System

Abstract

A system is provided.SOLUTION: A system comprising: means for obtaining paper data; means for tokenizing and analyzing the paper data using a generative artificial intelligence; means for generating a summary from an analysis result of the generative artificial intelligence; means for providing peer review support information based on the generated summary; and means for displaying the peer review support information through a user interface.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] The paper review process in modern scientific journals involves a huge amount of manual work, placing a heavy burden on reviewers. This lengthens the time required for peer review, hindering the rapid and efficient development of academic research. Furthermore, with a limited number of reviewers, maintaining the quality of peer review becomes a challenge as the number of submitted papers increases. Furthermore, while attempts to automate the process using generative artificial intelligence are progressing, current technology still has issues with the reliability and accuracy of the information generated. [Means for solving the problem]

[0005] In order to solve the above problems, the present invention proposes a system for automating the paper peer review process and providing efficient and accurate support. The system of the present invention includes the following means:

[0006] 1. Methods for obtaining paper data: Automatically obtain paper data from the Internet or databases.

[0007] 2. Using generative AI to tokenize and analyze paper data: Generative AI is used to tokenize and analyze the acquired paper data to understand the content of the paper.

[0008] 3. A means of generating summaries from the analysis results of generative artificial intelligence: Based on the analysis results, the main points of the paper are summarized.

[0009] 4. A means of providing peer review support information based on the generated abstract: Provide supporting information that can be used by reviewers based on the generated abstract.

[0010] 5. A means of displaying peer review support information through the user interface: Display peer review support information in a visually easy-to-understand format.

[0011] Through these measures, the system of this invention significantly streamlines the peer review process, enabling rapid and accurate evaluation of academic research. Furthermore, the generative AI's ability to detect hallucination risks and suggest corrections provides reliable peer review support. Furthermore, with its multilingual processing capabilities and flexibility to handle patent document acceptance and rejection, the system is expected to be used in a wide range of fields.

[0012] "Article data" refers to digital data containing the contents of academic papers, including information such as text, images, and charts.

[0013] "Generative AI" is a type of AI technology designed to perform tasks such as natural language processing and pattern recognition, and is particularly used for text generation and sentence analysis.

[0014] "Tokenization" is the process of dividing text data into analyzable units (tokens), and is part of the preprocessing required for input to a language model.

[0015] "Analysis" refers to the procedures and techniques used to process acquired data and information and decipher its meaning and structure.

[0016] A summary is a short summary of the main points or important information of an original text, capturing the essence of its content.

[0017] "Peer review support information" is information provided to support the evaluation of the content of a paper, including summaries of key points and feedback based on evaluation criteria.

[0018] A "user interface" is a facility that allows interaction between a system and a user and visually displays data input and output.

[0019] "Hallucination" is a phenomenon in which a generative artificial intelligence generates information that does not exist, resulting in the output of incorrect content.

[0020] A "database" is a system or structure that organizes and stores data so that it can be easily accessed and managed.

[0021] "Multilingual processing ability" means having the ability to analyze and process text data written in different languages.

[0022] "Patent documents" refer to documents and information related to patent applications and patent rights, which describe technical details and the scope of patent rights. [Brief explanation of the drawings]

[0023] [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

[0024] 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.

[0025] First, the terms used in the following description will be explained.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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."

[0031] [First embodiment]

[0032] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0033] 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.

[0034] 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).

[0035] 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.

[0036] 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.

[0037] 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.

[0038] 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.

[0039] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0040] 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.

[0041] 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.

[0042] 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.

[0043] 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."

[0044] The present invention relates to a system for automating the peer review of academic papers and providing efficient peer review support. This system utilizes generative artificial intelligence to acquire and analyze paper data, generate summaries, and provide peer review support information.

[0045] System configuration

[0046] 1. Data Acquisition Module: The server uses this module to acquire paper data from the Internet or databases. It automatically downloads the paper data from the URL or path specified by the user and converts the format as necessary.

[0047] 2. Generative AI module: The server uses generative AI to tokenize and analyze the acquired paper data. The generative AI uses natural language processing models, for example, to understand the content of the paper and extract key points.

[0048] 3. Summary generation module: The server generates a summary of the paper based on the analysis results of the AI. The generated summary is a concise summary of the main points of the paper, allowing reviewers to quickly grasp the overall picture of the paper.

[0049] 4. Peer review support module: Based on the generated abstract, the server provides specific peer review support information, including an evaluation of the paper's novelty, logic, and content consistency.

[0050] 5. User Interface Module: The terminal displays peer review support information to the user. This interface is designed to be visually easy to understand, allowing users to easily access and understand the information.

[0051] Program processing flow

[0052] 1. Acquisition of paper data: The server acquires the paper data from the URL or database specified by the user. This acquired data is temporarily stored in the server's memory.

[0053] 2. Analyzing the paper data: The server tokenizes the acquired data and analyzes it using a generative AI, which understands the main parts of the paper and summarizes its contents.

[0054] 3. Summary generation: The server creates a summary from the analysis results of the AI. This summary concisely presents important information to reviewers, streamlining the peer review process.

[0055] 4. Providing peer review support information: The server generates peer review support information based on the abstract, including an evaluation of the paper's novelty and logic.

[0056] 5. Display of peer review support information: The terminal displays peer review support information to the user, allowing the user to review papers efficiently based on this information.

[0057] Usage example

[0058] For example, if a user reviews a paper located at "https: / / example.com / sample_paper.pdf," the server accesses the URL and retrieves the paper data. It then uses generative AI to analyze the paper and generate a summary. It then creates peer review support information based on the generated summary and provides it to the user via their device. The user can use the provided support information to quickly and accurately review the paper.

[0059] This system will significantly improve the efficiency of the peer review process and support the rapid evaluation of academic research. Furthermore, since it can handle multilingual processing and patent document acceptance / rejection, it is expected to be applied in a wide range of fields.

[0060] The processing flow will be explained below.

[0061] Step 1:

[0062] The user inputs the URL of the paper they wish to review and sends it to the server via their device. The user then inputs this URL into a dedicated interface and issues a request to retrieve it.

[0063] Step 2:

[0064] The server retrieves the paper data from the specified URL. The server sends an HTTP request and retrieves the paper data as a response. If the data retrieval fails, the server generates an error message and notifies the user.

[0065] Step 3:

[0066] The server tokenizes the acquired paper data and converts it into an input format for the generative AI, which is a preprocessing step to enable the generative AI to accurately analyze the data.

[0067] Step 4:

[0068] The server uses a generative AI to analyze the tokenized paper data, which analyzes each part of the paper and understands its main points and meaning.

[0069] Step 5:

[0070] The server generates a summary from the AI ​​analysis results, extracting the most important information from the analysis results and concisely summarizing the main points of the paper.

[0071] Step 6:

[0072] The server provides peer review support information based on the abstracts generated by the server, which then generates feedback including evaluations of the paper's novelty, logic, and content consistency.

[0073] Step 7:

[0074] The server sends the generated peer review support information to the terminal, which displays the received information on a user interface for the reviewer to use.

[0075] Step 8:

[0076] Users can check the peer review support information provided through the device interface, evaluate the paper based on this information, and efficiently proceed with the peer review process.

[0077] Step 9:

[0078] The user enters the peer review results and sends them to the server via their terminal. The final peer review results are stored in the system and used to make decisions on whether to accept or reject the paper.

[0079] Through this series of processes, the system streamlines the paper peer review process and supports the rapid evaluation of academic research.

[0080] Example 1

[0081] 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."

[0082] The current peer review process is time-consuming and labor-intensive, making it difficult to evaluate many papers in a short period of time. Furthermore, there is a lack of consistency in the evaluation of papers' novelty, logic, and content consistency, resulting in inconsistent evaluations among reviewers. Therefore, there is a need for a system that streamlines the peer review process and improves the quality of evaluations.

[0083] 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.

[0084] In this invention, the server includes means for acquiring paper data from the Internet or a database, means for tokenizing the acquired paper data using a generative artificial intelligence and analyzing it using a natural language processing model, means for extracting important points from the analysis results of the generative artificial intelligence and generating a paper summary, means for providing peer review support information including an evaluation of the paper's novelty, logic, and content consistency based on the generated summary, and means for displaying the peer review support information through a user interface, thereby making it possible to streamline the peer review process and improve the quality of evaluation.

[0085] The "Internet" is a global communications network for accessing distributed information resources and retrieving, transmitting, and sharing data through computer networks.

[0086] A "database" is a system for managing data efficiently and systematically, allowing data to be searched, retrieved, and updated using specific queries.

[0087] "Generative AI" is an AI technology that has the ability to automatically handle specific tasks, learn like a human, analyze data, and generate deliverables.

[0088] "Tokenization" is the process of dividing text data into the smallest meaningful units, such as words and sentences, and is a preprocessing step in natural language processing.

[0089] A "natural language processing model" is a machine learning model for understanding, generating, and manipulating natural language, enabling computers to interpret human language.

[0090] A "summary" is a short, concise summary of the main points and conclusions from a longer document.

[0091] "Novelty" is a concept that indicates that a particular research or invention is something new that has not existed before.

[0092] "Logicality" refers to whether an argument or conclusion is based on a consistent logical structure, and is a standard for evaluating whether it is logically sound.

[0093] "Content relevance" is a concept that evaluates whether the content of a document is appropriate to its purpose or theme and meets the expected items and standards.

[0094] A "user interface" is a means by which a user interacts with a computer system, providing a visually easy-to-understand display of information and a method of operation.

[0095] This invention relates to a system that automates and improves the efficiency of the paper peer review process by using a generative AI model to acquire paper data, analyze it, generate summaries, and provide and display peer review support information.

[0096] 1. Acquisition of paper data

[0097] First, the user specifies the URL of the paper data they want to retrieve from the Internet or a database. The server accesses the specified URL and downloads the paper data via the HTTP protocol. During this process, the server uses the Python requests library to send an HTTP request, saves the paper data locally, and converts the format if necessary.

[0098] 2. Analysis of paper data

[0099] The acquired paper data is tokenized by the server. For example, if the data is in PDF format, text extraction is performed using the PyMuPDF library. This text data is input to a generative AI model, such as GPT-3 (registered trademark), for analysis. The generative AI analyzes the content of the paper using the specified prompt sentence and extracts key points.

[0100] 3. Summary Generation

[0101] The server generates a summary of the paper based on the analysis results of the generative AI. This summary is a concise summary of the main points of the paper, allowing reviewers to quickly grasp the overall picture of the paper.

[0102] 4. Providing peer review support information

[0103] From the generated abstract, the server generates specific peer review support information, including ratings of novelty, logic, and relevance of the content, allowing reviewers to quickly determine which parts of the paper are worthy of evaluation.

[0104] 5. Viewing peer review support information

[0105] Finally, the terminal displays the peer review support information sent from the server to the user. The user interface is designed to be visually easy to understand, allowing users to easily access and understand the information. For example, an interface using a web browser is used.

[0106] Specific examples

[0107] Consider a case where a user reviews a paper located at "https: / / example.com / sample_paper.pdf." First, the server accesses the URL and retrieves the paper data. Next, it analyzes this data using the generative AI model "GPT-3" and uses the following prompt:

[0108] "Please write an abstract of this paper. Include the main points of your conclusions, an outline of your methodology, and any key findings from your results."

[0109] Based on the generated summary, the server creates peer review support information on the paper's novelty, logic, and content consistency, which is displayed on the terminal, allowing users to efficiently proceed with the peer review.

[0110] This system will significantly improve the efficiency of the peer review process, enabling the rapid evaluation of academic research. Furthermore, it is capable of handling multilingual processing and the acceptance or rejection of patent documents, and is expected to be applied in a wide range of fields.

[0111] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0112] Step 1:

[0113] Acquisition of paper data

[0114] The server receives a URL specified by the user as input. It accesses the specified URL, for example, "https: / / example.com / sample_paper.pdf," and downloads the paper data using the HTTP protocol. During this process, the server uses the requests library to send an HTTP request and saves the retrieved data locally. The output is the paper data saved in the server's temporary memory or storage.

[0115] Step 2:

[0116] Tokenization and formatting

[0117] The server receives the paper data obtained in step 1 as input. If the paper data is in PDF format, it extracts the text using a library such as PyMuPDF and tokenizes the data. Specifically, it extracts the text of each page from the PDF file and splits it into a series of words and sentences as text data. The output is text data that can be analyzed.

[0118] Step 3:

[0119] Analysis by generative AI

[0120] The server receives the text data generated in step 2 as input. It uses a generative AI model, such as GPT-3, to analyze the paper data. It provides the generative AI with the prompt, "Write a summary of this paper. Include the main points of the conclusion, an overview of the methodology, and important findings from the results." and performs the analysis. The output is the extracted main points and summary text.

[0121] Step 4:

[0122] Generate a summary

[0123] The server receives as input the analysis results generated in step 3. From the analysis results generated, it creates a summary of the paper. The summary is a concise summary of the main sections, such as conclusions, methodology, and results. The summary is stored in a data structure on the server. The output is a summary of the completed paper.

[0124] Step 5:

[0125] Providing peer review support information

[0126] The server receives the abstract generated in step 4 as input. Based on the abstract, it creates peer review support information regarding novelty, logic, and relevance of the content. This information includes evaluations such as "This paper is innovative compared to previous research," "The logical development is consistent and the conclusion is persuasive," and "The content is highly relevant to the research objectives." The output is specific peer review support information.

[0127] Step 6:

[0128] View peer review support information

[0129] The terminal receives as input the peer review support information generated in step 5. It displays this information to the user through a visually easy-to-understand user interface. For example, it formats the information using HTML and CSS in the browser, providing a display format that is easily accessible and understandable to the user. The output is a screen displaying the peer review support information that the user can view.

[0130] Through these steps, the system will be able to streamline the peer review process and support the rapid evaluation of academic research.

[0131] (Application example 1)

[0132] 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."

[0133] Traditional peer review processes are often manual and require a great deal of time and effort. Furthermore, the lack of uniform evaluation criteria makes it difficult to obtain consistent peer review results. Furthermore, when multilingual processing is required, language barriers exist, limiting the access and sharing of information in a field of knowledge. This creates a challenge for researchers and students to quickly grasp the latest research information.

[0134] 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.

[0135] In this invention, the server includes: means for acquiring paper data; means for tokenizing and analyzing the paper data using a generative AI; means for generating summaries from the analysis results of the generative AI; means for providing peer review support information based on the generated summaries; means for generating prompts for evaluating the novelty, logic, and content consistency of the paper based on the summaries in generating the peer review support information; and means for displaying the peer review support information through a user interface. This automates the peer review process, enabling efficient and consistent peer review. Furthermore, multilingual processing is possible, allowing researchers and students to quickly and easily access international research information.

