficient collection and utilization of academic information and an improved user experience.

[1334] An "academic database" is a collection of data containing academic papers and research materials that researchers can access to obtain information.

[1335] An "academic paper" is a document in which a researcher describes in detail the results of his or her research, and is usually published in a specialist journal or at an academic conference.

[1336] "Collection means" refers to a method or device for systematically collecting specific information, and in this system it mainly refers to APIs and scraping technology.

[1337] "Natural language processing technology" is a technology for understanding and processing human language, and is used for generating summaries and text analysis.

[1338] A "summarization means" is a method or device for summarizing lengthy information in a short and concise manner.

[1339] "Terminology" is a word with a specific meaning used in a particular technical area or field.

[1340] "Explanatory means" refers to a method or device that explains specialized information in an easy-to-understand manner.

[1341] A database is a collection of information that efficiently stores and manages data and allows necessary information to be quickly searched.

[1342] A "search means" is a method or device for searching for information in a database based on specific keywords or conditions.

[1343] "Related information" is information that is determined to be semantically related to a particular theme or keyword.

[1344] An "emotion recognition engine" is a system that analyzes and recognizes emotions from a user's facial expressions, voice, text input, etc.

[1345] "User interface optimization" refers to improving the system's display and operation methods to enhance the user experience.

[1346] This invention is a system for promoting efficient utilization of academic research resources in local areas and collaborative research between universities and companies, and also aims to improve the user experience by combining it with an emotion engine that recognizes the user's emotions.

[1347] System Configuration

[1348] The system includes the following main components:

[1349] Server: Includes a means for automatically collecting academic papers from academic databases and summarizing the collected academic data using natural language processing technology.

[1350] Server: Includes a means to automatically detect and explain technical terms that appear in summarized academic data.

[1351] Server: Contains a means for storing abstracts and descriptions in a database, searching the database based on keywords, and providing relevant abstract information.

[1352] Server: Includes an emotion recognition engine that recognizes the user's emotional state and optimizes search results and the user interface.

[1353] Terminal: Includes a means for users to input research needs and issues of companies and for universities to input requirements for collaborative research.

[1354] Server: Includes the means to analyze the input data of companies and universities and match suitable pairs.

[1355] Server: Includes a means to notify matching results.

[1356] Server: Includes means to use high-powered GPUs to process large amounts of data and perform simulations.

[1357] Hardware and software used

[1358] The following hardware and software are used to realize this system.

[1359] Hardware: General purpose personal computer, webcam, powerful GPU

[1360] Software: Python, requests, BeautifulSoup, Transformers, OpenCV, Tkinter

[1361] Specific processing

[1362] 1. The server collects academic papers in a specified field from each academic database using APIs and scraping technology. For example, it collects papers in the field of machine learning and generates summaries.

[1363] 2. The server uses natural language processing (NLP) technology to clean the collected academic data, generate summaries, and automatically detect technical terms and provide explanations.

[1364] 3. The server stores the generated abstracts and descriptions in a database, where they are associated with metadata and can be efficiently searched.

[1365] 4. The user inputs keywords through the terminal, and the server searches the database based on the keywords and returns relevant summary information. For example, it presents search results related to "environmentally friendly plastic manufacturing technology."

[1366] 5. The server analyzes the user's facial expressions in real time using OpenCV to recognize the user's emotional state. The emotion recognition engine analyzes the user's emotions and, if they indicate dissatisfaction or confusion, optimizes the search results.

[1367] 6. The server and terminals provide a means for companies to input their research needs and issues, and for universities to input their collaborative research requirements. The data is analyzed and suitable pairs are matched. Notifications are sent, enabling companies and universities to conduct collaborative research.

[1368] 7. The server uses a powerful GPU to accept requests for computing resources, perform large-scale data processing and simulations, and return the results to the user's device. For example, it provides the results of a simulation of a plastic manufacturing process.

[1369] Examples of concrete examples and prompts

[1370] Examples:

[1371] User (factory researcher): "I want to know about the latest research on manufacturing processes for new materials," and enters the search keywords "new materials manufacturing process."

[1372] System: Collects relevant recent papers from arXiv and generates summaries for each.

[1373] User: Review the generated summary and refer to the glossary of technical terms.

[1374] Emotion engine: Analyzes the user's facial expressions and re-suggests search results if they indicate dissatisfaction or confusion.

[1375] Example prompt for a generative AI model:

[1376] "Generate a paper summary: 'Please generate a summary of the following abstract. Keep it concise and highlight the key points.'"

[1377] "Emotion recognition: 'User appears to be frustrated with the current search results. Suggest alternative keywords or improve the relevance of the displayed results.'"

[1378] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1379] Step 1:

[1380] The server collects academic papers in a specified field from academic databases using APIs or scraping technology. The input of this step is "academic papers in the specified field," and the output is "collected academic papers." Specifically, the server sends queries to academic databases such as arXiv and downloads relevant papers in HTML format.

[1381] Step 2:

[1382] The server uses natural language processing (NLP) technology to clean the collected academic data and generate summaries. The input for this step is the collected academic papers, and the output is a summarized academic paper. Specifically, the server uses Python's Transformers library to run a generative AI model to summarize the paper text.

[1383] Step 3:

[1384] The server automatically detects technical terms that appear in the summarized academic data and provides explanations. The input for this step is the "summarized academic paper," and the output is the "summary with explanations of the technical terms." Specifically, the server uses NLP technology to extract technical terms and obtain explanations from a pre-prepared technical terminology dictionary database.

[1385] Step 4:

[1386] The server stores the generated abstract and commentary in a database. The input to this step is the "abstract with terminology commentary" and the output is the "abstract and commentary stored in the database." The server associates this data with metadata and saves it in the database.

[1387] Step 5:

[1388] The user inputs keywords into the terminal, and the server searches the database based on the keywords and returns relevant summary information. The input for this step is the "keyword entered by the user," and the output is the "related summary information." Specifically, the user inputs keywords on a smartphone or PC, and the server queries the database using the keywords and sends the results to the terminal.

[1389] Step 6:

[1390] The server analyzes the user's facial expressions in real time using OpenCV to recognize the user's emotional state. The input of this step is "user's facial expression data" and the output is "emotion recognition results." Specifically, a webcam connected to the device captures the user's face, and the server receives and analyzes the video data.

[1391] Step 7:

[1392] The device optimizes search results and the user interface based on the results of the emotion recognition engine. The inputs to this step are the "emotion recognition results" and "search results," and the output is "optimized search results and user interface." Specifically, if the user expresses dissatisfaction with the search results, the server reanalyzes them and presents more relevant results.

[1393] Step 8:

[1394] Users input their company's research needs and issues through their terminals, and universities input the requirements for collaborative research. The input for this step is the "company's research needs and university research requirements," and the output is the "input research needs and requirements data." Users input the data into their respective terminals, and the server receives the data.

[1395] Step 9:

[1396] The server analyzes the input data of companies and universities and matches compatible pairs. The inputs of this step are "company research needs" and "university research requirements," and the output is "matching results." The server calculates the degree of compatibility and associates the research needs and requirements of companies and universities.

[1397] Step 10:

[1398] The server notifies the matching results. The input of this step is the "matching result" and the output is "notification to companies and universities." The server sends emails and pop-up notifications to companies and universities that have been successfully matched.

[1399] Step 11:

[1400] The server uses a high-power GPU to accept requests for computational resources and execute large-volume data processing or simulations. The input to this step is a "request for computational resources" and the output is the "computational results." The server schedules the high-power GPU based on the user's request to execute data processing or simulations.

[1401] Step 12:

[1402] The server returns the result to the user's terminal. The input of this step is the "computation result" and the output is the "computation result returned to the user's terminal." The server formats the computation result and sends it to the terminal in a format that the user can analyze.

[1403] 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.

[1404] 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.

[1405] 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.

[1406] [Third embodiment]

[1407] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1408] 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.

[1409] 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).

[1410] 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.

[1411] 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.

[1412] 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).

[1413] 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.

[1414] 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.

[1415] 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.

[1416] 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.

[1417] 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.

[1418] 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."

[1419] This invention is a system for efficiently utilizing academic research resources in rural areas and promoting collaborative research between universities and companies. The system utilizes a high-power GPU data center to support large-scale data processing and simulation. It also provides academic paper collection, summarization, terminology explanations, and search functions, facilitating the application of research results.

[1420] Collecting and summarizing academic papers

[1421] Program operation description:

[1422] Server: Automatically collects academic papers in a specified field from each academic database. For example, to obtain the latest papers on machine learning, a query is sent to the database using APIs or scraping technology and the relevant papers are downloaded.

[1423] Server: Summarizes the collected papers using natural language processing (NLP) techniques. Specifically, after text cleaning, a summary generation algorithm is applied to extract key points.

[1424] Server: Automatically detects technical terms contained in the summary and provides explanations. Refers to a pre-prepared technical terminology dictionary database and adds appropriate explanations to the summary text.

[1425] Examples:

[1426] Server: Collects 100 of the latest papers in the field of machine learning, summarizes their contents, and extracts key points. When the technical term "backpropagation" appears, an explanation of this term is added and included in the summary.

[1427] Providing a search system

[1428] Program operation description:

[1429] User: Enter search keywords from the terminal to search for abstracts of academic papers.

[1430] Server: Based on the entered keywords, the database is searched and relevant summary information is obtained.

[1431] On the device: Display search results to the user, including links to terminology explanations and original papers.

[1432] Examples:

[1433] User: Enter the keywords "environmentally friendly plastic manufacturing techniques."