[0136] "Paper data" refers to documents that describe academic research content and are stored electronically.

[0137] "Generative AI" refers to artificial intelligence technology that uses machine learning and deep learning to analyze and generate natural language.

[0138] "Tokenization" refers to the process of breaking down text into smaller units such as words or phrases.

[0139] "Analysis" refers to the process of understanding the acquired data and grasping its meaning and structure.

[0140] "Generative AI" refers to artificial intelligence that has the ability to generate new information or summaries based on data.

[0141] An "abstract" is a short sentence or paragraph that succinctly summarizes the contents of the entire document.

[0142] "Peer review support information" is specific information provided to assist in the evaluation of a paper, including an assessment of the paper's importance and quality.

[0143] A "prompt sentence" refers to an input sentence that causes a generation AI to generate a specific output.

[0144] "User interface" refers to the interface through which a system and a user communicate with each other.

[0145] In this invention, the server acquires and analyzes paper data using the following means to generate peer review support information.

[0146] 1. Acquisition of paper data

[0147] The server retrieves paper data from a URL or database. The hardware used for this is a cloud server, and the software used is an "HTTP request library" (e.g., Retrofit, OkHttp). The server accesses the URL specified by the user and downloads the paper data. For example, if the user specifies a paper at "https: / / example.com / sample_paper.pdf," the server retrieves the paper data from that URL.

[0148] 2. Analysis of paper data

[0149] The server tokenizes the acquired paper data and analyzes it using generative AI. The hardware used is a cloud server, and the software used is a "natural language processing model" (e.g., GPT-4 (registered trademark), BERT). Tokenization breaks the paper data down into small units (words and phrases) that are then analyzed by generative AI.

[0150] 3. Summary Generation

[0151] The server generates a summary from the analysis results of the generative AI. The generative AI model used (e.g., BART, a large-scale language model) is used to summarize the analysis results as short sentences or paragraphs. For example, a BART model is used to generate a summary that concisely summarizes the main points of a paper.

[0152] 4. Generating peer review support information

[0153] Based on the generated abstract, we provide peer review support information, including generating prompts to evaluate the novelty, logic, and relevance of the content of the paper based on the abstract. For example, the generated prompts are as follows:

[0154] "Based on the following paper abstract, please rate the paper's novelty, logic, and relevance to the content. Abstract: [generated abstract content]"

[0155] 5. Display in the user interface

[0156] The device displays the generated peer review support information to the user through a user interface. The hardware used is a smartphone or computer, and the software is a "UI framework" (e.g., Flutter (registered trademark) or React Native). Based on this information, the user can review papers efficiently and accurately.

[0157] Program execution example

[0158] For example, if a user reviews a paper located at "https: / / example.com / sample_paper.pdf," the server accesses the URL and retrieves the paper data. Next, it uses GPT-4 to analyze the paper and generates a summary using the BART model. Based on the summary, it inputs a prompt to GPT-3.5-turbo to generate specific review support information, which is then displayed on the smartphone's user interface.

[0159] As a result, this invention automates the peer review process, making it possible to provide fast and consistent peer review. Furthermore, its multilingual capabilities make it easier to access international research information.

[0160] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0161] Step 1:

[0162] The server retrieves the paper data from a URL specified by the user or from a database. The input is a URL provided by the user, and an HTTP request is made based on this to download the paper data. Specifically, the server sends a request using an HTTP request library (e.g., Retrofit, OkHttp) and retrieves PDF or text data as a response. The output is the retrieved paper data.

[0163] Step 2:

[0164] The server tokenizes and analyzes the acquired paper data. The input is the paper data acquired in step 1, which is tokenized using generative artificial intelligence. Specifically, it uses a natural language processing library (e.g., SpaCy, NLTK) to split (tokenize) the paper data into words and phrases. It then analyzes the content using a model such as GPT-4 or BERT. The output is the analyzed data (the tokenized data and its analysis results).

[0165] Step 3:

[0166] The server generates a summary from the analysis results. The input is the analysis result data obtained in step 2, which is input into a summary generation model (e.g., BART, a large-scale language model). The specific operation is to use a summary generation algorithm to extract the main points from the content of the entire paper and generate a concise summary. The output is the generated summary.

[0167] Step 4:

[0168] The server provides peer review support information based on the generated summary. The input is the summary generated in step 3, which is used to generate a prompt and provide evaluation information for peer review. Specifically, it uses a model such as GPT-3.5-turbo to generate a prompt to evaluate the paper's novelty, logic, and content relevance based on the summary, and then generates specific peer review support information based on the prompt. The output is peer review support information. For example, the generated prompt is as follows:

[0169] "Based on the following paper abstract, please rate the paper's novelty, logic, and relevance to the content. Abstract: [generated abstract content]"

[0170] Step 5:

[0171] The terminal displays the generated peer review support information to the user through a user interface. The input is the peer review support information generated in step 4, and a UI framework (e.g., Flutter, React Native) is used to display this information in an easy-to-understand manner for the user. The specific operation is to create a layout for visually displaying the peer review support information and display it on the screen. The output is the peer review support information displayed in a format that the user can view.

[0172] 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.

[0173] The present invention relates to a system that automates the peer review of papers and provides efficient peer review support. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide a more personalized peer review experience. The following describes embodiments of the present invention.

[0174] System configuration

[0175] 1. Data Acquisition Module: The server uses this module to acquire paper data from the Internet or databases. It downloads the paper data from the URL specified by the user and converts the format if necessary.

[0176] 2. Generative AI module: The server uses generative AI to tokenize and analyze the acquired paper data. The generative AI uses a natural language processing model to understand the content of the paper and extract key points.

[0177] 3. Summary generation module: The server generates a summary of the paper based on the analysis results of the AI. The summary is a concise summary of the main points of the paper, allowing reviewers to quickly grasp the overall picture.

[0178] 4. Peer review support module: The server provides specific peer review support information based on the generated abstract, including an evaluation of the paper's novelty, logic, and content consistency.

[0179] 5. Emotion Engine: The emotion engine installed on the server analyzes the user's facial expressions, voice, keyboard input, etc. to recognize the user's emotional state. Based on this information, the peer review support information is dynamically adjusted.

[0180] 6. User interface module: The terminal displays peer review support information to the user. The user interface is designed to be visually easy to understand, allowing users to easily access and understand the information.

[0181] Program processing flow

[0182] 1. Acquisition of paper data:

[0183] The user enters the URL of the paper they wish to review and sends it to the server via their device. The server then retrieves the paper data from the specified URL.

[0184] 2. Analysis of paper data:

[0185] The server tokenizes the acquired paper data and analyzes it using a generative AI, which analyzes each part of the paper to understand its main points and meaning.

[0186] 3. Generate a summary:

[0187] The server generates a summary from the analysis results of the AI. The summary provides important information to reviewers in a concise manner, streamlining the peer review process.

[0188] 4. Providing peer review support information:

[0189] Based on the server-generated abstract, the system provides specific peer review support information, including an assessment of the paper's novelty, logic, and content relevance.

[0190] 5. Emotion Recognition with Emotion Engine:

[0191] The server recognizes the user's emotions from their facial expressions, voice, keyboard input, etc. The emotion engine analyzes this data and understands the user's current emotional state.

[0192] 6. Dynamic adjustment according to emotions:

[0193] The server adjusts the presentation and design of peer review support information according to the user's emotional state. For example, if the user is feeling stressed, the server will simplify the presentation of information.

[0194] 7. View peer review support information:

[0195] The terminal displays the adjusted peer review support information to the user, who then performs the peer review efficiently based on the provided information.

[0196] Usage example

[0197] For example, if a user reviews a paper located at "https: / / example.com / sample_paper.pdf," the server accesses the URL to retrieve the paper data. It then uses generative artificial intelligence to analyze the paper and generate a summary. It then creates review support information based on the generated summary and uses an emotion engine to recognize the user's emotional state. Finally, it adjusts the information according to the emotion and provides it to the user via their device. The user can use the provided support information to quickly and accurately review the paper.

[0198] This allows our system to significantly streamline the review process, and when combined with emotion recognition, it can provide a more personalized review experience.

[0199] The processing flow will be explained below.

[0200] The present invention relates to a system that automates the peer review of papers and provides efficient peer review support. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide a more personalized peer review experience. Below, an embodiment of the present invention will be described using specific processing steps.

[0201] Step 1:

[0202] The user enters the URL of the paper they wish to review and sends it to the server via their device. The user then enters the URL into a dedicated interface and issues a command to retrieve the paper data.

[0203] Step 2:

[0204] The server retrieves the paper data from the specified URL. The server sends an HTTP request, receives the paper data as a response, and temporarily stores the data in memory. If the retrieval fails, an error message is generated to notify the user.

[0205] Step 3:

[0206] The server tokenizes the acquired paper data and converts it into an input format for the generation AI. Tokenization is the process of dividing text data into analyzable units.

[0207] Step 4:

[0208] The server analyzes the tokenized paper data using a generative AI, which uses a natural language processing model to understand the content of the paper and extract key points.

[0209] Step 5:

[0210] The server generates a summary from the analysis results of the AI. Based on the analysis results, a summary that concisely summarizes the main points of the paper is created. The summary is important information that allows reviewers to quickly grasp the overall picture of the paper.

[0211] Step 6:

[0212] The server generates abstracts and provides peer review support information, including an assessment of the paper's novelty, logic, and content consistency.

[0213] Step 7:

[0214] The emotion engine installed on the server analyzes the user's facial expressions, voice, keyboard input, etc. to recognize the user's emotional state. The emotion engine uses this data to understand the user's current emotional state.

[0215] Step 8:

[0216] The server dynamically adjusts the presentation and design of peer review support information according to the user's perceived emotional state: for example, if the user is feeling stressed, it will simplify the information or add an encouraging message.

[0217] Step 9:

[0218] The terminal displays tailored peer review support information to the user. The user interface is designed to be visually intuitive, allowing users to easily access and understand the information.

[0219] Step 10:

[0220] Users can check the peer review support information provided through the device interface and evaluate papers based on that information, allowing them to efficiently carry out their peer review work.

[0221] Step 11:

[0222] The user enters the peer review results and sends them to the server via their terminal. The final peer review results are stored in the system and used to make decisions on whether to accept or reject the paper.

[0223] Through this series of processes, the system of the present invention significantly streamlines the peer review process and supports rapid evaluation of academic research. Furthermore, by combining it with an emotion engine, it is possible to provide users with a more appropriate and personalized peer review experience.

[0224] Example 2

[0225] 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."

[0226] The traditional paper peer review process is time-consuming and places a heavy burden on reviewers. This burden is particularly pronounced when the paper is written in multiple languages ​​or when the reviewer's emotional state affects their work efficiency. Furthermore, traditional systems provide static peer review support information and are not sufficiently personalized to meet the needs of individual users. There is a need for a system that can solve these problems and provide efficient and personalized paper peer review support.

[0227] 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.

[0228] In this invention, the server includes means for acquiring paper data, means for tokenizing and analyzing the paper data using a generating AI, means for generating a summary from the analysis result of the generating AI, means for providing peer review support information based on the generated summary, means for recognizing a user's emotional state, means for dynamically adjusting the peer review support information based on the recognized emotional state, and means for displaying the peer review support information through a user interface, thereby enabling efficient and personalized paper peer review support.

[0229] "Means for obtaining paper data" refers to a device or function for downloading paper data from a specified URL and converting the format as necessary.

[0230] "Means for tokenizing and analyzing paper data using generative artificial intelligence" refers to a device or function that uses a natural language processing model to divide paper data into individual tokens and analyze their contents.

[0231] "Means for generating summaries from the analysis results of generative artificial intelligence" refers to a device or function that generates a summary that succinctly summarizes the main points of a paper based on the analyzed data.

[0232] "Means for providing peer review support information based on the generated abstract" refers to a device or function that provides peer review support information such as the novelty, logic, and content consistency of a paper based on the generated abstract.

[0233] The "means for recognizing the emotional state of the user" refers to a device or function that analyzes emotions from the user's facial expressions, voice, keyboard input, etc., and recognizes that emotional state.

[0234] "Means for dynamically adjusting peer review support information based on a recognized emotional state" refers to a device or function that dynamically adjusts the display method and content of peer review support information in accordance with the recognized emotional state of the user.

[0235] A "means for displaying peer review support information through a user interface" is an interface designed to visually display the tailored peer review support information and to allow easy user access.

[0236] The present invention relates to a system for automating the peer review of papers and providing efficient and personalized peer review support to users. The system functions as follows through interactions between a server, a terminal, and a user.

[0237] The system includes the following functions: acquiring paper data, analyzing it using generative artificial intelligence, generating summaries, providing peer review support information, recognizing the user's emotional state and dynamically adjusting information, and displaying information through a user interface.

[0238] Acquisition of paper data

[0239] First, the user enters the URL of the paper to be peer-reviewed. The device sends this URL to the server, and the server accesses the specified URL to retrieve the paper data. The retrieved paper data is in PDF format, for example, and the server converts it into an appropriate format as needed. This format conversion is performed using an OCR tool (e.g., Tesseract).

[0240] Analysis using generative artificial intelligence

[0241] The server tokenizes the acquired paper data using natural language processing libraries (e.g., NLTK, SpaCy). This tokenized data is then fed into a generative AI model (e.g., GPT-4) that analyzes each part of the paper and extracts key points. This analysis process clarifies the paper's subject, purpose, results, etc.

[0242] Generate a summary

[0243] The server uses topic modeling and summary generation algorithms to generate summaries based on the analysis results obtained from the generative AI model. These summaries provide a compact summary of important information for reviewers, allowing for quick understanding.

[0244] Providing peer review support information

[0245] Based on the generated summary, the server generates peer review support information, including an evaluation of the paper's novelty, logic, and content consistency. The server scores the paper based on these criteria and uses a rule-based engine (e.g., if-then rules) to generate specific advice and comments.

[0246] Recognizing the user's emotional state

[0247] The server analyzes the user's facial expressions, voice, keyboard input, etc. collected through the device and uses an emotion engine to recognize the user's emotional state. For example, facial expression analysis and voice emotion recognition technology are used to determine whether the user is feeling stressed.

[0248] Dynamic adjustment of information

[0249] Based on the user's recognized emotional state, the server dynamically adjusts the display method and content of peer review support information. For example, if the user is feeling stressed, the server will adjust the display to show only the most important information in a concise manner.