[1434] Server: Retrieves abstracts of relevant papers from the database and sends them to the user's device.

[1435] On the device: The search results show the user 20 abstracts, each containing a link to an explanation of the term "biodegradable plastic."

[1436] Intermediary for joint research

[1437] Program operation description:

[1438] User (company representative): Enters the company's research needs and issues from the terminal.

[1439] User (university professor): Enters the requirements for collaborative research from a terminal.

[1440] Server: Analyzes input data from companies and universities and makes appropriate matches.

[1441] Server: Notifies companies and universities of the matching results.

[1442] Examples:

[1443] User (company representative): Enter the topic "Research on optimizing the manufacturing process of biodegradable plastics."

[1444] User (university professor): Enters the requirement, "I would like to conduct joint research on materials development."

[1445] Server: Analyzes the data from both parties and matches compatible pairs. Upon notification, both parties contact each other and start collaboration.

[1446] Utilizing GPU data centers

[1447] Program operation description:

[1448] Users (researchers, company personnel): Input requests for computing resources from their terminals.

[1449] Server: Receives requests and schedules computing resources.

[1450] Server: Executes computational tasks in a high-powered GPU data center and returns the results to the user.

[1451] Examples:

[1452] User (company representative): Requests the necessary computing resources to conduct a simulation of the plastics manufacturing process.

[1453] Server: Assigns requests to a powerful GPU and starts the simulation.

[1454] Server: Acquires the simulation results and sends them to the user's device.

[1455] Terminal: Receives the resulting data and analyzes it using visualization tools.

[1456] This will strengthen cooperative relationships between local companies and universities, and by introducing the latest research results to local areas, it will provide a system that will revitalize the local economy and achieve sustainable growth.

[1457] The processing flow will be explained below.

[1458] Collecting and summarizing academic papers

[1459] Step 1:

[1460] Server: Sends queries to the API endpoints of each academic database, including subject areas and publication date ranges.

[1461] Example: To retrieve the latest papers in the field of machine learning, send a specified query to the API of "Academic Database A."

[1462] Step 2:

[1463] Server: Receives article data returned from academic databases and stores metadata (title, author, publication date, etc.) and the text in local storage.

[1464] Example: Parse the paper data returned from the API and save the title "Machine Learning Algorithms", the author "John Doe", and the body of the paper.

[1465] Step 3:

[1466] Server: Performs text cleaning on the received paper text, removing unnecessary parts (headers, footers, tables, etc.).

[1467] Example: Analyzing the text of a paper using regular expressions to remove unnecessary strings and structures.

[1468] Step 4:

[1469] Server: Apply natural language processing (NLP) algorithms to the cleaned-up paper text to generate summaries.

[1470] Example: Using a text summarization algorithm (e.g., Textrank) to extract the main points of an article.

[1471] Step 5:

[1472] Server: Automatically detects technical terms contained in the abstract and retrieves explanations by referencing a pre-prepared technical terminology dictionary database.

[1473] Example: Detect "backpropagation" from a summary and add its explanation "Learning algorithm in neural networks."

[1474] Step 6:

[1475] Server: Stores abstracts and their terminology in a database, and associates them with metadata.

[1476] Example: Save the abstract "This paper demonstrates the efficiency of an algorithm using backpropagation" and its explanation in a database.

[1477] Providing a search system

[1478] Step 1:

[1479] User: Enters search keywords into the device's search interface. The input is sent to the server.

[1480] Example: A user types in "environmentally friendly plastic manufacturing techniques" and presses the search button.

[1481] Step 2:

[1482] Server: Searches for abstracts in the database based on received keywords. Sorts the results based on relevance.

[1483] Example: Obtain multiple summaries that match the search keywords and sort them by relevance.

[1484] Step 3:

[1485] Server: Structures search results and sends them to the user's device, including abstracts, glossary links, and links to original articles.

[1486] Example: Send a summary and its link information to the user's device in JSON format.

[1487] Step 4:

[1488] Terminal: Displays the received search results. When the user clicks on a result, a detailed summary and glossary of terms is displayed.

[1489] Example: Search results are displayed in a list of 20 abstracts, and when the user clicks on the link for "biodegradable plastic," a pop-up explanation appears.

[1490] Intermediary for joint research

[1491] Step 1:

[1492] User (company representative): Enters the company's research needs and issues from the terminal and sends them to the server.

[1493] Example: A company enters "Research on optimizing the manufacturing process of biodegradable plastics."

[1494] Step 2:

[1495] User (university professor): Enters university research requirements from the terminal and sends them to the server.

[1496] Example: A university professor enters, "I would like to conduct joint research on materials development."

[1497] Step 3:

[1498] Server: Analyzes input data from companies and universities, calculates the degree of compatibility between research needs and requirements, and matches compatible pairs.

[1499] Example: Applying machine learning algorithms to analyze input data and calculate a matching score.

[1500] Step 4:

[1501] Server: Notifies companies and universities of match results, including contact information and next step instructions.

[1502] Example: Notify the company and university of the match results and provide contact information for both parties.

[1503] Utilizing GPU data centers

[1504] Step 1:

[1505] Users (researchers, company personnel): Enter requests for computing resources from their terminals and send them to the server.

[1506] Example: Requesting specific computational resources (number of GPUs, time, etc.) for a simulation of a plastics manufacturing process.

[1507] Step 2:

[1508] Server: Analyzes received requests and schedules them by allocating available computing resources.

[1509] Example: Check the resource status of the GPU data center and allocate an appropriate GPU.

[1510] Step 3:

[1511] Server: Executes computational tasks and initiates data processing and simulations using high-powered GPUs.

[1512] Example: Loading required datasets and launching scripts to perform computational tasks.

[1513] Step 4:

[1514] Server: Receives the calculation results and sends them to the user's device. The results are formatted in a format that can be analyzed.

[1515] Example: Convert simulation results into JSON or CSV format and send them to the user's device.

[1516] Step 5:

[1517] Terminal: Displays the received calculation results and helps the user analyze them.

[1518] Example: Visualizing calculation results in graphs and charts and presenting them to the user.

[1519] By using the above specific processing steps, the system of the present invention can realize efficient utilization of academic research, promotion of collaborative research, and large-scale data processing.

[1520] Example 1

[1521] 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."

[1522] The lack of effective methods for utilizing academic research resources in rural areas, the lack of promotion of collaborative research between universities and companies, and the lack of efficient processing of large amounts of data and simulation environments are issues that need to be addressed. Furthermore, there is a need to improve the situation where academic literature collection, summaries, explanations of technical terms, and effective search functions are not adequately provided.

[1523] 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.

[1524] In this invention, the server includes means for automatically collecting academic literature from an academic information repository, means for summarizing the collected academic literature using natural language processing technology, means for automatically detecting and defining technical terms that appear in the summarized academic literature, means for storing the summaries and definitions in the information repository, means for searching the information repository based on keywords and providing related summary information, means for displaying related information extracted from the information repository on a user terminal, means for performing large-volume data processing and simulations using a high-performance computing unit, and means for accepting and scheduling requests for computing resources, thereby enabling the effective collection and use of academic information, the promotion of collaborative research, and the efficient processing of large volumes of data.

[1525] An "academic information repository" is a database for storing, organizing, and accessing academic literature and research data.

[1526] "Academic literature" refers to documents such as papers, articles, and reports published as the results of academic research.

[1527] "Automatic collection means" refers to technology that automatically retrieves relevant literature from academic information repositories based on specific conditions or keywords.

[1528] "Natural language processing technology" is a computational technology for understanding and generating human language, and is a method for summarizing, translating, and analyzing documents.

[1529] "Summarizing" is the technique of summarizing the main points and conclusions of academic literature in a short form.

[1530] "Jargon" refers to specialized words and phrases used in a particular field or discipline.

[1531] "Means for automatic detection and definition assignment" refers to technology that identifies technical terms in text and automatically adds their meanings and explanations.

[1532] "Means for storing information in a repository" refers to the technology for storing collected literature, its summaries, and definitions in a database.

[1533] "Means for searching information repositories based on keywords" refers to a technology for searching documents and abstracts in information repositories based on keywords entered by the user.

[1534] The "means for providing related summary information" is a technique for providing the user with a summary obtained as a search result.

[1535] "Means for displaying on the user's terminal" refers to technology that displays search results, summary information, definitions of technical terms, etc. on the user's device.

[1536] A "high-performance computing unit" is high-performance hardware for efficiently processing large amounts of data and performing complex calculations.

[1537] "Means for accepting requests for computing resources and scheduling" refers to a technique in which users request the computing resources they need and determine the execution order of computing tasks based on that request.

[1538] This invention is a system for efficiently utilizing academic research resources in local areas and promoting collaborative research between academic institutions and companies. The system utilizes high-performance computing units to support large-scale data processing and simulation. It also provides academic literature collection, summarization, glossary, and search functions, facilitating the application of research results.

[1539] The main components of the system include an academic information repository, a server, terminals, and a high-performance computing unit. The specific processes and technologies used for each component are described below.

[1540] Collecting and summarizing academic literature

[1541] The server does the following:

[1542] Automatically collect academic literature in a specified field from each academic information repository. For example, by sending queries to the database using APIs or scraping technology and downloading the relevant literature. For example, by using a Python library to automate the collection process.

[1543] The collected literature is summarized using natural language processing techniques. Specifically, text cleaning is performed using Python libraries such as NLTK and Spacy, and summaries are generated by applying generative AI models such as BERT and GPT-3.

[1544] Automatically detects technical terms contained in the summary and assigns definitions to them. Refers to a pre-prepared technical terminology dictionary database and adds appropriate definitions to the summary.