[0250] Displaying Information Through a User Interface

[0251] Finally, the adjusted peer review support information is displayed on the user interface on the device. The user interface has an intuitive and easy-to-use design, allowing users to easily check the information. Users can utilize the provided support information to review papers quickly and accurately.

[0252] Prompt Sentence Examples

[0253] An example of a prompt that our system could input to the generative AI model would be:

[0254] Generate an abstract for the following paper:

[0255] URL: https: / / example.com / sample_paper.pdf

[0256] Summarize the contents of the paper and state the main points and conclusions.

[0257] This allows our system to significantly streamline the review process and, when combined with emotion recognition, provide a more personalized review experience.

[0258] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0259] Step 1:

[0260] The user enters the URL of the paper they wish to review, and the device sends the URL to the server.

[0261] Input: URL of the paper to be reviewed

[0262] Output: Request for paper data from server

[0263] Specific operation: The user enters a URL into the input field of the terminal and clicks the send button. The terminal sends this URL as an HTTP request to the server.

[0264] Step 2:

[0265] The server accesses the specified URL and retrieves the paper data.

[0266] Input: URL of the paper sent from the device

[0267] Output: Retrieved paper data (PDF format, etc.)

[0268] Specific operation: The server accesses the URL, downloads the paper data in PDF format, and temporarily stores it on the server.

[0269] Step 3:

[0270] The server tokenizes the acquired paper data and analyzes it using generative AI.

[0271] Input: Retrieved paper data

[0272] Output: Tokenized text data and analysis results

[0273] How it works: The server uses a natural language processing library (e.g., NLTK, SpaCy) to extract and tokenize text from the PDF, then feeds the tokenized data into a generative AI model (e.g., GPT-4) to parse out key points.

[0274] Step 4:

[0275] The server generates a summary from the analysis results of the generation AI.

[0276] Input: Analysis results by the generation AI

[0277] Output: Abstract text of the paper

[0278] Specific operation: The server uses topic modeling and summary generation algorithms to generate summaries from the analysis results, which are then stored in an internal database.

[0279] Step 5:

[0280] The server generates peer review support information based on the abstract.

[0281] Input: Generated summary

[0282] Output: Peer review support information (evaluation of novelty, logic, and consistency)

[0283] Specific Actions: The server analyzes the summary, calculates a score based on the evaluation criteria, and uses a rule-based engine to generate specific advice and comments.

[0284] Step 6:

[0285] The server recognizes the user's emotional state.

[0286] Input: User facial expression, voice, and keyboard input data collected from the device

[0287] Output: User's emotional state data

[0288] Specific operation: The device's camera and microphone collect the user's facial expressions and voice, and send them to the server. The server then uses emotion recognition technology to analyze the user's emotional state.

[0289] Step 7:

[0290] The server dynamically adjusts the peer review support information based on the perceived emotional state.

[0291] Input: User's emotional state data

[0292] Output: Adjusted peer review supporting information

[0293] Specific operation: The server receives the results of the emotion engine and adjusts the display method and content of the information. For example, if the user is feeling stressed, it will simply display only the important information.

[0294] Step 8:

[0295] The terminal displays the adjusted peer review support information on the user interface.

[0296] Input: Coordinated peer review supporting information

[0297] Output: Information displayed in the user interface

[0298] Specific operation: The terminal displays the adjusted peer review support information received from the server on the user interface. The user uses this information to perform the peer review efficiently.

[0299] The above is the specific processing flow of the program of this system.

[0300] (Application example 2)

[0301] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0302] Conventional paper peer review systems make it difficult for reviewers to efficiently review a large number of papers. Furthermore, because information is provided uniformly without considering the reviewer's emotional state, stress and dissatisfaction are likely to occur. To address these issues, a system that recognizes the user's emotions and provides dynamically adjusted peer review support information is needed.

[0303] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring paper data, means for tokenizing and analyzing the paper data using a generation AI, means for generating a summary from the analysis results of the generation AI, means for providing peer review support information based on the generated summary, means for displaying the peer review support information through a user interface, means for recognizing user emotions, and means for dynamically adjusting the peer review support information based on the user emotions. This allows reviewers to review papers efficiently and also allows them to receive personalized information according to their own emotional state.

[0304] "Paper data" refers to scientific or academic documents written by researchers or authors, and is material that presents specific research results or findings.

[0305] "Generative AI" refers to an AI technology that performs advanced processing such as natural language processing, tokenization, analysis, and summary generation based on large datasets.

[0306] "Tokenization" is the process of breaking down text into smaller units such as words and phrases in order to analyze it.

[0307] "Analysis" refers to the process of understanding the content of acquired data and extracting important information.

[0308] A summary is a concise summary of the main points of an entire text.

[0309] "Peer review support information" refers to information to assist reviewers in their work, such as evaluations of the paper's novelty, logic, and content consistency, as well as specific feedback.

[0310] A "user interface" is a means by which a user interacts with a computer system, and refers to a screen on which information is visually displayed and an operating means.

[0311] "User emotions" refer to the psychological states such as stress, excitement, and concentration that users experience when reviewing papers.

[0312] "Dynamic adjustment" refers to the process of changing the content and presentation of information in real time according to the user's emotions and situation.

[0313] The system that realizes this application example is composed of a series of modules that make up the entire system. The detailed operation of each module is shown below.

[0314] Program processing

[0315] Acquisition of paper data

[0316] First, the server retrieves the paper data from the URL specified by the user. This retrieved paper data is temporarily stored for subsequent analysis. Specifically, the server uses a network library (e.g., the requests library) to download the paper data via the Internet.

[0317] Analysis of paper data

[0318] The server then uses a generative AI model to tokenize the acquired paper data and analyze its content. The generative AI uses a natural language processing library (e.g., the Transformers library). Key points and important information are extracted from the analysis results, which then serve as the basis for generating summaries, as described below.

[0319] Generate a summary

[0320] The server generates a summary based on the analyzed data. The generated summary is intended to allow reviewers to quickly grasp the overall picture of the paper, concisely summarizing only the most important points. The summary is generated using the generative AI model mentioned above.

[0321] Emotion recognition by emotion engine

[0322] Furthermore, the server is equipped with an emotion engine that analyzes the user's facial expressions, voice, keyboard input, etc. to recognize their emotional state. This emotion data is then used to dynamically adjust the review support information. The emotion engine uses the dlib library, which provides facial recognition technology, and hardware such as a webcam.

[0323] Dynamic Adjustment

[0324] The server dynamically adjusts the peer review support information it provides based on the user's recognized emotional state. For example, if the user is feeling stressed, it displays information in a simplified manner to reduce the burden on the user. This allows the user to review efficiently in a relaxed state.

[0325] User Interface Display

[0326] Finally, the adjusted peer review support information is displayed on the user's device. The user interface is designed to be visually easy to understand, allowing users to quickly access the information they need. This display can be done using a web application framework (e.g., Flask or Django).

[0327] Specific examples

[0328] For example, if a user enters a URL such as "https: / / example.com / sample_paper.pdf" to review a paper, the server accesses the URL to retrieve the paper data. The retrieved data is analyzed using a generative AI model to generate a summary. The system then creates peer review support information based on the generated summary and uses an emotion engine to recognize the user's emotional state. Finally, the system adjusts the information according to the user's emotions and provides it to the user through a user interface.

[0329] An example of a prompt could be a question such as "What are the main points?", which allows the generative AI model to extract key points from the paper and support efficient peer review.

[0330] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0331] Step 1:

[0332] The user enters the URL of the paper they wish to review and sends it to the server via their terminal. When the server receives this URL, it uses a network library (for example, the requests library) to retrieve the paper data from the URL. The retrieved paper data is temporarily stored on the server. The input is the "URL of the paper" and the output is the "paper data."

[0333] Step 2:

[0334] The server tokenizes and analyzes the acquired paper data using a generative AI model (e.g., the transformers library). This process understands the structure and content of the entire paper and extracts key points. The input is "paper data" and the output is "analysis results."

[0335] Step 3:

[0336] The server generates a summary based on the analysis results. The generated summary is intended to allow reviewers to quickly grasp the overall picture of the paper. This summary generation is also performed using a generative AI model. The input is the "analysis results" and the output is the "summary."

[0337] Step 4:

[0338] The server generates peer review support information based on the abstract. This information includes an evaluation of the paper's novelty, logic, and content consistency. The input is the "abstract" and the output is the "peer review support information."

[0339] Step 5:

[0340] The server uses an emotion engine to analyze the user's facial expressions, voice, and keyboard input to recognize their emotional state. The emotion engine uses the dlib library, which provides facial recognition technology, and hardware such as a webcam. The input is "user's emotional data (facial expressions, voice, and input)" and the output is "recognized emotional state."

[0341] Step 6:

[0342] The server dynamically adjusts the peer review support information based on the recognized emotional state. For example, if the user is feeling stressed, it simplifies the information and makes the display easier to understand. The inputs are the "recognized emotional state" and "peer review support information," and the output is "adjusted peer review support information."

[0343] Step 7:

[0344] The terminal displays the adjusted peer review support information to the user. The information is provided in a visually easy-to-understand format through the user interface. The input is the adjusted peer review support information, and the output is the display to the user.

[0345] 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.

[0346] 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.

[0347] 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.

[0348] [Second embodiment]

[0349] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0350] 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.

[0351] 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).

[0352] 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.

[0353] 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.

[0354] 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).

[0355] 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.

[0356] 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.

[0357] 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.

[0358] 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.

[0359] 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.

[0360] 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."

[0361] The present invention relates to a system for automating the peer review of academic papers and providing efficient peer review support. This system utilizes generative artificial intelligence to acquire and analyze paper data, generate summaries, and provide peer review support information.

[0362] System configuration

[0363] 1. Data Acquisition Module: The server uses this module to acquire paper data from the Internet or databases. It automatically downloads the paper data from the URL or path specified by the user and converts the format as necessary.

[0364] 2. Generative AI module: The server uses generative AI to tokenize and analyze the acquired paper data. The generative AI uses natural language processing models, for example, to understand the content of the paper and extract key points.

[0365] 3. Summary generation module: The server generates a summary of the paper based on the analysis results of the AI. The generated summary is a concise summary of the main points of the paper, allowing reviewers to quickly grasp the overall picture of the paper.

[0366] 4. Peer review support module: Based on the generated abstract, the server provides specific peer review support information, including an evaluation of the paper's novelty, logic, and content consistency.

[0367] 5. User Interface Module: The terminal displays peer review support information to the user. This interface is designed to be visually easy to understand, allowing users to easily access and understand the information.

[0368] Program processing flow

[0369] 1. Acquisition of paper data: The server acquires the paper data from the URL or database specified by the user. This acquired data is temporarily stored in the server's memory.

[0370] 2. Analyzing the paper data: The server tokenizes the acquired data and analyzes it using a generative AI, which understands the main parts of the paper and summarizes its contents.

[0371] 3. Summary generation: The server creates a summary from the analysis results of the AI. This summary concisely presents important information to reviewers, streamlining the peer review process.

[0372] 4. Providing peer review support information: The server generates peer review support information based on the abstract, including an evaluation of the paper's novelty and logic.

[0373] 5. Display of peer review support information: The terminal displays peer review support information to the user, allowing the user to review papers efficiently based on this information.

[0374] Usage example

[0375] For example, if a user reviews a paper located at "https: / / example.com / sample_paper.pdf," the server accesses the URL and retrieves the paper data. It then uses generative AI to analyze the paper and generate a summary. It then creates peer review support information based on the generated summary and provides it to the user via their device. The user can use the provided support information to quickly and accurately review the paper.

[0376] This system will significantly improve the efficiency of the peer review process and support the rapid evaluation of academic research. Furthermore, since it can handle multilingual processing and patent document acceptance / rejection, it is expected to be applied in a wide range of fields.

[0377] The processing flow will be explained below.

[0378] Step 1:

[0379] The user inputs the URL of the paper they wish to review and sends it to the server via their device. The user then inputs this URL into a dedicated interface and issues a request to retrieve it.

[0380] Step 2:

[0381] The server retrieves the paper data from the specified URL. The server sends an HTTP request and retrieves the paper data as a response. If the data retrieval fails, the server generates an error message and notifies the user.

[0382] Step 3:

[0383] The server tokenizes the acquired paper data and converts it into an input format for the generative AI, which is a preprocessing step to enable the generative AI to accurately analyze the data.

[0384] Step 4:

[0385] The server uses a generative AI to analyze the tokenized paper data, which analyzes each part of the paper and understands its main points and meaning.

[0386] Step 5:

[0387] The server generates a summary from the AI ​​analysis results, extracting the most important information from the analysis results and concisely summarizing the main points of the paper.

[0388] Step 6:

[0389] The server provides peer review support information based on the abstracts generated by the server, which then generates feedback including evaluations of the paper's novelty, logic, and content consistency.

[0390] Step 7:

[0391] The server sends the generated peer review support information to the terminal, which displays the received information on a user interface for the reviewer to use.

[0392] Step 8:

[0393] Users can check the peer review support information provided through the device interface, evaluate the paper based on this information, and efficiently proceed with the peer review process.

[0394] Step 9:

[0395] The user enters the peer review results and sends them to the server via their terminal. The final peer review results are stored in the system and used to make decisions on whether to accept or reject the paper.

[0396] Through this series of processes, the system streamlines the paper peer review process and supports the rapid evaluation of academic research.

[0397] Example 1

[0398] 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."

[0399] The current peer review process is time-consuming and labor-intensive, making it difficult to evaluate many papers in a short period of time. Furthermore, there is a lack of consistency in the evaluation of papers' novelty, logic, and content consistency, resulting in inconsistent evaluations among reviewers. Therefore, there is a need for a system that streamlines the peer review process and improves the quality of evaluations.

[0400] 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.

[0401] In this invention, the server includes means for acquiring paper data from the Internet or a database, means for tokenizing the acquired paper data using a generative artificial intelligence and analyzing it using a natural language processing model, means for extracting important points from the analysis results of the generative artificial intelligence and generating a paper summary, means for providing peer review support information including an evaluation of the paper's novelty, logic, and content consistency based on the generated summary, and means for displaying the peer review support information through a user interface, thereby making it possible to streamline the peer review process and improve the quality of evaluation.

[0402] The "Internet" is a global communications network for accessing distributed information resources and retrieving, transmitting, and sharing data through computer networks.

[0403] A "database" is a system for managing data efficiently and systematically, allowing data to be searched, retrieved, and updated using specific queries.