[1545] Examples:

[1546] The server collects 100 of the latest articles in the field of machine learning, summarizes their contents, and extracts key points. When the technical term "backpropagation" appears, it adds a definition of the term and includes it in the summary.

[1547] Example prompt sentence:

[1548] "Collect the latest papers in the field of machine learning and generate summaries, including an explanation of the technical term 'backpropagation'."

[1549] Providing a search system

[1550] The server and the device perform the following operations:

[1551] Users enter search keywords into their terminals to search for abstracts of academic literature.

[1552] The server searches the information repository based on the input keywords and retrieves relevant summary information. The search process is performed using a database such as Elasticsearch or MySQL.

[1553] The terminal displays the search results to the user, including links to explanations of technical terms and original literature.

[1554] Examples:

[1555] The user inputs the keywords "environmentally friendly plastic manufacturing technology."

[1556] The server retrieves abstracts of relevant documents from the information repository and sends them to the user's terminal.

[1557] The device displays 20 search results and summaries to the user, each of which includes a link to an explanation of the technical term "biodegradable plastic."

[1558] Example prompt sentence:

[1559] "Displays abstracts of research papers for the search keyword 'environmentally friendly plastic manufacturing technology,' and includes links to explanations of the technical term 'biodegradable plastic.'"

[1560] Intermediary for joint research

[1561] The server does the following:

[1562] Users (company personnel) input their company's research needs and issues from the terminal.

[1563] Users (academic institutions) input the requirements for collaborative research from their terminals.

[1564] The server analyzes input information from companies and academic institutions to make appropriate matches, using machine learning algorithms to compare research needs and skill sets.

[1565] The server notifies the companies and academic institutions of the matching results.

[1566] Examples:

[1567] The user (company representative) inputs the topic of "Research on optimizing the manufacturing process of biodegradable plastics."

[1568] The user (academic institution) enters the requirement that "I would like to conduct joint research on materials development."

[1569] The server analyzes the information of both parties and matches suitable pairs. Upon notification, both parties can contact each other and begin collaborative research.

[1570] Example prompt sentence:

[1571] "Matching the corporate research needs of 'Optimizing the manufacturing process of biodegradable plastics' with the academic institution's requirements of 'Joint research on materials development'"

[1572] Utilizing high-performance computing units

[1573] The server and the device perform the following operations:

[1574] Users (researchers, company personnel) input requests for computing resources from their terminals.

[1575] The server receives requests and schedules computing resources, using Slurm or Kubernetes to efficiently manage computing tasks.

[1576] The computational task is performed on a high-performance computing unit (e.g., a high-performance GPU cluster) and the results are returned to the user.

[1577] The terminal receives the resulting data and analyzes it using a visualization tool (e.g., Tableau or Matplotlib).

[1578] Examples:

[1579] A user (company representative) requests the necessary computing resources to perform a simulation of a plastic manufacturing process.

[1580] The server assigns the request to a high-performance GPU and starts the simulation.

[1581] The server obtains the simulation results and sends them to the user's terminal.

[1582] The terminal receives the resulting data and analyzes it using visualization tools.

[1583] Example prompt sentence:

[1584] "To simulate the plastics manufacturing process, the necessary computational resources are requested from a high-performance computing unit, and the results are analyzed using visualization tools."

[1585] In this way, local businesses and academic institutions can strengthen their collaborative relationships, introduce the latest research results to local areas, and provide a system for revitalizing the local economy and achieving sustainable growth.

[1586] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1587] Collecting and summarizing academic literature

[1588] Step 1: Collecting academic literature

[1589] Server: Automatically collects academic literature in specified fields from each academic information repository.

[1590] Input: Access information for academic repository and search query

[1591] What it does: It sends a query to the API of Elsevier or arXiv to get a list of relevant documents, from which the server downloads the documents in PDF or text format.

[1592] Output: Collected academic literature

[1593] Step 2: Generate summaries of academic literature

[1594] Server: Summarizes the collected literature using natural language processing technology.

[1595] Input: Collected academic literature

[1596] How it works: It uses Python's NLTK and Spacy to preprocess text data (cleaning, tokenization, stop word removal), and then generates summaries using generative AI models such as BERT and GPT-3.

[1597] Output: Summary

[1598] Step 3: Adding explanations of technical terms

[1599] Server: Automatically detects technical terms contained in the abstract and assigns definitions.

[1600] Input: Abstract, technical dictionary database

[1601] Specific operation: Analyze the summary sentence, refer to the terminology dictionary database, and add definitions corresponding to the detected terminology to the summary sentence.

[1602] Output: Summary with definition

[1603] Providing a search system

[1604] Step 1: Enter search keywords

[1605] User: Enter search keywords from the terminal and search for abstracts of academic literature.

[1606] Input: Search keyword

[1607] Specific operation: The user enters a keyword into the device's search bar and presses the "Search" button.

[1608] Output: The search keywords are sent to the server.

[1609] Step 2: Database search

[1610] Server: Based on the entered keywords, the information repository is searched and relevant summary information is obtained.

[1611] Input: Search keyword

[1612] How it works: The server uses databases such as Elasticsearch or MySQL to perform full-text searches for keywords, and prioritizes retrieving relevant summaries.

[1613] Output: List of relevant summary sentences

[1614] Step 3: Viewing search results

[1615] Terminal: Display search results to the user.

[1616] Input: List of relevant abstract sentences

[1617] Specific operation: The terminal displays a list of abstracts, allowing the user to check their contents. The abstracts also include links to explanations of technical terms and original documents.

[1618] Output: Display of search results

[1619] Intermediary for joint research

[1620] Step 1: Enter your research needs and requirements

[1621] User (company representative): Enters the company's research needs and issues from the terminal.

[1622] Input: Details of your research needs or problems

[1623] Specific operation: The company representative enters the issue into the input form and presses the submit button.

[1624] Output: The input research needs are sent to the server.

[1625] User (academic institution): Enters the requirements for collaborative research from a terminal.

[1626] Input: Details of collaboration requirements

[1627] Specific operations: The person in charge at the academic institution enters the requirements for the collaborative research into the input form and presses the submit button.

[1628] Output: The entered collaborative research requirements are sent to the server.

[1629] Step 2: Analyze the input data

[1630] Server: Analyzes the information entered by companies and academic institutions and makes appropriate matches.

[1631] Input: Corporate research needs, academic institutions' collaborative research requirements

[1632] What it does: It uses machine learning algorithms to compare research needs with available skill sets and select the best match.

[1633] Output: Matching results

[1634] Step 3: Notification of match results

[1635] Server: Notifies companies and academic institutions of the matching results.

[1636] Input: Matching results

[1637] Specific actions: Companies and academic institutions will be notified of the matching results via email and in-system notifications.

[1638] Output: Matching notification

[1639] Utilizing high-performance computing units

[1640] Step 1: Submit a request for computing resources

[1641] Users (researchers, company personnel): Input requests for computing resources from their terminals.

[1642] Input: Details of computing resources (number of GPUs, amount of memory, etc.)

[1643] Specific operation: A researcher or company representative inputs the required computing resources and presses the request submit button.

[1644] Output: A request for computational resources is sent to the server.

[1645] Step 2: Scheduling resources

[1646] Server: Receives requests and schedules computing resources.

[1647] Input: Computational resource request

[1648] How it works: Servers use resource management tools such as Slurm and Kubernetes to efficiently allocate and schedule computational tasks.

[1649] Output: Scheduling information

[1650] Step 3: Execute the computational task

[1651] Server: Executes computational tasks on high performance computing units and produces results.

[1652] Input: Scheduling information

[1653] Specific operation: The computation task is executed on a high-performance computing unit (e.g., a high-performance GPU cluster) and the computation is monitored until completion.

[1654] Output: calculation result

[1655] Step 4: Return and display results

[1656] Server: Returns the calculation results to the user.

[1657] Input: Calculation result

[1658] Specific operation: Send the calculation result to the user's device.

[1659] Output: The calculation result is sent to the terminal.

[1660] Terminal: Receives the resulting data and analyzes it using visualization tools.

[1661] Input: Submitted calculation result

[1662] Specific operation: Download the result data and perform data analysis and visualization using visualization tools such as Tableau or Matplotlib.

[1663] Output: Visualized result data

[1664] The above steps will concretely implement the collection of academic literature, the generation of summaries, the provision of a search system, the mediation of collaborative research, and the effective utilization of high-performance computing units.

[1665] (Application example 1)

[1666] 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."

[1667] Modern manufacturing requires the rapid adoption of new technologies and research results, and the optimization of manufacturing processes. However, there has been insufficient promotion of efficient use of academic research resources and collaborative research between universities and companies in regional areas. As a result, regional manufacturing companies are unable to quickly utilize the latest research results, which could lead to a decline in their competitiveness. Another issue is the difficulty of securing the high-power computing resources required for manufacturing process simulations and data calculations.

[1668] 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.

[1669] In this invention, the server includes means for automatically collecting academic papers from academic databases, means for summarizing the collected academic papers using natural language processing technology, means for automatically detecting technical terms appearing in the summarized academic papers and providing explanations, means for storing the summaries and explanations in a database, means for searching the database based on keywords and providing related summary information, means for displaying search results on a user interface, means for utilizing a group of high-performance computers to support simulations and data calculations, means for requesting computing resources, scheduling and executing tasks, and means for returning calculation results to the user. This enables the utilization of the latest academic papers, the improvement of manufacturers' technologies, and the efficient use of high-performance computing resources.

[1670] An "academic database" is a database that collects and stores academic papers and research results, and allows them to be searched and viewed.