[0404] "Generative AI" is an AI technology that has the ability to automatically handle specific tasks, learn like a human, analyze data, and generate deliverables.

[0405] "Tokenization" is the process of dividing text data into the smallest meaningful units, such as words and sentences, and is a preprocessing step in natural language processing.

[0406] A "natural language processing model" is a machine learning model for understanding, generating, and manipulating natural language, enabling computers to interpret human language.

[0407] A "summary" is a short, concise summary of the main points and conclusions from a longer document.

[0408] "Novelty" is a concept that indicates that a particular research or invention is something new that has not existed before.

[0409] "Logicality" refers to whether an argument or conclusion is based on a consistent logical structure, and is a standard for evaluating whether it is logically sound.

[0410] "Content relevance" is a concept that evaluates whether the content of a document is appropriate to its purpose or theme and meets the expected items and standards.

[0411] A "user interface" is a means by which a user interacts with a computer system, providing a visually easy-to-understand display of information and a method of operation.

[0412] This invention relates to a system that automates and improves the efficiency of the paper peer review process by using a generative AI model to acquire paper data, analyze it, generate summaries, and provide and display peer review support information.

[0413] 1. Acquisition of paper data

[0414] First, the user specifies the URL of the paper data they want to retrieve from the Internet or a database. The server accesses the specified URL and downloads the paper data via the HTTP protocol. During this process, the server uses the Python requests library to send an HTTP request, saves the paper data locally, and converts the format if necessary.

[0415] 2. Analysis of paper data

[0416] The acquired paper data is tokenized by the server. For example, PDF format data is subjected to text extraction using the PyMuPDF library. This text data is then input into a generative AI model, such as GPT-3, for analysis. The generative AI uses the specified prompt to analyze the content of the paper and extract key points.

[0417] 3. Summary Generation

[0418] The server generates a summary of the paper based on the analysis results of the generative AI. This summary is a concise summary of the main points of the paper, allowing reviewers to quickly grasp the overall picture of the paper.

[0419] 4. Providing peer review support information

[0420] From the generated abstract, the server generates specific peer review support information, including ratings of novelty, logic, and relevance of the content, allowing reviewers to quickly determine which parts of the paper are worthy of evaluation.

[0421] 5. Viewing peer review support information

[0422] Finally, the terminal displays the peer review support information sent from the server to the user. The user interface is designed to be visually easy to understand, allowing users to easily access and understand the information. For example, an interface using a web browser is used.

[0423] Specific examples

[0424] Consider a case where a user reviews a paper located at "https: / / example.com / sample_paper.pdf." First, the server accesses the URL and retrieves the paper data. Next, it analyzes this data using the generative AI model "GPT-3" and uses the following prompt:

[0425] "Please write an abstract of this paper. Include the main points of your conclusions, an outline of your methodology, and any key findings from your results."

[0426] Based on the generated summary, the server creates peer review support information on the paper's novelty, logic, and content consistency, which is displayed on the terminal, allowing users to efficiently proceed with the peer review.

[0427] This system will significantly improve the efficiency of the peer review process, enabling the rapid evaluation of academic research. Furthermore, it is capable of handling multilingual processing and the acceptance or rejection of patent documents, and is expected to be applied in a wide range of fields.

[0428] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0429] Step 1:

[0430] Acquisition of paper data

[0431] The server receives a URL specified by the user as input. It accesses the specified URL, for example, "https: / / example.com / sample_paper.pdf," and downloads the paper data using the HTTP protocol. During this process, the server uses the requests library to send an HTTP request and saves the retrieved data locally. The output is the paper data saved in the server's temporary memory or storage.

[0432] Step 2:

[0433] Tokenization and formatting

[0434] The server receives the paper data obtained in step 1 as input. If the paper data is in PDF format, it extracts the text using a library such as PyMuPDF and tokenizes the data. Specifically, it extracts the text of each page from the PDF file and splits it into a series of words and sentences as text data. The output is text data that can be analyzed.

[0435] Step 3:

[0436] Analysis by generative AI

[0437] The server receives the text data generated in step 2 as input. It uses a generative AI model, such as GPT-3, to analyze the paper data. It provides the generative AI with the prompt, "Write a summary of this paper. Include the main points of the conclusion, an overview of the methodology, and important findings from the results." and performs the analysis. The output is the extracted main points and summary text.

[0438] Step 4:

[0439] Generate a summary

[0440] The server receives as input the analysis results generated in step 3. From the analysis results generated, it creates a summary of the paper. The summary is a concise summary of the main sections, such as conclusions, methodology, and results. The summary is stored in a data structure on the server. The output is a summary of the completed paper.

[0441] Step 5:

[0442] Providing peer review support information

[0443] The server receives the abstract generated in step 4 as input. Based on the abstract, it creates peer review support information regarding novelty, logic, and relevance of the content. This information includes evaluations such as "This paper is innovative compared to previous research," "The logical development is consistent and the conclusion is persuasive," and "The content is highly relevant to the research objectives." The output is specific peer review support information.

[0444] Step 6:

[0445] View peer review support information

[0446] The terminal receives as input the peer review support information generated in step 5. It displays this information to the user through a visually easy-to-understand user interface. For example, it formats the information using HTML and CSS in the browser, providing a display format that is easily accessible and understandable to the user. The output is a screen displaying the peer review support information that the user can view.

[0447] Through these steps, the system will be able to streamline the peer review process and support the rapid evaluation of academic research.

[0448] (Application example 1)

[0449] 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."

[0450] Traditional peer review processes are often manual and require a great deal of time and effort. Furthermore, the lack of uniform evaluation criteria makes it difficult to obtain consistent peer review results. Furthermore, when multilingual processing is required, language barriers exist, limiting the access and sharing of information in a field of knowledge. This creates a challenge for researchers and students to quickly grasp the latest research information.

[0451] 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.

[0452] In this invention, the server includes: means for acquiring paper data; means for tokenizing and analyzing the paper data using a generative AI; means for generating summaries from the analysis results of the generative AI; means for providing peer review support information based on the generated summaries; means for generating prompts for evaluating the novelty, logic, and content consistency of the paper based on the summaries in generating the peer review support information; and means for displaying the peer review support information through a user interface. This automates the peer review process, enabling efficient and consistent peer review. Furthermore, multilingual processing is possible, allowing researchers and students to quickly and easily access international research information.

[0453] "Paper data" refers to documents that describe academic research content and are stored electronically.

[0454] "Generative AI" refers to artificial intelligence technology that uses machine learning and deep learning to analyze and generate natural language.

[0455] "Tokenization" refers to the process of breaking down text into smaller units such as words or phrases.

[0456] "Analysis" refers to the process of understanding the acquired data and grasping its meaning and structure.

[0457] "Generative AI" refers to artificial intelligence that has the ability to generate new information or summaries based on data.

[0458] An "abstract" is a short sentence or paragraph that succinctly summarizes the contents of the entire document.

[0459] "Peer review support information" is specific information provided to assist in the evaluation of a paper, including an assessment of the paper's importance and quality.

[0460] A "prompt sentence" refers to an input sentence that causes a generation AI to generate a specific output.

[0461] "User interface" refers to the interface through which a system and a user communicate with each other.

[0462] In this invention, the server acquires and analyzes paper data using the following means to generate peer review support information.

[0463] 1. Acquisition of paper data

[0464] The server retrieves paper data from a URL or database. The hardware used for this is a cloud server, and the software used is an "HTTP request library" (e.g., Retrofit, OkHttp). The server accesses the URL specified by the user and downloads the paper data. For example, if the user specifies a paper at "https: / / example.com / sample_paper.pdf," the server retrieves the paper data from that URL.

[0465] 2. Analysis of paper data

[0466] The server tokenizes the acquired paper data and analyzes it using generative AI. The hardware used is a cloud server, and the software used is a "natural language processing model" (e.g., GPT-4, BERT). Tokenization breaks the paper data down into small units (words and phrases) that are then analyzed by generative AI.

[0467] 3. Summary Generation

[0468] The server generates a summary from the analysis results of the generative AI. The generative AI model used (e.g., BART, a large-scale language model) is used to summarize the analysis results as short sentences or paragraphs. For example, a BART model is used to generate a summary that concisely summarizes the main points of a paper.

[0469] 4. Generating peer review support information

[0470] Based on the generated abstract, we provide peer review support information, including generating prompts to evaluate the novelty, logic, and relevance of the content of the paper based on the abstract. For example, the generated prompts are as follows:

[0471] "Based on the following paper abstract, please rate the paper's novelty, logic, and relevance to the content. Abstract: [generated abstract content]"

[0472] 5. Display in the user interface

[0473] The device displays the generated peer review support information to the user through a user interface. The hardware used is a smartphone or computer, and the software is a "UI framework" (e.g., Flutter, React Native). Based on this information, the user can review papers efficiently and accurately.

[0474] Program execution example

[0475] For example, if a user reviews a paper located at "https: / / example.com / sample_paper.pdf," the server accesses the URL and retrieves the paper data. Next, it uses GPT-4 to analyze the paper and generates a summary using the BART model. Based on the summary, it inputs a prompt to GPT-3.5-turbo to generate specific review support information, which is then displayed on the smartphone's user interface.

[0476] As a result, this invention automates the peer review process, making it possible to provide fast and consistent peer review. Furthermore, its multilingual capabilities make it easier to access international research information.

[0477] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0478] Step 1:

[0479] The server retrieves the paper data from a URL specified by the user or from a database. The input is a URL provided by the user, and an HTTP request is made based on this to download the paper data. Specifically, the server sends a request using an HTTP request library (e.g., Retrofit, OkHttp) and retrieves PDF or text data as a response. The output is the retrieved paper data.

[0480] Step 2:

[0481] The server tokenizes and analyzes the acquired paper data. The input is the paper data acquired in step 1, which is tokenized using generative artificial intelligence. Specifically, it uses a natural language processing library (e.g., SpaCy, NLTK) to split (tokenize) the paper data into words and phrases. It then analyzes the content using a model such as GPT-4 or BERT. The output is the analyzed data (the tokenized data and its analysis results).

[0482] Step 3:

[0483] The server generates a summary from the analysis results. The input is the analysis result data obtained in step 2, which is input into a summary generation model (e.g., BART, a large-scale language model). The specific operation is to use a summary generation algorithm to extract the main points from the content of the entire paper and generate a concise summary. The output is the generated summary.

[0484] Step 4:

[0485] The server provides peer review support information based on the generated summary. The input is the summary generated in step 3, which is used to generate a prompt and provide evaluation information for peer review. Specifically, it uses a model such as GPT-3.5-turbo to generate a prompt to evaluate the paper's novelty, logic, and content relevance based on the summary, and then generates specific peer review support information based on the prompt. The output is peer review support information. For example, the generated prompt is as follows:

[0486] "Based on the following paper abstract, please rate the paper's novelty, logic, and relevance to the content. Abstract: [generated abstract content]"

[0487] Step 5:

[0488] The terminal displays the generated peer review support information to the user through a user interface. The input is the peer review support information generated in step 4, and a UI framework (e.g., Flutter, React Native) is used to display this information in an easy-to-understand manner for the user. The specific operation is to create a layout for visually displaying the peer review support information and display it on the screen. The output is the peer review support information displayed in a format that the user can view.

[0489] 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.

[0490] The present invention relates to a system that automates the peer review of papers and provides efficient peer review support. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide a more personalized peer review experience. The following describes embodiments of the present invention.

[0491] System configuration

[0492] 1. Data Acquisition Module: The server uses this module to acquire paper data from the Internet or databases. It downloads the paper data from the URL specified by the user and converts the format if necessary.

[0493] 2. Generative AI module: The server uses generative AI to tokenize and analyze the acquired paper data. The generative AI uses a natural language processing model to understand the content of the paper and extract key points.

[0494] 3. Summary generation module: The server generates a summary of the paper based on the analysis results of the AI. The summary is a concise summary of the main points of the paper, allowing reviewers to quickly grasp the overall picture.

[0495] 4. Peer review support module: The server provides specific peer review support information based on the generated abstract, including an evaluation of the paper's novelty, logic, and content consistency.

[0496] 5. Emotion Engine: The emotion engine installed on the server analyzes the user's facial expressions, voice, keyboard input, etc. to recognize the user's emotional state. Based on this information, the peer review support information is dynamically adjusted.

[0497] 6. User interface module: The terminal displays peer review support information to the user. The user interface is designed to be visually easy to understand, allowing users to easily access and understand the information.

[0498] Program processing flow

[0499] 1. Acquisition of paper data:

[0500] The user enters the URL of the paper they wish to review and sends it to the server via their device. The server then retrieves the paper data from the specified URL.

[0501] 2. Analysis of paper data:

[0502] The server tokenizes the acquired paper data and analyzes it using a generative AI, which analyzes each part of the paper to understand its main points and meaning.

[0503] 3. Generate a summary:

[0504] The server generates a summary from the analysis results of the AI. The summary provides important information to reviewers in a concise manner, streamlining the peer review process.

[0505] 4. Providing peer review support information:

[0506] Based on the server-generated abstract, the system provides specific peer review support information, including an assessment of the paper's novelty, logic, and content relevance.

[0507] 5. Emotion Recognition with Emotion Engine:

[0508] The server recognizes the user's emotions from their facial expressions, voice, keyboard input, etc. The emotion engine analyzes this data and understands the user's current emotional state.

[0509] 6. Dynamic adjustment according to emotions:

[0510] The server adjusts the presentation and design of peer review support information according to the user's emotional state. For example, if the user is feeling stressed, the server will simplify the presentation of information.

[0511] 7. View peer review support information:

[0512] The terminal displays the adjusted peer review support information to the user, who then performs the peer review efficiently based on the provided information.

[0513] Usage example

[0514] For example, if a user reviews a paper located at "https: / / example.com / sample_paper.pdf," the server accesses the URL to retrieve the paper data. It then uses generative artificial intelligence to analyze the paper and generate a summary. It then creates review support information based on the generated summary and uses an emotion engine to recognize the user's emotional state. Finally, it adjusts the information according to the emotion and provides it to the user via their device. The user can use the provided support information to quickly and accurately review the paper.

[0515] This allows our system to significantly streamline the review process, and when combined with emotion recognition, it can provide a more personalized review experience.

[0516] The processing flow will be explained below.

[0517] The present invention relates to a system that automates the peer review of papers and provides efficient peer review support. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide a more personalized peer review experience. Below, an embodiment of the present invention will be described using specific processing steps.

[0518] Step 1:

[0519] The user enters the URL of the paper they wish to review and sends it to the server via their device. The user then enters the URL into a dedicated interface and issues a command to retrieve the paper data.