[1671] An "academic paper" is a written document that summarizes the results of academic research and is published in academic journals, conference proceedings, etc.

[1672] "Natural language processing technology" is a technology that allows computers to understand human language and analyze and generate it.

[1673] A "summary" is a short, concise summary of the most important parts of an original text.

[1674] "Terminology" refers to terms or words used professionally in a particular field.

[1675] "Explanation" means explaining technical terms and complex content in an easy-to-understand manner.

[1676] A "database" is a computer device that systematically organizes and stores large amounts of information, making it easy to search and extract.

[1677] "Keywords" are typical words used when searching and classifying data.

[1678] A "user interface" is a screen or operating means for interaction between a user and a system.

[1679] "Simulation" is a method of virtually reproducing real-world events and processes for analysis and prediction.

[1680] A "high-power computing group" is a group of multiple high-performance computing devices that can simultaneously execute large-scale computational processes.

[1681] "Computational resources" refers to the hardware and software required to execute computational processes.

[1682] "Task scheduling" refers to efficiently managing the use of computing resources and determining the execution order and allocation of each task.

[1683] The purpose of this invention is to provide a system that promotes the efficient use of academic research resources and collaborative research between universities and companies in order to contribute to industrial applications. This system includes academic paper collection, abstract generation, terminology explanation, search functionality, and the use of a group of powerful computers. The specific processing and details of each part of the system are explained below.

[1684] The system consists of a server and client terminals, and the server has the following main functions:

[1685] 1. Collection of academic papers

[1686] The server automatically collects academic papers from academic databases using API calls and web scraping techniques. Specifically, it periodically collects papers in a specified field and stores them in a database.

[1687] 2. Abstract generation for papers

[1688] The server then uses natural language processing (NLP) techniques to summarize the collected academic papers. This process uses generative AI models such as BERT and GPT. After cleaning the text of the papers, it extracts key points and generates concise summaries.

[1689] 3. Adding technical term explanations

[1690] The server automatically detects technical terms contained in the abstract and provides explanations by referencing a pre-prepared technical terminology dictionary database.

[1691] 4. Providing a search system

[1692] The server provides a system that allows users to search for abstracts of academic papers by entering keywords, using Elasticsearch and other search algorithms to provide fast and accurate search results.

[1693] 5. Intermediary for collaborative research

[1694] Users can input their company's research needs and issues, and universities can input their collaborative research requirements. The server analyzes this data, matches and notifies compatible pairs.

[1695] 6. Utilizing high-power computers

[1696] The server uses high-performance computing resources to process large amounts of data and perform simulations. For example, we use NVIDIA GPU Cloud. The system schedules computational tasks and returns the results to the user after execution.

[1697] Examples:

[1698] If a user enters the keyword "optimization of an automated storage system using an AI robot," the server will collect abstracts of related papers and provide search results with explanations of the technical terms "backpropagation" and "reinforcement learning." The specific prompt for the generative AI model is as follows:

[1699] Example prompt sentence:

[1700] "Collect the latest academic papers on manufacturing technology and summarize the content related to 'Automated Storage System Using AI Robots'. Also, provide explanations of technical terms such as 'backpropagation' and 'reinforcement learning'."

[1701] As described above, this system will enable the rapid introduction of new technologies in the manufacturing industry and the efficient utilization of research results.

[1702] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1703] Step 1:

[1704] The server collects academic papers from academic databases using API calls and web scraping technology. The input is query information in a specified field, and the output is the collected academic papers. This allows for efficient collection of highly relevant academic papers.

[1705] Step 2:

[1706] The server performs text cleaning on the collected academic papers. Specific examples of cleaning include removing HTML tags and unnecessary spaces. The input is the raw paper text collected in step 1, and the output is the cleaned paper text. This creates a foundation for subsequent processing to be performed accurately.

[1707] Step 3:

[1708] The server summarizes the paper using natural language processing technology. Specifically, it applies a generative AI model (e.g., BERT or GPT). The input is the paper text cleaned in step 2, and the output is a summary that extracts the important points. This allows users to grasp the gist of a lengthy paper in a short amount of time.

[1709] Step 4:

[1710] The server automatically detects technical terms that appear in the summarized paper and provides explanations. This process refers to a technical dictionary. The input is the summary generated in step 3, and the output is the summary with explanations. This makes it easier for users with insufficient technical knowledge to understand the content.

[1711] Step 5:

[1712] The server stores the summary and commentary in a database. The input is the summary with commentary obtained in step 4, and the output is the data stored in the database. This makes it easy to search and reference later.

[1713] Step 6:

[1714] Users can search for abstracts of academic papers by entering keywords from a client terminal. The input is the keywords entered by the user, and the output is a list of abstracts of the search results. This allows users to quickly obtain the information they need.

[1715] Step 7:

[1716] The server searches the database using a search algorithm (e.g., Elasticsearch). The input is the keyword entered by the user in step 6, and the output is a list of related summary sentences. This returns highly accurate search results.

[1717] Step 8:

[1718] The server displays the search results on a user interface. The input is the list of abstracts obtained in step 7, and the output is a search result screen displayed on the user's device. This allows the user to obtain information in an easy-to-view format.

[1719] Step 9:

[1720] Corporate users input their research needs and issues. The input is information about the company's research topics and needs, and the output is that information saved in a database. This provides the basic information for collaborative research.

[1721] Step 10:

[1722] University users input the requirements for collaborative research. The input is the university's research requirements and desired items, and the output is that information saved in a database. This enables matching with researchers.

[1723] Step 11:

[1724] The server analyzes the input data from companies and universities and matches suitable pairs. The input is the data obtained in steps 9 and 10, and the output is the matching results. This determines the optimal collaborative research partner.

[1725] Step 12:

[1726] The server notifies the company and the university of the matching results. The input is the matching result obtained in step 11, and the output is the state in which the notification has been sent. This allows the company and the university to communicate with each other and begin collaborative research.

[1727] Step 13:

[1728] The server uses a group of powerful computers to perform simulations and data processing. The input is the user request, and the output is the calculation result. This allows for efficient processing of large amounts of data.

[1729] Step 14:

[1730] The server schedules computing resources. The input is a request for computing resources, and the output is a scheduled state. This allows for optimal allocation of computing resources.

[1731] Step 15:

[1732] The server returns the calculation results to the user. The input is the calculation result obtained in step 13, and the output is the calculation result sent to the user's terminal. This allows the user to analyze and utilize the results.

[1733] Step 16:

[1734] The user receives the calculation results and analyzes them using a visualization tool. The input is the calculation results obtained in step 15, and the output is visualized data. This makes it easier for the user to intuitively understand the data.

[1735] 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.

[1736] This system aims to efficiently utilize academic research resources in rural areas and promote collaborative research between universities and companies. It also aims to improve the user experience by incorporating an emotion engine that recognizes user emotions. The system utilizes a high-power GPU data center to support large-scale data processing and simulation. It also provides academic paper collection, summarization, terminology explanations, and search functions, as well as an interface that takes the user's psychological state into account through the emotion engine.

[1737] Collecting and summarizing academic papers

[1738] Program operation description:

[1739] Server: Automatically collects academic papers in specified fields from each academic database. It sends queries to the database using APIs and scraping technology and downloads the relevant papers.

[1740] Server: Generates summaries of collected papers using natural language processing (NLP) techniques, performs text cleaning, and applies summary generation algorithms to extract key points.

[1741] Server: Automatically detects technical terms contained in the abstract and retrieves explanations by referencing a pre-prepared technical terminology dictionary database.

[1742] Server: Stores abstracts and their terminology in a database, and associates them with metadata.

[1743] Examples:

[1744] Server: Collects 100 of the latest papers in the field of machine learning, summarizes their contents, and extracts key points. For example, it adds an explanation of the term "backpropagation" that appears in the paper to the summary.

[1745] Providing a search system

[1746] Program operation description:

[1747] User: Enter search keywords from the terminal to search for abstracts of academic papers.

[1748] Server: Searches the database based on the entered keywords and retrieves relevant abstracts.

[1749] On the device: Display search results to the user, including links to terminology explanations and original papers.

[1750] Examples:

[1751] User: Enter the keywords "environmentally friendly plastic manufacturing techniques."

[1752] Server: Retrieves abstracts of relevant papers from the database and sends them to the user's device.

[1753] On the device: The search results show the user 20 abstracts, each containing a link to an explanation of the term "biodegradable plastic."

[1754] Intermediary for joint research

[1755] Program operation description:

[1756] User (company representative): Enters the company's research needs and issues from the terminal and sends them to the server.

[1757] User (university professor): Enters university research requirements from the terminal and sends them to the server.

[1758] Server: Analyzes input data from companies and universities, calculates the degree of compatibility between research needs and requirements, and matches compatible pairs.

[1759] Server: Notifies companies and universities of the matching results.

[1760] Examples:

[1761] User (company representative): Enter the topic "Research on optimizing the manufacturing process of biodegradable plastics."

[1762] User (university professor): Enters the requirement, "I would like to conduct joint research on materials development."

[1763] Server: Analyzes the data from both parties and matches compatible pairs. Upon notification, both parties contact each other and start collaboration.

[1764] Utilizing GPU data centers

[1765] Program operation description:

[1766] Users (researchers, company personnel): Enter requests for computing resources from their terminals and send them to the server.

[1767] Server: Analyzes received requests and schedules them by allocating available computing resources.

[1768] Server: Executes computational tasks and initiates data processing and simulations using high-powered GPUs.

[1769] Server: Receives the calculation results and sends them to the user's device. The results are formatted in a format that can be analyzed.