[0520] Step 2:

[0521] The server retrieves the paper data from the specified URL. The server sends an HTTP request, receives the paper data as a response, and temporarily stores the data in memory. If the retrieval fails, an error message is generated to notify the user.

[0522] Step 3:

[0523] The server tokenizes the acquired paper data and converts it into an input format for the generation AI. Tokenization is the process of dividing text data into analyzable units.

[0524] Step 4:

[0525] The server analyzes the tokenized paper data using a generative AI, which uses a natural language processing model to understand the content of the paper and extract key points.

[0526] Step 5:

[0527] The server generates a summary from the analysis results of the AI. Based on the analysis results, a summary that concisely summarizes the main points of the paper is created. The summary is important information that allows reviewers to quickly grasp the overall picture of the paper.

[0528] Step 6:

[0529] The server generates abstracts and provides peer review support information, including an assessment of the paper's novelty, logic, and content consistency.

[0530] Step 7:

[0531] The emotion engine installed on the server analyzes the user's facial expressions, voice, keyboard input, etc. to recognize the user's emotional state. The emotion engine uses this data to understand the user's current emotional state.

[0532] Step 8:

[0533] The server dynamically adjusts the presentation and design of peer review support information according to the user's perceived emotional state: for example, if the user is feeling stressed, it will simplify the information or add an encouraging message.

[0534] Step 9:

[0535] The terminal displays tailored peer review support information to the user. The user interface is designed to be visually intuitive, allowing users to easily access and understand the information.

[0536] Step 10:

[0537] Users can check the peer review support information provided through the device interface and evaluate papers based on that information, allowing them to efficiently carry out their peer review work.

[0538] Step 11:

[0539] The user enters the peer review results and sends them to the server via their terminal. The final peer review results are stored in the system and used to make decisions on whether to accept or reject the paper.

[0540] Through this series of processes, the system of the present invention significantly streamlines the peer review process and supports rapid evaluation of academic research. Furthermore, by combining it with an emotion engine, it is possible to provide users with a more appropriate and personalized peer review experience.

[0541] Example 2

[0542] 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."

[0543] The traditional paper peer review process is time-consuming and places a heavy burden on reviewers. This burden is particularly pronounced when the paper is written in multiple languages ​​or when the reviewer's emotional state affects their work efficiency. Furthermore, traditional systems provide static peer review support information and are not sufficiently personalized to meet the needs of individual users. There is a need for a system that can solve these problems and provide efficient and personalized paper peer review support.

[0544] 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.

[0545] In this invention, the server includes means for acquiring paper data, means for tokenizing and analyzing the paper data using a generating AI, means for generating a summary from the analysis result of the generating AI, means for providing peer review support information based on the generated summary, means for recognizing a user's emotional state, means for dynamically adjusting the peer review support information based on the recognized emotional state, and means for displaying the peer review support information through a user interface, thereby enabling efficient and personalized paper peer review support.

[0546] "Means for obtaining paper data" refers to a device or function for downloading paper data from a specified URL and converting the format as necessary.

[0547] "Means for tokenizing and analyzing paper data using generative artificial intelligence" refers to a device or function that uses a natural language processing model to divide paper data into individual tokens and analyze their contents.

[0548] "Means for generating summaries from the analysis results of generative artificial intelligence" refers to a device or function that generates a summary that succinctly summarizes the main points of a paper based on the analyzed data.

[0549] "Means for providing peer review support information based on the generated abstract" refers to a device or function that provides peer review support information such as the novelty, logic, and content consistency of a paper based on the generated abstract.

[0550] The "means for recognizing the emotional state of the user" refers to a device or function that analyzes emotions from the user's facial expressions, voice, keyboard input, etc., and recognizes that emotional state.

[0551] "Means for dynamically adjusting peer review support information based on a recognized emotional state" refers to a device or function that dynamically adjusts the display method and content of peer review support information in accordance with the recognized emotional state of the user.

[0552] A "means for displaying peer review support information through a user interface" is an interface designed to visually display the tailored peer review support information and to allow easy user access.

[0553] The present invention relates to a system for automating the peer review of papers and providing efficient and personalized peer review support to users. The system functions as follows through interactions between a server, a terminal, and a user.

[0554] The system includes the following functions: acquiring paper data, analyzing it using generative artificial intelligence, generating summaries, providing peer review support information, recognizing the user's emotional state and dynamically adjusting information, and displaying information through a user interface.

[0555] Acquisition of paper data

[0556] First, the user enters the URL of the paper to be peer-reviewed. The device sends this URL to the server, and the server accesses the specified URL to retrieve the paper data. The retrieved paper data is in PDF format, for example, and the server converts it into an appropriate format as needed. This format conversion is performed using an OCR tool (e.g., Tesseract).

[0557] Analysis using generative artificial intelligence

[0558] The server tokenizes the acquired paper data using natural language processing libraries (e.g., NLTK, SpaCy). This tokenized data is then fed into a generative AI model (e.g., GPT-4) that analyzes each part of the paper and extracts key points. This analysis process clarifies the paper's subject, purpose, results, etc.

[0559] Generate a summary

[0560] The server uses topic modeling and summary generation algorithms to generate summaries based on the analysis results obtained from the generative AI model. These summaries provide a compact summary of important information for reviewers, allowing for quick understanding.

[0561] Providing peer review support information

[0562] Based on the generated summary, the server generates peer review support information, including an evaluation of the paper's novelty, logic, and content consistency. The server scores the paper based on these criteria and uses a rule-based engine (e.g., if-then rules) to generate specific advice and comments.

[0563] Recognizing the user's emotional state

[0564] The server analyzes the user's facial expressions, voice, keyboard input, etc. collected through the device and uses an emotion engine to recognize the user's emotional state. For example, facial expression analysis and voice emotion recognition technology are used to determine whether the user is feeling stressed.

[0565] Dynamic adjustment of information

[0566] Based on the user's recognized emotional state, the server dynamically adjusts the display method and content of peer review support information. For example, if the user is feeling stressed, the server will adjust the display to show only the most important information in a concise manner.

[0567] Displaying Information Through a User Interface

[0568] Finally, the adjusted peer review support information is displayed on the user interface on the device. The user interface has an intuitive and easy-to-use design, allowing users to easily check the information. Users can utilize the provided support information to review papers quickly and accurately.

[0569] Prompt Sentence Examples

[0570] An example of a prompt that our system could input to the generative AI model would be:

[0571] Generate an abstract for the following paper:

[0572] URL: https: / / example.com / sample_paper.pdf

[0573] Summarize the contents of the paper and state the main points and conclusions.

[0574] This allows our system to significantly streamline the review process and, when combined with emotion recognition, provide a more personalized review experience.

[0575] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0576] Step 1:

[0577] The user enters the URL of the paper they wish to review, and the device sends the URL to the server.

[0578] Input: URL of the paper to be reviewed

[0579] Output: Request for paper data from server

[0580] Specific operation: The user enters a URL into the input field of the terminal and clicks the send button. The terminal sends this URL as an HTTP request to the server.

[0581] Step 2:

[0582] The server accesses the specified URL and retrieves the paper data.

[0583] Input: URL of the paper sent from the device

[0584] Output: Retrieved paper data (PDF format, etc.)

[0585] Specific operation: The server accesses the URL, downloads the paper data in PDF format, and temporarily stores it on the server.

[0586] Step 3:

[0587] The server tokenizes the acquired paper data and analyzes it using generative AI.

[0588] Input: Retrieved paper data

[0589] Output: Tokenized text data and analysis results

[0590] How it works: The server uses a natural language processing library (e.g., NLTK, SpaCy) to extract and tokenize text from the PDF, then feeds the tokenized data into a generative AI model (e.g., GPT-4) to parse out key points.

[0591] Step 4:

[0592] The server generates a summary from the analysis results of the generation AI.

[0593] Input: Analysis results by the generation AI

[0594] Output: Abstract text of the paper

[0595] Specific operation: The server uses topic modeling and summary generation algorithms to generate summaries from the analysis results, which are then stored in an internal database.

[0596] Step 5:

[0597] The server generates peer review support information based on the abstract.

[0598] Input: Generated summary

[0599] Output: Peer review support information (evaluation of novelty, logic, and consistency)

[0600] Specific Actions: The server analyzes the summary, calculates a score based on the evaluation criteria, and uses a rule-based engine to generate specific advice and comments.

[0601] Step 6:

[0602] The server recognizes the user's emotional state.

[0603] Input: User facial expression, voice, and keyboard input data collected from the device

[0604] Output: User's emotional state data

[0605] Specific operation: The device's camera and microphone collect the user's facial expressions and voice, and send them to the server. The server then uses emotion recognition technology to analyze the user's emotional state.

[0606] Step 7:

[0607] The server dynamically adjusts the peer review support information based on the perceived emotional state.

[0608] Input: User's emotional state data

[0609] Output: Adjusted peer review supporting information

[0610] Specific operation: The server receives the results of the emotion engine and adjusts the display method and content of the information. For example, if the user is feeling stressed, it will simply display only the important information.

[0611] Step 8:

[0612] The terminal displays the adjusted peer review support information on the user interface.

[0613] Input: Coordinated peer review supporting information

[0614] Output: Information displayed in the user interface

[0615] Specific operation: The terminal displays the adjusted peer review support information received from the server on the user interface. The user uses this information to perform the peer review efficiently.

[0616] The above is the specific processing flow of the program of this system.

[0617] (Application example 2)

[0618] 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."

[0619] Conventional paper peer review systems make it difficult for reviewers to efficiently review a large number of papers. Furthermore, because information is provided uniformly without considering the reviewer's emotional state, stress and dissatisfaction are likely to occur. To address these issues, a system that recognizes the user's emotions and provides dynamically adjusted peer review support information is needed.

[0620] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring paper data, means for tokenizing and analyzing the paper data using a generation AI, means for generating a summary from the analysis results of the generation AI, means for providing peer review support information based on the generated summary, means for displaying the peer review support information through a user interface, means for recognizing user emotions, and means for dynamically adjusting the peer review support information based on the user emotions. This allows reviewers to review papers efficiently and also allows them to receive personalized information according to their own emotional state.

[0621] "Paper data" refers to scientific or academic documents written by researchers or authors, and is material that presents specific research results or findings.

[0622] "Generative AI" refers to an AI technology that performs advanced processing such as natural language processing, tokenization, analysis, and summary generation based on large datasets.

[0623] "Tokenization" is the process of breaking down text into smaller units such as words and phrases in order to analyze it.

[0624] "Analysis" refers to the process of understanding the content of acquired data and extracting important information.

[0625] A summary is a concise summary of the main points of an entire text.

[0626] "Peer review support information" refers to information to assist reviewers in their work, such as evaluations of the paper's novelty, logic, and content consistency, as well as specific feedback.

[0627] A "user interface" is a means by which a user interacts with a computer system, and refers to a screen on which information is visually displayed and an operating means.

[0628] "User emotions" refer to the psychological states such as stress, excitement, and concentration that users experience when reviewing papers.

[0629] "Dynamic adjustment" refers to the process of changing the content and presentation of information in real time according to the user's emotions and situation.

[0630] The system that realizes this application example is composed of a series of modules that make up the entire system. The detailed operation of each module is shown below.

[0631] Program processing

[0632] Acquisition of paper data

[0633] First, the server retrieves the paper data from the URL specified by the user. This retrieved paper data is temporarily stored for subsequent analysis. Specifically, the server uses a network library (e.g., the requests library) to download the paper data via the Internet.

[0634] Analysis of paper data

[0635] The server then uses a generative AI model to tokenize the acquired paper data and analyze its content. The generative AI uses a natural language processing library (e.g., the Transformers library). Key points and important information are extracted from the analysis results, which then serve as the basis for generating summaries, as described below.

[0636] Generate a summary

[0637] The server generates a summary based on the analyzed data. The generated summary is intended to allow reviewers to quickly grasp the overall picture of the paper, concisely summarizing only the most important points. The summary is generated using the generative AI model mentioned above.

[0638] Emotion recognition by emotion engine

[0639] Furthermore, the server is equipped with an emotion engine that analyzes the user's facial expressions, voice, keyboard input, etc. to recognize their emotional state. This emotion data is then used to dynamically adjust the review support information. The emotion engine uses the dlib library, which provides facial recognition technology, and hardware such as a webcam.

[0640] Dynamic Adjustment

[0641] The server dynamically adjusts the peer review support information it provides based on the user's recognized emotional state. For example, if the user is feeling stressed, it displays information in a simplified manner to reduce the burden on the user. This allows the user to review efficiently in a relaxed state.

[0642] User Interface Display

[0643] Finally, the adjusted peer review support information is displayed on the user's device. The user interface is designed to be visually easy to understand, allowing users to quickly access the information they need. This display can be done using a web application framework (e.g., Flask or Django).

[0644] Specific examples

[0645] For example, if a user enters a URL such as "https: / / example.com / sample_paper.pdf" to review a paper, the server accesses the URL to retrieve the paper data. The retrieved data is analyzed using a generative AI model to generate a summary. The system then creates peer review support information based on the generated summary and uses an emotion engine to recognize the user's emotional state. Finally, the system adjusts the information according to the user's emotions and provides it to the user through a user interface.

[0646] An example of a prompt could be a question such as "What are the main points?", which allows the generative AI model to extract key points from the paper and support efficient peer review.

[0647] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0648] Step 1:

[0649] The user enters the URL of the paper they wish to review and sends it to the server via their terminal. When the server receives this URL, it uses a network library (for example, the requests library) to retrieve the paper data from the URL. The retrieved paper data is temporarily stored on the server. The input is the "URL of the paper" and the output is the "paper data."

[0650] Step 2:

[0651] The server tokenizes and analyzes the acquired paper data using a generative AI model (e.g., the transformers library). This process understands the structure and content of the entire paper and extracts key points. The input is "paper data" and the output is "analysis results."

[0652] Step 3:

[0653] The server generates a summary based on the analysis results. The generated summary is intended to allow reviewers to quickly grasp the overall picture of the paper. This summary generation is also performed using a generative AI model. The input is the "analysis results" and the output is the "summary."

[0654] Step 4:

[0655] The server generates peer review support information based on the abstract. This information includes an evaluation of the paper's novelty, logic, and content consistency. The input is the "abstract" and the output is the "peer review support information."

[0656] Step 5:

[0657] The server uses an emotion engine to analyze the user's facial expressions, voice, and keyboard input to recognize their emotional state. The emotion engine uses the dlib library, which provides facial recognition technology, and hardware such as a webcam. The input is "user's emotional data (facial expressions, voice, and input)" and the output is "recognized emotional state."