[1770] Terminal: Displays the received calculation results and helps the user analyze them.

[1771] Examples:

[1772] User (company representative): Requests the necessary computing resources to conduct a simulation of the plastics manufacturing process.

[1773] Server: Assigns requests to a powerful GPU and starts the simulation.

[1774] Server: Acquires the simulation results and sends them to the user's device.

[1775] Terminal: Receives the resulting data and analyzes it using visualization tools.

[1776] Combining Emotion Engines

[1777] Program operation description:

[1778] User: When searching for academic papers or entering collaborative research on the device, the emotion recognition system analyzes the user's facial expressions, voice, and text input.

[1779] Server: The emotion engine analyzes the user's emotions in real time and optimizes the user experience based on the results.

[1780] Example: Adjusting the relevance or presentation of search results when users express dissatisfaction or confusion with the results.

[1781] Terminal: Presents information optimized by the emotion engine to the user.

[1782] Examples:

[1783] User: If you express frustration while searching for "eco-friendly plastic manufacturing techniques," the emotion engine will detect that.

[1784] Server: Re-analyzes the search results and presents more relevant results.

[1785] On your device: Show new search results and give you quick access to the information you want.

[1786] This will provide a system that takes user emotions into consideration, enable the introduction of cutting-edge technology to local areas, and make efficient use of academic research, thereby realizing regional digital transformation and regional revitalization.

[1787] The processing flow will be explained below.

[1788] Collecting and summarizing academic papers

[1789] Step 1:

[1790] Server: Sends queries to the API endpoints of each academic database, including subject areas and publication date ranges.

[1791] Example: To retrieve the latest papers in the field of machine learning, send the query "machine learning AND 2023" to the API of "Academic Database A."

[1792] Step 2:

[1793] Server: Receives article data returned from academic databases and stores metadata (title, author, publication date, etc.) and the text in local storage.

[1794] Example: Analyze paper data returned from an API and save the title "Machine Learning Algorithms", author "John Doe", and text to local storage.

[1795] Step 3:

[1796] Server: Performs text cleaning on the received paper text, removing unnecessary parts (headers, footers, tables, etc.).

[1797] Example: Using regular expressions to remove unnecessary tags and special characters from the body of a paper.

[1798] Step 4:

[1799] Server: Apply natural language processing (NLP) algorithms to the cleaned-up paper text to generate summaries.

[1800] Example: Extracting key sentences using the Textrank algorithm for text summarization.

[1801] Step 5:

[1802] Server: Automatically detects technical terms contained in the abstract and retrieves explanations by referencing a pre-prepared technical terminology dictionary database.

[1803] Example: Detect "backpropagation" from a summary and add an explanation "neural network learning algorithm."

[1804] Step 6:

[1805] Server: Stores the abstracts and their terminology in a database, and associates them with metadata.

[1806] Example: Save the abstract "This paper discusses the performance of an algorithm using backpropagation" in a database.

[1807] Providing a search system

[1808] Step 1:

[1809] User: Enter search keywords into the device's search interface and press the search button.

[1810] Example: Enter "environmentally friendly plastic manufacturing techniques" and click the search button.

[1811] Step 2:

[1812] Server: Searches for abstracts in the database based on received keywords. Sorts the results based on relevance.

[1813] Example: Search the database using the keyword "environmentally friendly plastic manufacturing technology," retrieve relevant abstracts, and sort them by relevance.

[1814] Step 3:

[1815] Server: Structures the search results and sends them to the user's device, including any necessary glossary links.

[1816] Example: Send a summary and link information in JSON format.

[1817] Step 4:

[1818] Terminal: Displays the received search results. When the user clicks on a result, a detailed summary and glossary is displayed.

[1819] Example: Search results are displayed in a list of 20 abstracts, and when the user clicks on the link for "biodegradable plastic," a pop-up explanation appears.

[1820] Intermediary for joint research

[1821] Step 1:

[1822] User (company representative): Enters the company's research needs and issues into the input form on the device.

[1823] Example: A company writes "Research into optimizing the manufacturing process of biodegradable plastics."

[1824] Step 2:

[1825] User (university professor): Enter the requirements for the collaborative research into the input form on the terminal.

[1826] Example: A university professor enters, "I would like to conduct joint research on materials development."

[1827] Step 3:

[1828] Server: Analyzes input data from companies and universities, calculates the degree of compatibility between research needs and requirements, and matches compatible pairs.

[1829] Example: Machine learning algorithms analyze input data and match companies and universities with good matches.

[1830] Step 4:

[1831] Server: Notifies companies and universities of match results, including contact information and next step instructions.

[1832] Example: Notify the match results by email and provide contact information for the company and university.

[1833] Utilizing GPU data centers

[1834] Step 1:

[1835] Users (researchers, company personnel): Enter requests for computing resources using the request form on their terminal.

[1836] Example: Requesting specific computational resources (number of GPUs, time, etc.) for a simulation of a plastics manufacturing process.

[1837] Step 2:

[1838] Server: Analyzes received requests and schedules them by allocating available computing resources.

[1839] Example: Check the resource status of the GPU data center and allocate an appropriate GPU.

[1840] Step 3:

[1841] Server: Executes computational tasks and initiates data processing and simulations using high-powered GPUs.

[1842] Example: Loading required datasets and launching scripts to perform computational tasks.

[1843] Step 4:

[1844] Server: Receives the calculation results and sends them to the user's device. The results are formatted in a format that can be analyzed.

[1845] Example: Convert simulation results into JSON or CSV format and send them to the user's device.

[1846] Step 5:

[1847] Terminal: Displays the received calculation results and helps the user analyze them.

[1848] Example: Visualizing calculation results in graphs and charts and presenting them to the user.

[1849] Combining Emotion Engines

[1850] Step 1:

[1851] User: When searching for academic papers or entering collaborative research on the device, the emotion recognition system analyzes the user's facial expressions, voice, and text input.

[1852] Example: A user provides facial expression data via a webcam and also provides voice input.

[1853] Step 2:

[1854] Server: The emotion engine analyzes the user's emotions in real time and optimizes the user experience based on the results.

[1855] Example: An emotion engine detects frustration or anxiety and adaptively modifies the interface.

[1856] Step 3:

[1857] Server: Re-analyzes search results and adjusts how information is presented based on the user's sentiment.

[1858] Example: If a user expresses frustration, reevaluate the relevance of search results and present more relevant results.

[1859] Step 4:

[1860] Terminal: Presents information optimized by the emotion engine to the user.

[1861] Example: Visually highlighting new search results and related information to help users quickly access the information they need.

[1862] This will provide a system that takes user emotions into consideration, enable the introduction of cutting-edge technology to local areas, and make efficient use of academic research, thereby realizing regional digital transformation and regional revitalization.

[1863] Example 2

[1864] 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."

[1865] Modern academic research is highly specialized, making it difficult for researchers to efficiently collect, analyze, and search large amounts of academic information. It is also challenging to facilitate collaborative research between companies and educational institutions, and to efficiently manage data processing and simulations that require high-performance computing resources. Furthermore, there is a need to improve the user experience by providing an interface that takes user emotions into consideration.

[1866] 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.

[1867] In this invention, the server includes: means for automatically collecting academic papers from an academic information database; means for summarizing the collected academic papers using natural language processing technology; means for automatically detecting technical terms appearing in the summarized academic papers and providing explanations; means for storing the summaries and explanations in an information storage device; means for searching the information storage device based on search keywords and providing relevant summary information; and means for analyzing a user's emotional state using emotion recognition technology and adjusting the information provided. This allows researchers and corporate personnel to efficiently collect, summarize, and search large amounts of academic information, and further enables them to provide appropriate information taking user emotions into consideration. The system can also be used as an intermediary system to facilitate collaborative research between companies and educational institutions, and efficiently execute data processing and simulations using high-performance computing devices.

[1868] An "academic information database" is a collection of information that stores academic papers and research results.

[1869] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate human language.

[1870] An "abstract" is a concise summary of the key points extracted from an academic paper.

[1871] "Terminology" is the specific terminology used in a particular field or area.

[1872] "Explanation" refers to documents or information that provide detailed explanations of technical terms, etc.

[1873] An "information storage device" is a physical or virtual device for storing data or information.

[1874] "Search keywords" are specific words or phrases used when searching for information.

[1875] "Emotion recognition technology" is a technology that determines emotions by analyzing a user's facial expressions, voice, text, etc.

[1876] "Collaborative research" refers to research activities conducted jointly by multiple researchers or organizations.

[1877] The "intermediary system" is a system that supports matching collaborative research between companies and educational institutions.

[1878] A "high performance computing device" is a device equipped with computing resources that can process large amounts of data and perform calculations at high speed.

[1879] MODE FOR CARRYING OUT THE INVENTION

[1880] The present invention is a system for efficiently collecting, analyzing, and searching academic information and mediating collaborative research. This system is composed of the following multiple modules:

[1881] Collecting and summarizing academic papers

[1882] Server: Sends queries to academic information databases (e.g., IEEE Xplore and PubMed) using APIs and scraping technologies (e.g., BeautifulSoup, Scrapy) to automatically collect academic papers in the specified field. This process allows researchers to collect the information they need without any effort.

[1883] Server: Summarizes the collected papers using natural language processing techniques (e.g., BERT, GPT-3). First, text cleaning is performed to remove unnecessary parts, and then a summary generation algorithm is used to extract the key points of the paper.

[1884] Server: Automatically detects technical terms contained in the abstract and provides explanations by referencing a pre-prepared technical terminology dictionary database. This method makes it possible to provide information that is easy to understand even for non-expert users.