[0658] Step 6:

[0659] The server dynamically adjusts the peer review support information based on the recognized emotional state. For example, if the user is feeling stressed, it simplifies the information and makes the display easier to understand. The inputs are the "recognized emotional state" and "peer review support information," and the output is "adjusted peer review support information."

[0660] Step 7:

[0661] The terminal displays the adjusted peer review support information to the user. The information is provided in a visually easy-to-understand format through the user interface. The input is the adjusted peer review support information, and the output is the display to the user.

[0662] 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.

[0663] 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.

[0664] 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.

[0665] [Third embodiment]

[0666] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0667] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0668] 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).

[0669] 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.

[0670] 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.

[0671] 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).

[0672] 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.

[0673] 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.

[0674] 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.

[0675] 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.

[0676] 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.

[0677] 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."

[0678] The present invention relates to a system for automating the peer review of academic papers and providing efficient peer review support. This system utilizes generative artificial intelligence to acquire and analyze paper data, generate summaries, and provide peer review support information.

[0679] System configuration

[0680] 1. Data Acquisition Module: The server uses this module to acquire paper data from the Internet or databases. It automatically downloads the paper data from the URL or path specified by the user and converts the format as necessary.

[0681] 2. Generative AI module: The server uses generative AI to tokenize and analyze the acquired paper data. The generative AI uses natural language processing models, for example, to understand the content of the paper and extract key points.

[0682] 3. Summary generation module: The server generates a summary of the paper based on the analysis results of the AI. The generated summary is a concise summary of the main points of the paper, allowing reviewers to quickly grasp the overall picture of the paper.

[0683] 4. Peer review support module: Based on the generated abstract, the server provides specific peer review support information, including an evaluation of the paper's novelty, logic, and content consistency.

[0684] 5. User Interface Module: The terminal displays peer review support information to the user. This interface is designed to be visually easy to understand, allowing users to easily access and understand the information.

[0685] Program processing flow

[0686] 1. Acquisition of paper data: The server acquires the paper data from the URL or database specified by the user. This acquired data is temporarily stored in the server's memory.

[0687] 2. Analyzing the paper data: The server tokenizes the acquired data and analyzes it using a generative AI, which understands the main parts of the paper and summarizes its contents.

[0688] 3. Summary generation: The server creates a summary from the analysis results of the AI. This summary concisely presents important information to reviewers, streamlining the peer review process.

[0689] 4. Providing peer review support information: The server generates peer review support information based on the abstract, including an evaluation of the paper's novelty and logic.

[0690] 5. Display of peer review support information: The terminal displays peer review support information to the user, allowing the user to review papers efficiently based on this information.

[0691] Usage example

[0692] For example, if a user reviews a paper located at "https: / / example.com / sample_paper.pdf," the server accesses the URL and retrieves the paper data. It then uses generative AI to analyze the paper and generate a summary. It then creates peer review support information based on the generated summary and provides it to the user via their device. The user can use the provided support information to quickly and accurately review the paper.

[0693] This system will significantly improve the efficiency of the peer review process and support the rapid evaluation of academic research. Furthermore, since it can handle multilingual processing and patent document acceptance / rejection, it is expected to be applied in a wide range of fields.

[0694] The processing flow will be explained below.

[0695] Step 1:

[0696] The user inputs the URL of the paper they wish to review and sends it to the server via their device. The user then inputs this URL into a dedicated interface and issues a request to retrieve it.

[0697] Step 2:

[0698] The server retrieves the paper data from the specified URL. The server sends an HTTP request and retrieves the paper data as a response. If the data retrieval fails, the server generates an error message and notifies the user.

[0699] Step 3:

[0700] The server tokenizes the acquired paper data and converts it into an input format for the generative AI, which is a preprocessing step to enable the generative AI to accurately analyze the data.

[0701] Step 4:

[0702] The server uses a generative AI to analyze the tokenized paper data, which analyzes each part of the paper and understands its main points and meaning.

[0703] Step 5:

[0704] The server generates a summary from the AI ​​analysis results, extracting the most important information from the analysis results and concisely summarizing the main points of the paper.

[0705] Step 6:

[0706] The server provides peer review support information based on the abstracts generated by the server, which then generates feedback including evaluations of the paper's novelty, logic, and content consistency.

[0707] Step 7:

[0708] The server sends the generated peer review support information to the terminal, which displays the received information on a user interface for the reviewer to use.

[0709] Step 8:

[0710] Users can check the peer review support information provided through the device interface, evaluate the paper based on this information, and efficiently proceed with the peer review process.

[0711] Step 9:

[0712] The user enters the peer review results and sends them to the server via their terminal. The final peer review results are stored in the system and used to make decisions on whether to accept or reject the paper.

[0713] Through this series of processes, the system streamlines the paper peer review process and supports the rapid evaluation of academic research.

[0714] Example 1

[0715] 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."

[0716] The current peer review process is time-consuming and labor-intensive, making it difficult to evaluate many papers in a short period of time. Furthermore, there is a lack of consistency in the evaluation of papers' novelty, logic, and content consistency, resulting in inconsistent evaluations among reviewers. Therefore, there is a need for a system that streamlines the peer review process and improves the quality of evaluations.

[0717] 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.

[0718] In this invention, the server includes means for acquiring paper data from the Internet or a database, means for tokenizing the acquired paper data using a generative artificial intelligence and analyzing it using a natural language processing model, means for extracting important points from the analysis results of the generative artificial intelligence and generating a paper summary, means for providing peer review support information including an evaluation of the paper's novelty, logic, and content consistency based on the generated summary, and means for displaying the peer review support information through a user interface, thereby making it possible to streamline the peer review process and improve the quality of evaluation.

[0719] The "Internet" is a global communications network for accessing distributed information resources and retrieving, transmitting, and sharing data through computer networks.

[0720] A "database" is a system for managing data efficiently and systematically, allowing data to be searched, retrieved, and updated using specific queries.

[0721] "Generative AI" is an AI technology that has the ability to automatically handle specific tasks, learn like a human, analyze data, and generate deliverables.

[0722] "Tokenization" is the process of dividing text data into the smallest meaningful units, such as words and sentences, and is a preprocessing step in natural language processing.

[0723] A "natural language processing model" is a machine learning model for understanding, generating, and manipulating natural language, enabling computers to interpret human language.

[0724] A "summary" is a short, concise summary of the main points and conclusions from a longer document.

[0725] "Novelty" is a concept that indicates that a particular research or invention is something new that has not existed before.

[0726] "Logicality" refers to whether an argument or conclusion is based on a consistent logical structure, and is a standard for evaluating whether it is logically sound.

[0727] "Content relevance" is a concept that evaluates whether the content of a document is appropriate to its purpose or theme and meets the expected items and standards.

[0728] A "user interface" is a means by which a user interacts with a computer system, providing a visually easy-to-understand display of information and a method of operation.

[0729] This invention relates to a system that automates and improves the efficiency of the paper peer review process by using a generative AI model to acquire paper data, analyze it, generate summaries, and provide and display peer review support information.

[0730] 1. Acquisition of paper data

[0731] First, the user specifies the URL of the paper data they want to retrieve from the Internet or a database. The server accesses the specified URL and downloads the paper data via the HTTP protocol. During this process, the server uses the Python requests library to send an HTTP request, saves the paper data locally, and converts the format if necessary.

[0732] 2. Analysis of paper data

[0733] The acquired paper data is tokenized by the server. For example, PDF format data is subjected to text extraction using the PyMuPDF library. This text data is then input into a generative AI model, such as GPT-3, for analysis. The generative AI uses the specified prompt to analyze the content of the paper and extract key points.

[0734] 3. Summary Generation

[0735] The server generates a summary of the paper based on the analysis results of the generative AI. This summary is a concise summary of the main points of the paper, allowing reviewers to quickly grasp the overall picture of the paper.

[0736] 4. Providing peer review support information

[0737] From the generated abstract, the server generates specific peer review support information, including ratings of novelty, logic, and relevance of the content, allowing reviewers to quickly determine which parts of the paper are worthy of evaluation.

[0738] 5. Viewing peer review support information

[0739] Finally, the terminal displays the peer review support information sent from the server to the user. The user interface is designed to be visually easy to understand, allowing users to easily access and understand the information. For example, an interface using a web browser is used.

[0740] Specific examples

[0741] Consider a case where a user reviews a paper located at "https: / / example.com / sample_paper.pdf." First, the server accesses the URL and retrieves the paper data. Next, it analyzes this data using the generative AI model "GPT-3" and uses the following prompt:

[0742] "Please write an abstract of this paper. Include the main points of your conclusions, an outline of your methodology, and any key findings from your results."

[0743] Based on the generated summary, the server creates peer review support information on the paper's novelty, logic, and content consistency, which is displayed on the terminal, allowing users to efficiently proceed with the peer review.

[0744] This system will significantly improve the efficiency of the peer review process, enabling the rapid evaluation of academic research. Furthermore, it is capable of handling multilingual processing and the acceptance or rejection of patent documents, and is expected to be applied in a wide range of fields.

[0745] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0746] Step 1:

[0747] Acquisition of paper data

[0748] The server receives a URL specified by the user as input. It accesses the specified URL, for example, "https: / / example.com / sample_paper.pdf," and downloads the paper data using the HTTP protocol. During this process, the server uses the requests library to send an HTTP request and saves the retrieved data locally. The output is the paper data saved in the server's temporary memory or storage.

[0749] Step 2:

[0750] Tokenization and formatting

[0751] The server receives the paper data obtained in step 1 as input. If the paper data is in PDF format, it extracts the text using a library such as PyMuPDF and tokenizes the data. Specifically, it extracts the text of each page from the PDF file and splits it into a series of words and sentences as text data. The output is text data that can be analyzed.

[0752] Step 3:

[0753] Analysis by generative AI

[0754] The server receives the text data generated in step 2 as input. It uses a generative AI model, such as GPT-3, to analyze the paper data. It provides the generative AI with the prompt, "Write a summary of this paper. Include the main points of the conclusion, an overview of the methodology, and important findings from the results." and performs the analysis. The output is the extracted main points and summary text.

[0755] Step 4:

[0756] Generate a summary

[0757] The server receives as input the analysis results generated in step 3. From the analysis results generated, it creates a summary of the paper. The summary is a concise summary of the main sections, such as conclusions, methodology, and results. The summary is stored in a data structure on the server. The output is a summary of the completed paper.

[0758] Step 5:

[0759] Providing peer review support information

[0760] The server receives the abstract generated in step 4 as input. Based on the abstract, it creates peer review support information regarding novelty, logic, and relevance of the content. This information includes evaluations such as "This paper is innovative compared to previous research," "The logical development is consistent and the conclusion is persuasive," and "The content is highly relevant to the research objectives." The output is specific peer review support information.

[0761] Step 6:

[0762] View peer review support information

[0763] The terminal receives as input the peer review support information generated in step 5. It displays this information to the user through a visually easy-to-understand user interface. For example, it formats the information using HTML and CSS in the browser, providing a display format that is easily accessible and understandable to the user. The output is a screen displaying the peer review support information that the user can view.

[0764] Through these steps, the system will be able to streamline the peer review process and support the rapid evaluation of academic research.

[0765] (Application example 1)

[0766] 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."

[0767] Traditional peer review processes are often manual and require a great deal of time and effort. Furthermore, the lack of uniform evaluation criteria makes it difficult to obtain consistent peer review results. Furthermore, when multilingual processing is required, language barriers exist, limiting the access and sharing of information in a field of knowledge. This creates a challenge for researchers and students to quickly grasp the latest research information.

[0768] 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.

[0769] In this invention, the server includes: means for acquiring paper data; means for tokenizing and analyzing the paper data using a generative AI; means for generating summaries from the analysis results of the generative AI; means for providing peer review support information based on the generated summaries; means for generating prompts for evaluating the novelty, logic, and content consistency of the paper based on the summaries in generating the peer review support information; and means for displaying the peer review support information through a user interface. This automates the peer review process, enabling efficient and consistent peer review. Furthermore, multilingual processing is possible, allowing researchers and students to quickly and easily access international research information.

[0770] "Paper data" refers to documents that describe academic research content and are stored electronically.

[0771] "Generative AI" refers to artificial intelligence technology that uses machine learning and deep learning to analyze and generate natural language.

[0772] "Tokenization" refers to the process of breaking down text into smaller units such as words or phrases.

[0773] "Analysis" refers to the process of understanding the acquired data and grasping its meaning and structure.

[0774] "Generative AI" refers to artificial intelligence that has the ability to generate new information or summaries based on data.

[0775] An "abstract" is a short sentence or paragraph that succinctly summarizes the contents of the entire document.

[0776] "Peer review support information" is specific information provided to assist in the evaluation of a paper, including an assessment of the paper's importance and quality.

[0777] A "prompt sentence" refers to an input sentence that causes a generation AI to generate a specific output.

[0778] "User interface" refers to the interface through which a system and a user communicate with each other.

[0779] In this invention, the server acquires and analyzes paper data using the following means to generate peer review support information.

[0780] 1. Acquisition of paper data

[0781] The server retrieves paper data from a URL or database. The hardware used for this is a cloud server, and the software used is an "HTTP request library" (e.g., Retrofit, OkHttp). The server accesses the URL specified by the user and downloads the paper data. For example, if the user specifies a paper at "https: / / example.com / sample_paper.pdf," the server retrieves the paper data from that URL.

[0782] 2. Analysis of paper data

[0783] The server tokenizes the acquired paper data and analyzes it using generative AI. The hardware used is a cloud server, and the software used is a "natural language processing model" (e.g., GPT-4, BERT). Tokenization breaks the paper data down into small units (words and phrases) that are then analyzed by generative AI.

[0784] 3. Summary Generation

[0785] The server generates a summary from the analysis results of the generative AI. The generative AI model used (e.g., BART, a large-scale language model) is used to summarize the analysis results as short sentences or paragraphs. For example, a BART model is used to generate a summary that concisely summarizes the main points of a paper.

[0786] 4. Generating peer review support information

[0787] Based on the generated abstract, we provide peer review support information, including generating prompts to evaluate the novelty, logic, and relevance of the content of the paper based on the abstract. For example, the generated prompts are as follows:

[0788] "Based on the following paper abstract, please rate the paper's novelty, logic, and relevance to the content. Abstract: [generated abstract content]"

[0789] 5. Display in the user interface

[0790] The device displays the generated peer review support information to the user through a user interface. The hardware used is a smartphone or computer, and the software is a "UI framework" (e.g., Flutter, React Native). Based on this information, the user can review papers efficiently and accurately.