[1885] Server: Stores the generated abstracts and their descriptions in an information storage device (e.g., a database), and also associates and stores the metadata of the collected papers (author names, paper titles, publication dates, etc.).

[1886] Example: A server collects 100 of the latest papers in the field of machine learning, summarizes their contents using the BERT model, adds an explanation to the technical term "backpropagation" in the paper, and saves it in an information storage device.

[1887] Providing a search system

[1888] User: Enter keywords to search for abstracts of academic papers from a terminal.

[1889] Server: Searches the information storage device based on the entered keywords and retrieves related abstracts.

[1890] Terminal: The search results are displayed to the user, including links to terminology explanations and original papers.

[1891] Example: A user enters the search keyword "environmentally friendly plastic manufacturing technology," and the server retrieves the corresponding abstracts and sends them to the terminal. The terminal displays 20 abstracts as search results, including a link to an explanation of "biodegradable plastic."

[1892] Intermediary for joint research

[1893] User (company representative): Enters the company's research needs and issues from the terminal and sends them to the server.

[1894] User (university professor): Enters university research requirements from the terminal and sends them to the server.

[1895] Server: Analyzes input data from companies and educational institutions, calculates the degree of compatibility between research needs and requirements, matches compatible pairs, and notifies both parties.

[1896] Example: A company representative inputs "Research into optimizing the manufacturing process of biodegradable plastics," and a university professor inputs "I would like to conduct joint research on materials development." The server analyzes the data of both parties and matches and notifies compatible pairs based on common research themes.

[1897] Utilizing high-performance computing devices

[1898] Users (researchers, company personnel): Enter requests for computing resources from their terminals and send them to the server.

[1899] Server: Analyzes received requests and schedules them by allocating available computing resources.

[1900] Server: Executes computational tasks and initiates data processing and simulations using high performance computing devices (e.g., NVIDIA Tesla).

[1901] Server: Receives the calculation results and sends them to the user's device. The results are formatted in a format that can be analyzed.

[1902] Terminal: Displays the received calculation results and helps the user analyze them.

[1903] Example: A company representative requests computing resources to run a simulation of a plastics manufacturing process. The server assigns the task to a high-performance computing device, runs the simulation, and sends the results to the terminal, which displays the results using a visualization tool.

[1904] Combining Emotion Engines

[1905] User: When searching for academic papers or entering collaborative research on the device, the emotion recognition system analyzes the user's facial expressions, voice, and text input.

[1906] Server: The emotion engine analyzes user emotions in real time and optimizes the user experience based on the results. For example, if a user expresses dissatisfaction or confusion about a search result, it adjusts the relevance and presentation of the search result.

[1907] Terminal: Presents information optimized by the emotion engine to the user.

[1908] Example: If a user expresses frustration while searching for "eco-friendly plastic manufacturing techniques," the emotion engine detects this and the server re-analyzes the search results to present more relevant results. The device then displays new search results, helping the user quickly access the information they are looking for.

[1909] This will provide a system that takes user emotions into consideration, enable the introduction of cutting-edge technology to local areas, and make efficient use of academic research, thereby realizing regional digital transformation (DX) and regional revitalization.

[1910] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1911] Collecting and summarizing academic papers

[1912] Step 1: Collecting academic papers

[1913] Server: Sends queries to each academic database (e.g., IEEE Xplore or PubMed) using APIs or scraping technologies (e.g., BeautifulSoup, Scrapy) to automatically collect academic papers in the specified field.

[1914] Input: URL of academic database API or web page

[1915] Output: Collected academic paper data (text format)

[1916] Specific operation: The server sends a query about "autonomous driving technology" to the IEEE Xplore API and receives data on relevant papers.

[1917] Step 2: Text Cleaning

[1918] Server: Cleans the collected text, removing unnecessary tags and characters, and formatting it into a parseable format.

[1919] Input: Collected academic paper data

[1920] Output: Cleaned text data

[1921] Specific operation: The server removes HTML tags from the paper text and standardizes line break codes.

[1922] Step 3: Summary generation

[1923] Server: Generates a summary of the paper using natural language processing (NLP) techniques (e.g., BERT, GPT-3), extracting key points and summarizing them concisely.

[1924] Input: Cleaned text data

[1925] Output: Summary

[1926] What it does: The server uses the BERT model to generate a summary about "sensor fusion techniques for self-driving cars" from the cleaned text.

[1927] Step 4: Detect and explain terminology

[1928] Server: Automatically detects technical terms contained in the abstract and provides explanations by referencing a technical term dictionary database.

[1929] Input: Abstract, technical dictionary database

[1930] Output: Summary with glossary

[1931] Specific operation: The server detects the technical term "sensor fusion" in the summary text and retrieves its definition from the dictionary database.

[1932] Step 5: Store in the database

[1933] Server: Stores the abstracts and their terminology in an information storage device. Metadata of collected papers (author names, paper titles, publication dates, etc.) is also associated and saved.

[1934] Input: Abstract, glossary, metadata

[1935] Output: Abstract and commentary stored in a database

[1936] Specific operation: The server stores the abstract and terminology in an information storage device and correctly associates the respective metadata.

[1937] Providing a search system

[1938] Step 1: Enter search keywords

[1939] User: Enter search keywords from the device.

[1940] Input: Search keyword

[1941] Output: Search request to the server

[1942] What happens: A user types "eco-friendly plastic manufacturing techniques" into their browser's search bar.

[1943] Step 2: Database search

[1944] Server: Searches the information storage device based on the entered keywords and retrieves related abstracts.

[1945] Input: Search keyword

[1946] Output: Related summary sentence data

[1947] Specific operation: The server queries the database based on the keyword "environmentally friendly plastic manufacturing technology" and retrieves related abstracts.

[1948] Step 3: Submit and view search results

[1949] Terminal: The search results are displayed to the user, including links to terminology explanations and original papers.

[1950] Input: relevant summary sentence data

[1951] Output: Search results displayed to the user

[1952] Specific behavior: The device displays the search results in a list format, including a technical link for "biodegradable plastic."

[1953] Intermediary for joint research

[1954] Step 1: Input your research needs and problems

[1955] User (company representative): Enters the company's research needs and issues from the terminal and sends them to the server.

[1956] Input: Corporate research needs and challenges

[1957] Output: Research needs data to server

[1958] Specific actions: A company representative enters "Research on optimizing the manufacturing process of biodegradable plastics" into a web form and presses the submit button.

[1959] Step 2: Enter your university research requirements

[1960] User (university professor): Enters university research requirements from the terminal and sends them to the server.

[1961] Input: University study requirements

[1962] Output: Research requirements data to the server

[1963] Specific operation: A university professor enters "I would like to conduct joint research on materials development" into a web form and presses the submit button.

[1964] Step 3: Data analysis and goodness-of-fit calculation

[1965] Server: Analyzes input data from companies and educational institutions, calculates the degree of compatibility between research needs and requirements, and matches compatible pairs.

[1966] Input: Corporate research needs data, University research requirements data

[1967] Output: Matching result data

[1968] How it works: The server compares the company's needs with the university's requirements and calculates the degree of fit based on common research themes.

[1969] Step 4: Notification of match results

[1970] Server: Notifies companies and universities of the matching results.

[1971] Input: Matching result data

[1972] Output: Send notification

[1973] Specific operation: The server sends the matching results to the company representative and university professor using email and notification functions.

[1974] Utilizing high-performance computing devices

[1975] Step 1: Request Computational Resources

[1976] Users (researchers, company personnel): Enter requests for computing resources from their terminals and send them to the server.

[1977] Input: A request for compute resources

[1978] Output: Request data to the server

[1979] Specific operation: A user requests computational resources to perform a "simulation of a plastics manufacturing process."

[1980] Step 2: Scheduling and resource allocation

[1981] Server: Analyzes received requests and schedules them by allocating available computing resources.

[1982] Input: Request data

[1983] Output: Resource allocation data

[1984] What it does: The server makes a powerful GPU (e.g., NVIDIA Tesla) available and assigns it to the requested task.

[1985] Step 3: Execute the computational task

[1986] Server: Executes computational tasks and initiates data processing and simulations using high-performance computing devices.

[1987] Input: Resource allocation data

[1988] Output: Calculation result data

[1989] What it does: The server runs simulations on a GPU cluster and monitors the execution status.

[1990] Step 4: Obtaining and sending the calculation results

[1991] Server: Receives the calculation results and sends them to the user's device. The results are formatted in a format that can be analyzed.

[1992] Input: Calculation result data

[1993] Output: Formatted calculation result data

[1994] Specific operation: The server collects the simulation results and sends them to the user in CSV format.

[1995] Step 5: Displaying results and supporting analysis

[1996] Terminal: Displays the received calculation results and helps the user analyze them.

[1997] Input: Formatted calculation result data

[1998] Output: Visualized result data

[1999] Specific operation: The terminal imports the CSV file and visualizes the data using a tool for displaying graphs (e.g., Matplotlib).

[2000] Combining Emotion Engines

[2001] Step 1: Obtaining emotion data

[2002] User: When searching for academic papers or entering collaborative research on the device, the emotion recognition system analyzes the user's facial expressions, voice, and text input.

[2003] Input: User's facial expression data, voice data, text data

[2004] Output: Emotion data to the server

[2005] How it works: The user performs a search using the camera and microphone, and the facial expressions and voice recordings are sent to the emotion engine in real time.

[2006] Step 2: Sentiment analysis and feedback

[2007] Server: The emotion engine analyzes user emotions in real time and optimizes the user experience based on the results. Example: If a user expresses dissatisfaction or confusion with a search result, the relevance and presentation of the search result can be adjusted.

[2008] Input: Emotion data

[2009] Output: Optimized search result data

[2010] What it does: The server uses Emotion AI to detect "irritation" from facial expressions and re-runs the search algorithm to optimize the results.

[2011] Step 3: Viewing optimized information

[2012] Terminal: Presents information optimized by the emotion engine to the user.

[2013] Input: Optimized search result data

[2014] Output: Optimized display of information to the user

[2015] What it does: Your device displays reparsed search results, allowing you to quickly access the information you want.

[2016] This will provide a system that takes user emotions into consideration, enable the introduction of cutting-edge technology to local areas, and make efficient use of academic research, thereby realizing regional digital transformation (DX) and regional revitalization.

[2017] (Application example 2)

[2018] 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."

[2019] In the field of modern research and development, it is important to efficiently utilize local academic research resources and promote collaborative research between universities and companies. However, collecting and understanding related academic information is difficult, and there is a lack of efficient means to search and use it. Furthermore, there is a lack of interfaces that take the user's psychological state into consideration, and there is a need to improve the user experience. No systems have been proposed to address these issues.

[2020] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[2021] In this invention, the server includes: means for automatically collecting academic papers from an academic database; means for summarizing the collected academic data using natural language processing technology; means for automatically detecting technical terms that appear in the summarized academic data and providing explanations; means for storing the summaries and explanations in a database; means for searching the database based on keywords and providing related summary information; and means including an emotion recognition engine that recognizes the user's emotional state and optimizes search results and the user interface. This enables efficient collection and utilization of academic information and an improved user experience.

[2022] An "academic database" is a collection of data containing academic papers and research materials that researchers can access to obtain information.

[2023] An "academic paper" is a document in which a researcher describes in detail the results of his or her research, and is usually published in a specialist journal or at an academic conference.

[2024] "Collection means" refers to a method or device for systematically collecting specific information, and in this system it mainly refers to APIs and scraping technology.

[2025] "Natural language processing technology" is a technology for understanding and processing human language, and is used for generating summaries and text analysis.

[2026] A "summarization means" is a method or device for summarizing lengthy information in a short and concise manner.

[2027] "Terminology" is a word with a specific meaning used in a particular technical area or field.

[2028] "Explanatory means" refers to a method or device that explains specialized information in an easy-to-understand manner.

[2029] A database is a collection of information that efficiently stores and manages data and allows necessary information to be quickly searched.

[2030] A "search means" is a method or device for searching for information in a database based on specific keywords or conditions.

[2031] "Related information" is information that is determined to be semantically related to a particular theme or keyword.

[2032] An "emotion recognition engine" is a system that analyzes and recognizes emotions from a user's facial expressions, voice, text input, etc.

[2033] "User interface optimization" refers to improving the system's display and operation methods to enhance the user experience.

[2034] This invention is a system for promoting efficient utilization of academic research resources in local areas and collaborative research between universities and companies, and also aims to improve the user experience by combining it with an emotion engine that recognizes the user's emotions.

[2035] System Configuration

[2036] The system includes the following main components:

[2037] Server: Includes a means for automatically collecting academic papers from academic databases and summarizing the collected academic data using natural language processing technology.

[2038] Server: Includes a means to automatically detect and explain technical terms that appear in summarized academic data.

[2039] Server: Contains a means for storing abstracts and descriptions in a database, searching the database based on keywords, and providing relevant abstract information.

[2040] Server: Includes an emotion recognition engine that recognizes the user's emotional state and optimizes search results and the user interface.

[2041] Terminal: Includes a means for users to input research needs and issues of companies and for universities to input requirements for collaborative research.

[2042] Server: Includes the means to analyze the input data of companies and universities and match suitable pairs.

[2043] Server: Includes a means to notify matching results.

[2044] Server: Includes means to use high-powered GPUs to process large amounts of data and perform simulations.

[2045] Hardware and software used

[2046] The following hardware and software are used to realize this system.

[2047] Hardware: General purpose personal computer, webcam, powerful GPU

[2048] Software: Python, requests, BeautifulSoup, Transformers, OpenCV, Tkinter

[2049] Specific processing

[2050] 1. The server collects academic papers in a specified field from each academic database using APIs and scraping technology. For example, it collects papers in the field of machine learning and generates summaries.

[2051] 2. The server uses natural language processing (NLP) technology to clean the collected academic data, generate summaries, and automatically detect technical terms and provide explanations.

[2052] 3. The server stores the generated abstracts and descriptions in a database, where they are associated with metadata and can be efficiently searched.

[2053] 4. The user inputs keywords through the terminal, and the server searches the database based on the keywords and returns relevant summary information. For example, it presents search results related to "environmentally friendly plastic manufacturing technology."

[2054] 5. The server analyzes the user's facial expressions in real time using OpenCV to recognize the user's emotional state. The emotion recognition engine analyzes the user's emotions and, if they indicate dissatisfaction or confusion, optimizes the search results.

[2055] 6. The server and terminals provide a means for companies to input their research needs and issues, and for universities to input their collaborative research requirements. The data is analyzed and suitable pairs are matched. Notifications are sent, enabling companies and universities to conduct collaborative research.

[2056] 7. The server uses a powerful GPU to accept requests for computing resources, perform large-scale data processing and simulations, and return the results to the user's device. For example, it provides the results of a simulation of a plastic manufacturing process.

[2057] Examples of concrete examples and prompts

[2058] Examples:

[2059] User (factory researcher): "I want to know about the latest research on manufacturing processes for new materials," and enters the search keywords "new materials manufacturing process."

[2060] System: Collects relevant recent papers from arXiv and generates summaries for each.

[2061] User: Review the generated summary and refer to the glossary of technical terms.

[2062] Emotion engine: Analyzes the user's facial expressions and re-suggests search results if they indicate dissatisfaction or confusion.

[2063] Example prompt for a generative AI model:

[2064] "Generate a paper summary: 'Please generate a summary of the following abstract. Keep it concise and highlight the key points.'"

[2065] "Emotion recognition: 'User appears to be frustrated with the current search results. Suggest alternative keywords or improve the relevance of the displayed results.'"

[2066] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2067] Step 1:

[2068] The server collects academic papers in a specified field from academic databases using APIs or scraping technology. The input of this step is "academic papers in the specified field," and the output is "collected academic papers." Specifically, the server sends queries to academic databases such as arXiv and downloads relevant papers in HTML format.

[2069] Step 2:

[2070] The server uses natural language processing (NLP) technology to clean the collected academic data and generate summaries. The input for this step is the collected academic papers, and the output is a summarized academic paper. Specifically, the server uses Python's Transformers library to run a generative AI model to summarize the paper text.

[2071] Step 3:

[2072] The server automatically detects technical terms that appear in the summarized academic data and provides explanations. The input for this step is the "summarized academic paper," and the output is the "summary with explanations of the technical terms." Specifically, the server uses NLP technology to extract technical terms and obtain explanations from a pre-prepared technical terminology dictionary database.

[2073] Step 4:

[2074] The server stores the generated abstract and commentary in a database. The input to this step is the "abstract with terminology commentary" and the output is the "abstract and commentary stored in the database." The server associates this data with metadata and saves it in the database.

[2075] Step 5:

[2076] The user inputs keywords into the terminal, and the server searches the database based on the keywords and returns relevant summary information. The input for this step is the "keyword entered by the user," and the output is the "related summary information." Specifically, the user inputs keywords on a smartphone or PC, and the server queries the database using the keywords and sends the results to the terminal.

[2077] Step 6:

[2078] The server analyzes the user's facial expressions in real time using OpenCV to recognize the user's emotional state. The input of this step is "user's facial expression data" and the output is "emotion recognition results." Specifically, a webcam connected to the device captures the user's face, and the server receives and analyzes the video data.

[2079] Step 7:

[2080] The device optimizes search results and the user interface based on the results of the emotion recognition engine. The inputs to this step are the "emotion recognition results" and "search results," and the output is "optimized search results and user interface." Specifically, if the user expresses dissatisfaction with the search results, the server reanalyzes them and presents more relevant results.

[2081] Step 8:

[2082] Users input their company's research needs and issues through their terminals, and universities input the requirements for collaborative research. The input for this step is the "company's research needs and university research requirements," and the output is the "input research needs and requirements data." Users input the data into their respective terminals, and the server receives the data.

[2083] Step 9:

[2084] The server analyzes the input data of companies and universities and matches compatible pairs. The inputs of this step are "company research needs" and "university research requirements," and the output is "matching results." The server calculates the degree of compatibility and associates the research needs and requirements of companies and universities.

[2085] Step 10:

[2086] The server notifies the matching results. The input of this step is the "matching result" and the output is "notification to companies and universities." The server sends emails and pop-up notifications to companies and universities that have been successfully matched.

[2087] Step 11:

[2088] The server uses a high-power GPU to accept requests for computational resources and execute large-volume data processing or simulations. The input to this step is a "request for computational resources" and the output is the "computational results." The server schedules the high-power GPU based on the user's request to execute data processing or simulations.

[2089] Step 12:

[2090] The server returns the result to the user's terminal. The input of this step is the "computation result" and the output is the "computation result returned to the user's terminal." The server formats the computation result and sends it to the terminal in a format that the user can analyze.

[2091] 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.

[2092] 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.

[2093] 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.

[2094] [Fourth embodiment]

[2095] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[2096] 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.

[2097] 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).

[2098] 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.

[2099] 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.

[2100] 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).

[2101] 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.

[2102] 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.