[0791] Program execution example

[0792] For example, if a user reviews a paper located at "https: / / example.com / sample_paper.pdf," the server accesses the URL and retrieves the paper data. Next, it uses GPT-4 to analyze the paper and generates a summary using the BART model. Based on the summary, it inputs a prompt to GPT-3.5-turbo to generate specific review support information, which is then displayed on the smartphone's user interface.

[0793] As a result, this invention automates the peer review process, making it possible to provide fast and consistent peer review. Furthermore, its multilingual capabilities make it easier to access international research information.

[0794] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0795] Step 1:

[0796] The server retrieves the paper data from a URL specified by the user or from a database. The input is a URL provided by the user, and an HTTP request is made based on this to download the paper data. Specifically, the server sends a request using an HTTP request library (e.g., Retrofit, OkHttp) and retrieves PDF or text data as a response. The output is the retrieved paper data.

[0797] Step 2:

[0798] The server tokenizes and analyzes the acquired paper data. The input is the paper data acquired in step 1, which is tokenized using generative artificial intelligence. Specifically, it uses a natural language processing library (e.g., SpaCy, NLTK) to split (tokenize) the paper data into words and phrases. It then analyzes the content using a model such as GPT-4 or BERT. The output is the analyzed data (the tokenized data and its analysis results).

[0799] Step 3:

[0800] The server generates a summary from the analysis results. The input is the analysis result data obtained in step 2, which is input into a summary generation model (e.g., BART, a large-scale language model). The specific operation is to use a summary generation algorithm to extract the main points from the content of the entire paper and generate a concise summary. The output is the generated summary.

[0801] Step 4:

[0802] The server provides peer review support information based on the generated summary. The input is the summary generated in step 3, which is used to generate a prompt and provide evaluation information for peer review. Specifically, it uses a model such as GPT-3.5-turbo to generate a prompt to evaluate the paper's novelty, logic, and content relevance based on the summary, and then generates specific peer review support information based on the prompt. The output is peer review support information. For example, the generated prompt is as follows:

[0803] "Based on the following paper abstract, please rate the paper's novelty, logic, and relevance to the content. Abstract: [generated abstract content]"

[0804] Step 5:

[0805] The terminal displays the generated peer review support information to the user through a user interface. The input is the peer review support information generated in step 4, and a UI framework (e.g., Flutter, React Native) is used to display this information in an easy-to-understand manner for the user. The specific operation is to create a layout for visually displaying the peer review support information and display it on the screen. The output is the peer review support information displayed in a format that the user can view.

[0806] 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.

[0807] The present invention relates to a system that automates the peer review of papers and provides efficient peer review support. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide a more personalized peer review experience. The following describes embodiments of the present invention.

[0808] System configuration

[0809] 1. Data Acquisition Module: The server uses this module to acquire paper data from the Internet or databases. It downloads the paper data from the URL specified by the user and converts the format if necessary.

[0810] 2. Generative AI module: The server uses generative AI to tokenize and analyze the acquired paper data. The generative AI uses a natural language processing model to understand the content of the paper and extract key points.

[0811] 3. Summary generation module: The server generates a summary of the paper based on the analysis results of the AI. The summary is a concise summary of the main points of the paper, allowing reviewers to quickly grasp the overall picture.

[0812] 4. Peer review support module: The server provides specific peer review support information based on the generated abstract, including an evaluation of the paper's novelty, logic, and content consistency.

[0813] 5. Emotion Engine: The emotion engine installed on the server analyzes the user's facial expressions, voice, keyboard input, etc. to recognize the user's emotional state. Based on this information, the peer review support information is dynamically adjusted.

[0814] 6. User interface module: The terminal displays peer review support information to the user. The user interface is designed to be visually easy to understand, allowing users to easily access and understand the information.

[0815] Program processing flow

[0816] 1. Acquisition of paper data:

[0817] The user enters the URL of the paper they wish to review and sends it to the server via their device. The server then retrieves the paper data from the specified URL.

[0818] 2. Analysis of paper data:

[0819] The server tokenizes the acquired paper data and analyzes it using a generative AI, which analyzes each part of the paper to understand its main points and meaning.

[0820] 3. Generate a summary:

[0821] The server generates a summary from the analysis results of the AI. The summary provides important information to reviewers in a concise manner, streamlining the peer review process.

[0822] 4. Providing peer review support information:

[0823] Based on the server-generated abstract, the system provides specific peer review support information, including an assessment of the paper's novelty, logic, and content relevance.

[0824] 5. Emotion Recognition with Emotion Engine:

[0825] The server recognizes the user's emotions from their facial expressions, voice, keyboard input, etc. The emotion engine analyzes this data and understands the user's current emotional state.

[0826] 6. Dynamic adjustment according to emotions:

[0827] The server adjusts the presentation and design of peer review support information according to the user's emotional state. For example, if the user is feeling stressed, the server will simplify the presentation of information.

[0828] 7. View peer review support information:

[0829] The terminal displays the adjusted peer review support information to the user, who then performs the peer review efficiently based on the provided information.

[0830] Usage example

[0831] For example, if a user reviews a paper located at "https: / / example.com / sample_paper.pdf," the server accesses the URL to retrieve the paper data. It then uses generative artificial intelligence to analyze the paper and generate a summary. It then creates review support information based on the generated summary and uses an emotion engine to recognize the user's emotional state. Finally, it adjusts the information according to the emotion and provides it to the user via their device. The user can use the provided support information to quickly and accurately review the paper.

[0832] This allows our system to significantly streamline the review process, and when combined with emotion recognition, it can provide a more personalized review experience.

[0833] The processing flow will be explained below.

[0834] The present invention relates to a system that automates the peer review of papers and provides efficient peer review support. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide a more personalized peer review experience. Below, an embodiment of the present invention will be described using specific processing steps.

[0835] Step 1:

[0836] The user enters the URL of the paper they wish to review and sends it to the server via their device. The user then enters the URL into a dedicated interface and issues a command to retrieve the paper data.

[0837] Step 2:

[0838] The server retrieves the paper data from the specified URL. The server sends an HTTP request, receives the paper data as a response, and temporarily stores the data in memory. If the retrieval fails, an error message is generated to notify the user.

[0839] Step 3:

[0840] The server tokenizes the acquired paper data and converts it into an input format for the generation AI. Tokenization is the process of dividing text data into analyzable units.

[0841] Step 4:

[0842] The server analyzes the tokenized paper data using a generative AI, which uses a natural language processing model to understand the content of the paper and extract key points.

[0843] Step 5:

[0844] The server generates a summary from the analysis results of the AI. Based on the analysis results, a summary that concisely summarizes the main points of the paper is created. The summary is important information that allows reviewers to quickly grasp the overall picture of the paper.

[0845] Step 6:

[0846] The server generates abstracts and provides peer review support information, including an assessment of the paper's novelty, logic, and content consistency.

[0847] Step 7:

[0848] The emotion engine installed on the server analyzes the user's facial expressions, voice, keyboard input, etc. to recognize the user's emotional state. The emotion engine uses this data to understand the user's current emotional state.

[0849] Step 8:

[0850] The server dynamically adjusts the presentation and design of peer review support information according to the user's perceived emotional state: for example, if the user is feeling stressed, it will simplify the information or add an encouraging message.

[0851] Step 9:

[0852] The terminal displays tailored peer review support information to the user. The user interface is designed to be visually intuitive, allowing users to easily access and understand the information.

[0853] Step 10:

[0854] Users can check the peer review support information provided through the device interface and evaluate papers based on that information, allowing them to efficiently carry out their peer review work.

[0855] Step 11:

[0856] The user enters the peer review results and sends them to the server via their terminal. The final peer review results are stored in the system and used to make decisions on whether to accept or reject the paper.

[0857] Through this series of processes, the system of the present invention significantly streamlines the peer review process and supports rapid evaluation of academic research. Furthermore, by combining it with an emotion engine, it is possible to provide users with a more appropriate and personalized peer review experience.

[0858] Example 2

[0859] 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."

[0860] The traditional paper peer review process is time-consuming and places a heavy burden on reviewers. This burden is particularly pronounced when the paper is written in multiple languages ​​or when the reviewer's emotional state affects their work efficiency. Furthermore, traditional systems provide static peer review support information and are not sufficiently personalized to meet the needs of individual users. There is a need for a system that can solve these problems and provide efficient and personalized paper peer review support.

[0861] 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.

[0862] In this invention, the server includes means for acquiring paper data, means for tokenizing and analyzing the paper data using a generating AI, means for generating a summary from the analysis result of the generating AI, means for providing peer review support information based on the generated summary, means for recognizing a user's emotional state, means for dynamically adjusting the peer review support information based on the recognized emotional state, and means for displaying the peer review support information through a user interface, thereby enabling efficient and personalized paper peer review support.

[0863] "Means for obtaining paper data" refers to a device or function for downloading paper data from a specified URL and converting the format as necessary.

[0864] "Means for tokenizing and analyzing paper data using generative artificial intelligence" refers to a device or function that uses a natural language processing model to divide paper data into individual tokens and analyze their contents.

[0865] "Means for generating summaries from the analysis results of generative artificial intelligence" refers to a device or function that generates a summary that succinctly summarizes the main points of a paper based on the analyzed data.

[0866] "Means for providing peer review support information based on the generated abstract" refers to a device or function that provides peer review support information such as the novelty, logic, and content consistency of a paper based on the generated abstract.

[0867] The "means for recognizing the emotional state of the user" refers to a device or function that analyzes emotions from the user's facial expressions, voice, keyboard input, etc., and recognizes that emotional state.

[0868] "Means for dynamically adjusting peer review support information based on a recognized emotional state" refers to a device or function that dynamically adjusts the display method and content of peer review support information in accordance with the recognized emotional state of the user.

[0869] A "means for displaying peer review support information through a user interface" is an interface designed to visually display the tailored peer review support information and to allow easy user access.

[0870] The present invention relates to a system for automating the peer review of papers and providing efficient and personalized peer review support to users. The system functions as follows through interactions between a server, a terminal, and a user.

[0871] The system includes the following functions: acquiring paper data, analyzing it using generative artificial intelligence, generating summaries, providing peer review support information, recognizing the user's emotional state and dynamically adjusting information, and displaying information through a user interface.

[0872] Acquisition of paper data

[0873] First, the user enters the URL of the paper to be peer-reviewed. The device sends this URL to the server, and the server accesses the specified URL to retrieve the paper data. The retrieved paper data is in PDF format, for example, and the server converts it into an appropriate format as needed. This format conversion is performed using an OCR tool (e.g., Tesseract).

[0874] Analysis using generative artificial intelligence

[0875] The server tokenizes the acquired paper data using natural language processing libraries (e.g., NLTK, SpaCy). This tokenized data is then fed into a generative AI model (e.g., GPT-4) that analyzes each part of the paper and extracts key points. This analysis process clarifies the paper's subject, purpose, results, etc.

[0876] Generate a summary

[0877] The server uses topic modeling and summary generation algorithms to generate summaries based on the analysis results obtained from the generative AI model. These summaries provide a compact summary of important information for reviewers, allowing for quick understanding.

[0878] Providing peer review support information

[0879] Based on the generated summary, the server generates peer review support information, including an evaluation of the paper's novelty, logic, and content consistency. The server scores the paper based on these criteria and uses a rule-based engine (e.g., if-then rules) to generate specific advice and comments.

[0880] Recognizing the user's emotional state

[0881] The server analyzes the user's facial expressions, voice, keyboard input, etc. collected through the device and uses an emotion engine to recognize the user's emotional state. For example, facial expression analysis and voice emotion recognition technology are used to determine whether the user is feeling stressed.

[0882] Dynamic adjustment of information

[0883] Based on the user's recognized emotional state, the server dynamically adjusts the display method and content of peer review support information. For example, if the user is feeling stressed, the server will adjust the display to show only the most important information in a concise manner.

[0884] Displaying Information Through a User Interface

[0885] Finally, the adjusted peer review support information is displayed on the user interface on the device. The user interface has an intuitive and easy-to-use design, allowing users to easily check the information. Users can utilize the provided support information to review papers quickly and accurately.

[0886] Prompt Sentence Examples

[0887] An example of a prompt that our system could input to the generative AI model would be:

[0888] Generate an abstract for the following paper:

[0889] URL: https: / / example.com / sample_paper.pdf

[0890] Summarize the contents of the paper and state the main points and conclusions.

[0891] This allows our system to significantly streamline the review process and, when combined with emotion recognition, provide a more personalized review experience.

[0892] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0893] Step 1:

[0894] The user enters the URL of the paper they wish to review, and the device sends the URL to the server.

[0895] Input: URL of the paper to be reviewed

[0896] Output: Request for paper data from server

[0897] Specific operation: The user enters a URL into the input field of the terminal and clicks the send button. The terminal sends this URL as an HTTP request to the server.

[0898] Step 2:

[0899] The server accesses the specified URL and retrieves the paper data.

[0900] Input: URL of the paper sent from the device

[0901] Output: Retrieved paper data (PDF format, etc.)

[0902] Specific operation: The server accesses the URL, downloads the paper data in PDF format, and temporarily stores it on the server.

[0903] Step 3:

[0904] The server tokenizes the acquired paper data and analyzes it using generative AI.

[0905] Input: Retrieved paper data

[0906] Output: Tokenized text data and analysis results

[0907] How it works: The server uses a natural language processing library (e.g., NLTK, SpaCy) to extract and tokenize text from the PDF, then feeds the tokenized data into a generative AI model (e.g., GPT-4) to parse out key points.

[0908] Step 4:

[0909] The server generates a summary from the analysis results of the generation AI.

[0910] Input: Analysis results by the generation AI

[0911] Output: Abstract text of the paper

[0912] Specific operation: The server uses topic modeling and summary generation algorithms to generate summaries from the analysis results, which are then stored in an internal database.

[0913] Step 5:

[0914] The server generates peer review support information based on the abstract.

[0915] Input: Generated summary

[0916] Output: Peer review support information (evaluation of novelty, logic, and consistency)

[0917] Specific Actions: The server analyzes the summary, calculates a score based on the evaluation criteria, and uses a rule-based engine to generate specific advice and comments.

[0918] Step 6:

[0919] The server recognizes the user's emotional state.

[0920] Input: User facial expression, voice, and keyboard input data collected from the device

[0921] Output: User's emotional state data

[0922] Specific operation: The device's camera and microphone collect the user's facial expressions and voice, and send them to the server. The server then uses emotion recognition technology to analyze the user's emotional state.