[2103] 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.

[2104] 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.

[2105] 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.

[2106] 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.

[2107] 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."

[2108] This invention is a system for efficiently utilizing academic research resources in rural areas and promoting collaborative research between universities and companies. The system utilizes a high-power GPU data center to support large-scale data processing and simulation. It also provides academic paper collection, summarization, terminology explanations, and search functions, facilitating the application of research results.

[2109] Collecting and summarizing academic papers

[2110] Program operation description:

[2111] Server: Automatically collects academic papers in a specified field from each academic database. For example, to obtain the latest papers on machine learning, a query is sent to the database using APIs or scraping technology and the relevant papers are downloaded.

[2112] Server: Summarizes the collected papers using natural language processing (NLP) techniques. Specifically, after text cleaning, a summary generation algorithm is applied to extract key points.

[2113] Server: Automatically detects technical terms contained in the summary and provides explanations. Refers to a pre-prepared technical terminology dictionary database and adds appropriate explanations to the summary text.

[2114] Examples:

[2115] Server: Collects 100 of the latest papers in the field of machine learning, summarizes their contents, and extracts key points. When the technical term "backpropagation" appears, an explanation of this term is added and included in the summary.

[2116] Providing a search system

[2117] Program operation description:

[2118] User: Enter search keywords from the terminal to search for abstracts of academic papers.

[2119] Server: Based on the entered keywords, the database is searched and relevant summary information is obtained.

[2120] On the device: Display search results to the user, including links to terminology explanations and original papers.

[2121] Examples:

[2122] User: Enter the keywords "environmentally friendly plastic manufacturing techniques."

[2123] Server: Retrieves abstracts of relevant papers from the database and sends them to the user's device.

[2124] On the device: The search results show the user 20 abstracts, each containing a link to an explanation of the term "biodegradable plastic."

[2125] Intermediary for joint research

[2126] Program operation description:

[2127] User (company representative): Enters the company's research needs and issues from the terminal.

[2128] User (university professor): Enters the requirements for collaborative research from a terminal.

[2129] Server: Analyzes input data from companies and universities and makes appropriate matches.

[2130] Server: Notifies companies and universities of the matching results.

[2131] Examples:

[2132] User (company representative): Enter the topic "Research on optimizing the manufacturing process of biodegradable plastics."

[2133] User (university professor): Enters the requirement, "I would like to conduct joint research on materials development."

[2134] Server: Analyzes the data from both parties and matches compatible pairs. Upon notification, both parties contact each other and start collaboration.

[2135] Utilizing GPU data centers

[2136] Program operation description:

[2137] Users (researchers, company personnel): Input requests for computing resources from their terminals.

[2138] Server: Receives requests and schedules computing resources.

[2139] Server: Executes computational tasks in a high-powered GPU data center and returns the results to the user.

[2140] Examples:

[2141] User (company representative): Requests the necessary computing resources to conduct a simulation of the plastics manufacturing process.

[2142] Server: Assigns requests to a powerful GPU and starts the simulation.

[2143] Server: Acquires the simulation results and sends them to the user's device.

[2144] Terminal: Receives the resulting data and analyzes it using visualization tools.

[2145] This will strengthen cooperative relationships between local companies and universities, and by introducing the latest research results to local areas, it will provide a system that will revitalize the local economy and achieve sustainable growth.

[2146] The processing flow will be explained below.

[2147] Collecting and summarizing academic papers

[2148] Step 1:

[2149] Server: Sends queries to the API endpoints of each academic database, including subject areas and publication date ranges.

[2150] Example: To retrieve the latest papers in the field of machine learning, send a specified query to the API of "Academic Database A."

[2151] Step 2:

[2152] Server: Receives article data returned from academic databases and stores metadata (title, author, publication date, etc.) and the text in local storage.

[2153] Example: Parse the paper data returned from the API and save the title "Machine Learning Algorithms", the author "John Doe", and the body of the paper.

[2154] Step 3:

[2155] Server: Performs text cleaning on the received paper text, removing unnecessary parts (headers, footers, tables, etc.).

[2156] Example: Analyzing the text of a paper using regular expressions to remove unnecessary strings and structures.

[2157] Step 4:

[2158] Server: Apply natural language processing (NLP) algorithms to the cleaned-up paper text to generate summaries.

[2159] Example: Using a text summarization algorithm (e.g., Textrank) to extract the main points of an article.

[2160] Step 5:

[2161] Server: Automatically detects technical terms contained in the abstract and retrieves explanations by referencing a pre-prepared technical terminology dictionary database.

[2162] Example: Detect "backpropagation" from a summary and add its explanation "Learning algorithm in neural networks."

[2163] Step 6:

[2164] Server: Stores abstracts and their terminology in a database, and associates them with metadata.

[2165] Example: Save the abstract "This paper demonstrates the efficiency of an algorithm using backpropagation" and its explanation in a database.

[2166] Providing a search system

[2167] Step 1:

[2168] User: Enters search keywords into the device's search interface. The input is sent to the server.

[2169] Example: A user types in "environmentally friendly plastic manufacturing techniques" and presses the search button.

[2170] Step 2:

[2171] Server: Searches for abstracts in the database based on received keywords. Sorts the results based on relevance.

[2172] Example: Obtain multiple summaries that match the search keywords and sort them by relevance.

[2173] Step 3:

[2174] Server: Structures search results and sends them to the user's device, including abstracts, glossary links, and links to original articles.

[2175] Example: Send a summary and its link information to the user's device in JSON format.

[2176] Step 4:

[2177] Terminal: Displays the received search results. When the user clicks on a result, a detailed summary and glossary of terms is displayed.

[2178] Example: Search results are displayed in a list of 20 abstracts, and when the user clicks on the link for "biodegradable plastic," a pop-up explanation appears.

[2179] Intermediary for joint research

[2180] Step 1:

[2181] User (company representative): Enters the company's research needs and issues from the terminal and sends them to the server.

[2182] Example: A company enters "Research on optimizing the manufacturing process of biodegradable plastics."

[2183] Step 2:

[2184] User (university professor): Enters university research requirements from the terminal and sends them to the server.

[2185] Example: A university professor enters, "I would like to conduct joint research on materials development."

[2186] Step 3:

[2187] Server: Analyzes input data from companies and universities, calculates the degree of compatibility between research needs and requirements, and matches compatible pairs.

[2188] Example: Applying machine learning algorithms to analyze input data and calculate a matching score.

[2189] Step 4:

[2190] Server: Notifies companies and universities of match results, including contact information and next step instructions.

[2191] Example: Notify the company and university of the match results and provide contact information for both parties.

[2192] Utilizing GPU data centers

[2193] Step 1:

[2194] Users (researchers, company personnel): Enter requests for computing resources from their terminals and send them to the server.

[2195] Example: Requesting specific computational resources (number of GPUs, time, etc.) for a simulation of a plastics manufacturing process.

[2196] Step 2:

[2197] Server: Analyzes received requests and schedules them by allocating available computing resources.

[2198] Example: Check the resource status of the GPU data center and allocate an appropriate GPU.

[2199] Step 3:

[2200] Server: Executes computational tasks and initiates data processing and simulations using high-powered GPUs.

[2201] Example: Loading required datasets and launching scripts to perform computational tasks.

[2202] Step 4:

[2203] Server: Receives the calculation results and sends them to the user's device. The results are formatted in a format that can be analyzed.

[2204] Example: Convert simulation results into JSON or CSV format and send them to the user's device.

[2205] Step 5:

[2206] Terminal: Displays the received calculation results and helps the user analyze them.

[2207] Example: Visualizing calculation results in graphs and charts and presenting them to the user.

[2208] By using the above specific processing steps, the system of the present invention can realize efficient utilization of academic research, promotion of collaborative research, and large-scale data processing.

[2209] Example 1

[2210] 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."

[2211] The lack of effective methods for utilizing academic research resources in rural areas, the lack of promotion of collaborative research between universities and companies, and the lack of efficient processing of large amounts of data and simulation environments are issues that need to be addressed. Furthermore, there is a need to improve the situation where academic literature collection, summaries, explanations of technical terms, and effective search functions are not adequately provided.

[2212] 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.

[2213] In this invention, the server includes means for automatically collecting academic literature from an academic information repository, means for summarizing the collected academic literature using natural language processing technology, means for automatically detecting and defining technical terms that appear in the summarized academic literature, means for storing the summaries and definitions in the information repository, means for searching the information repository based on keywords and providing related summary information, means for displaying related information extracted from the information repository on a user terminal, means for performing large-volume data processing and simulations using a high-performance computing unit, and means for accepting and scheduling requests for computing resources, thereby enabling the effective collection and use of academic information, the promotion of collaborative research, and the efficient processing of large volumes of data.

[2214] An "academic information repository" is a database for storing, organizing, and accessing academic literature and research data.

[2215] "Academic literature" refers to documents such as papers, articles, and reports published as t...

Claims

1. A means for automatically collecting academic papers from academic databases; A means for summarizing collected academic papers using natural language processing technology; A method for automatically detecting and explaining technical terms that appear in abstracted academic papers; a means for storing the abstract and commentary in a database; a means for searching the database based on keywords and providing relevant summary information; A system including:

2. A means for companies to input their research needs and issues, A means for universities to input their collaborative research requirements; A means of analyzing input data from companies and universities and matching compatible pairs; A means for notifying the matching result; The system of claim 1 , comprising:

3. A means of processing large amounts of data and performing simulations using high-power GPUs, a means for accepting and scheduling requests for computing resources; means for returning the results of the calculation to the user; The system of claim 1 , comprising:

Images