[0923] Step 7:

[0924] The server dynamically adjusts the peer review support information based on the perceived emotional state.

[0925] Input: User's emotional state data

[0926] Output: Adjusted peer review supporting information

[0927] Specific operation: The server receives the results of the emotion engine and adjusts the display method and content of the information. For example, if the user is feeling stressed, it will simply display only the important information.

[0928] Step 8:

[0929] The terminal displays the adjusted peer review support information on the user interface.

[0930] Input: Coordinated peer review supporting information

[0931] Output: Information displayed in the user interface

[0932] Specific operation: The terminal displays the adjusted peer review support information received from the server on the user interface. The user uses this information to perform the peer review efficiently.

[0933] The above is the specific processing flow of the program of this system.

[0934] (Application example 2)

[0935] 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."

[0936] Conventional paper peer review systems make it difficult for reviewers to efficiently review a large number of papers. Furthermore, because information is provided uniformly without considering the reviewer's emotional state, stress and dissatisfaction are likely to occur. To address these issues, a system that recognizes the user's emotions and provides dynamically adjusted peer review support information is needed.

[0937] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring paper data, means for tokenizing and analyzing the paper data using a generation AI, means for generating a summary from the analysis results of the generation AI, means for providing peer review support information based on the generated summary, means for displaying the peer review support information through a user interface, means for recognizing user emotions, and means for dynamically adjusting the peer review support information based on the user emotions. This allows reviewers to review papers efficiently and also allows them to receive personalized information according to their own emotional state.

[0938] "Paper data" refers to scientific or academic documents written by researchers or authors, and is material that presents specific research results or findings.

[0939] "Generative AI" refers to an AI technology that performs advanced processing such as natural language processing, tokenization, analysis, and summary generation based on large datasets.

[0940] "Tokenization" is the process of breaking down text into smaller units such as words and phrases in order to analyze it.

[0941] "Analysis" refers to the process of understanding the content of acquired data and extracting important information.

[0942] A summary is a concise summary of the main points of an entire text.

[0943] "Peer review support information" refers to information to assist reviewers in their work, such as evaluations of the paper's novelty, logic, and content consistency, as well as specific feedback.

[0944] A "user interface" is a means by which a user interacts with a computer system, and refers to a screen on which information is visually displayed and an operating means.

[0945] "User emotions" refer to the psychological states such as stress, excitement, and concentration that users experience when reviewing papers.

[0946] "Dynamic adjustment" refers to the process of changing the content and presentation of information in real time according to the user's emotions and situation.

[0947] The system that realizes this application example is composed of a series of modules that make up the entire system. The detailed operation of each module is shown below.

[0948] Program processing

[0949] Acquisition of paper data

[0950] First, the server retrieves the paper data from the URL specified by the user. This retrieved paper data is temporarily stored for subsequent analysis. Specifically, the server uses a network library (e.g., the requests library) to download the paper data via the Internet.

[0951] Analysis of paper data

[0952] The server then uses a generative AI model to tokenize the acquired paper data and analyze its content. The generative AI uses a natural language processing library (e.g., the Transformers library). Key points and important information are extracted from the analysis results, which then serve as the basis for generating summaries, as described below.

[0953] Generate a summary

[0954] The server generates a summary based on the analyzed data. The generated summary is intended to allow reviewers to quickly grasp the overall picture of the paper, concisely summarizing only the most important points. The summary is generated using the generative AI model mentioned above.

[0955] Emotion recognition by emotion engine

[0956] Furthermore, the server is equipped with an emotion engine that analyzes the user's facial expressions, voice, keyboard input, etc. to recognize their emotional state. This emotion data is then used to dynamically adjust the review support information. The emotion engine uses the dlib library, which provides facial recognition technology, and hardware such as a webcam.

[0957] Dynamic Adjustment

[0958] The server dynamically adjusts the peer review support information it provides based on the user's recognized emotional state. For example, if the user is feeling stressed, it displays information in a simplified manner to reduce the burden on the user. This allows the user to review efficiently in a relaxed state.

[0959] User Interface Display

[0960] Finally, the adjusted peer review support information is displayed on the user's device. The user interface is designed to be visually easy to understand, allowing users to quickly access the information they need. This display can be done using a web application framework (e.g., Flask or Django).

[0961] Specific examples

[0962] For example, if a user enters a URL such as "https: / / example.com / sample_paper.pdf" to review a paper, the server accesses the URL to retrieve the paper data. The retrieved data is analyzed using a generative AI model to generate a summary. The system then creates peer review support information based on the generated summary and uses an emotion engine to recognize the user's emotional state. Finally, the system adjusts the information according to the user's emotions and provides it to the user through a user interface.

[0963] An example of a prompt could be a question such as "What are the main points?", which allows the generative AI model to extract key points from the paper and support efficient peer review.

[0964] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0965] Step 1:

[0966] The user enters the URL of the paper they wish to review and sends it to the server via their terminal. When the server receives this URL, it uses a network library (for example, the requests library) to retrieve the paper data from the URL. The retrieved paper data is temporarily stored on the server. The input is the "URL of the paper" and the output is the "paper data."

[0967] Step 2:

[0968] The server tokenizes and analyzes the acquired paper data using a generative AI model (e.g., the transformers library). This process understands the structure and content of the entire paper and extracts key points. The input is "paper data" and the output is "analysis results."

[0969] Step 3:

[0970] The server generates a summary based on the analysis results. The generated summary is intended to allow reviewers to quickly grasp the overall picture of the paper. This summary generation is also performed using a generative AI model. The input is the "analysis results" and the output is the "summary."

[0971] Step 4:

[0972] The server generates peer review support information based on the abstract. This information includes an evaluation of the paper's novelty, logic, and content consistency. The input is the "abstract" and the output is the "peer review support information."

[0973] Step 5:

[0974] The server uses an emotion engine to analyze the user's facial expressions, voice, and keyboard input to recognize their emotional state. The emotion engine uses the dlib library, which provides facial recognition technology, and hardware such as a webcam. The input is "user's emotional data (facial expressions, voice, and input)" and the output is "recognized emotional state."

[0975] Step 6:

[0976] The server dynamically adjusts the peer review support information based on the recognized emotional state. For example, if the user is feeling stressed, it simplifies the information and makes the display easier to understand. The inputs are the "recognized emotional state" and "peer review support information," and the output is "adjusted peer review support information."

[0977] Step 7:

[0978] The terminal displays the adjusted peer review support information to the user. The information is provided in a visually easy-to-understand format through the user interface. The input is the adjusted peer review support information, and the output is the display to the user.

[0979] 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.

[0980] 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.

[0981] 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.

[0982] [Fourth embodiment]

[0983] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0984] 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.

[0985] 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).

[0986] 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.

[0987] 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.

[0988] 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).

[0989] 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.

[0990] 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.

[0991] 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.

[0992] 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.

[0993] 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.

[0994] 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.

[0995] 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."

[0996] The present invention relates to a system for automating the peer review of academic papers and providing efficient peer review support. This system utilizes generative artificial intelligence to acquire and analyze paper data, generate summaries, and provide peer review support information.

[0997] System configuration

[0998] 1. Data Acquisition Module: The server uses this module to acquire paper data from the Internet or databases. It automatically downloads the paper data from the URL or path specified by the user and converts the format as necessary.

[0999] 2. Generative AI module: The server uses generative AI to tokenize and analyze the acquired paper data. The generative AI uses natural language processing models, for example, to understand the content of the paper and extract key points.

[1000] 3. Summary generation module: The server generates a summary of the paper based on the analysis results of the AI. The generated summary is a concise summary of the main points of the paper, allowing reviewers to quickly grasp the overall picture of the paper.

[1001] 4. Peer review support module: Based on the generated abstract, the server provides specific peer review support information, including an evaluation of the paper's novelty, logic, and content consistency.

[1002] 5. User Interface Module: The terminal displays peer review support information to the user. This interface is designed to be visually easy to understand, allowing users to easily access and understand the information.

[1003] Program processing flow

[1004] 1. Acquisition of paper data: The server acquires the paper data from the URL or database specified by the user. This acquired data is temporarily stored in the server's memory.

[1005] 2. Analyzing the paper data: The server tokenizes the acquired data and analyzes it using a generative AI, which understands the main parts of the paper and summarizes its contents.

[1006] 3. Summary generation: The server creates a summary from the analysis results of the AI. This summary concisely presents important information to reviewers, streamlining the peer review process.

[1007] 4. Providing peer review support information: The server generates peer review support information based on the abstract, including an evaluation of the paper's novelty and logic.

[1008] 5. Display of peer review support information: The terminal displays peer review support information to the user, allowing the user to review papers efficiently based on this information.

[1009] Usage example

[1010] For example, if a user reviews a paper located at "https: / / example.com / sample_paper.pdf," the server accesses the URL and retrieves the paper data. It then uses generative AI to analyze the paper and generate a summary. It then creates peer review support information based on the generated summary and provides it to the user via their device. The user can use the provided support information to quickly and accurately review the paper.

[1011] This system will significantly improve the efficiency of the peer review process and support the rapid evaluation of academic research. Furthermore, since it can handle multilingual processing and patent document acceptance / rejection, it is expected to be applied in a wide range of fields.

[1012] The processing flow will be explained below.

[1013] Step 1:

[1014] The user inputs the URL of the paper they wish to review and sends it to the server via their device. The user then inputs this URL into a dedicated interface and issues a request to retrieve it.

[1015] Step 2:

[1016] The server retrieves the paper data from the specified URL. The server sends an HTTP request and retrieves the paper data as a response. If the data retrieval fails, the server generates an error message and notifies the user.

[1017] Step 3:

[1018] The server tokenizes the acquired paper data and converts it into an input format for the generative AI, which is a preprocessing step to enable the generative AI to accurately analyze the data.

[1019] Step 4:

[1020] The server uses a generative AI to analyze the tokenized paper data, which analyzes each part of the paper and understands its main points and meaning.

[1021] Step 5:

[1022] The server generates a summary from the AI ​​analysis results, extracting the most important information from the analysis results and concisely summarizing the main points of the paper.

[1023] Step 6:

[1024] The server provides peer review support information based on the abstracts generated by the server, which then generates feedback including evaluations of the paper's novelty, logic, and content consistency.

[1025] Step 7:

[1026] The server sends the generated peer review support information to the terminal, which displays the received information on a user interface for the reviewer to use.

[1027] Step 8:

[1028] Users can check the peer review support information provided through the device interface, evaluate the paper based on this information, and efficiently proceed with the peer review process.

[1029] Step 9:

[1030] The user enters the peer review results and sends them to the server via their terminal. The final peer review results are stored in the system and used to make decisions on whether to accept or reject the paper.

[1031] Through this series of processes, the system streamlines the paper peer review process and supports the rapid evaluation of academic research.

[1032] Example 1

[1033] 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."

[1034] The current peer review process is time-consuming and labor-intensive, making it difficult to evaluate many papers in a short period of time. Furthermore, there is a lack of consistency in the evaluation of papers' novelty, logic, and content consistency, resulting in inconsistent evaluations among reviewers. Therefore, there is a need for a system that streamlines the peer review process and improves the quality of evaluations.

[1035] 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.

[1036] In this invention, the server includes means for acquiring paper data from the Internet or a database, means for tokenizing the acquired paper data using a generative artificial intelligence and analyzing it using a natural language processing model, means for extracting important points from the analysis results of the generative artificial intelligence and generating a paper summary, means for providing peer review support information including an evaluation of the paper's novelty, logic, and content consistency based on the generated summary, and means for displaying the peer review support information through a user interface, thereby making it possible to streamline the peer review process and improve the quality of evaluation.

[1037] The "Internet" is a global communications network for accessing distributed information resources and retrieving, transmitting, and sharing data through computer networks.

[1038] A "database" is a system for managing data efficiently and systematically, allowing data to be searched, retrieved, and updated using specific queries.

[1039] "Generative AI" is an AI technology that has the ability to automatically handle specific tasks, learn like a human, analyze data, and generate deliverables.

[1040] "Tokenization" is the process of dividing text data into the smallest meaningful units, such as words and sentences, and is a preprocessing step in natural language processing.

[1041] A "natural language processing model" is a machine learning model for understanding, generating, and manipulating natural language, enabling computers to interpret human language.

[1042] A "summary" is a short, concise summary of the main points and conclusions from a longer document.

[1043] "Novelty" is a concept that indicates that a particular research or invention is something new that has not existed before.

[1044] "Logicality" refers to whether an argument or conclusion is based on a consistent logical structure, and is a standard for evaluating whether it is logically sound.

[1045] "Content relevance" is a concept that evaluates whether the content of a document is appropriate to its purpose or theme and meets the expected items and standards.

[1046] A "user interface" is a means by which a user interacts with a computer system, providing a visually easy-to-understand display of information and a method of operation.

[1047] This invention relates to a system that automates and improves the efficiency of the paper peer review process by using a generative AI model to acquire paper data, analyze it, generate summaries, and provide and display peer review support information.

[1048] 1. Acquisition of paper data

[1049] First, the user specifies the URL of the paper data they want to retrieve from the Internet or a database. The server accesses the specified URL and downloads the paper data via the HTTP protocol. During this process, the server uses the Python requests library to send an HTTP request, saves the paper data locally, and converts the format if necessary.

[1050] 2. Analysis of paper data

[1051] The acquired paper data is tokenized by the server. For example, PDF format data is subjected to text extraction using the PyMuPDF library. This text data is then input into a generative AI model, such as GPT-3, for analysis. The generative AI uses the specified prompt to analyze the content of the paper and extract key points.

[1052] 3. Summary Generation

[1053] The server generates a summary of the paper based on the analysis results of the generative AI. This summary is a concise summary of the main points of the paper, allowing reviewers to quickly grasp the overall picture of the paper.

[1054] 4. Providing peer review support information

[1055] From the generated abstract, the server generates specific peer review support information, including ratings of novelty, logic, and relevance of the content, allowing reviewers to quickly determine which parts of the paper are worthy of evaluation.

[1056] 5. Viewing peer review support information

[1057] Finally, the terminal displays the peer review support information sent from the server to the user. The user interface is designed to be visually easy to understand, allowing users to easily access and understand the information. For example, an interface using a web browser is used.

[1058] Specific examples

[1059] Consider a case where a user reviews a paper located at "https: / / example.com / sample_paper.pdf." First, the server accesses the URL and retrieves the paper data. Next, it analyzes this data using the generative AI model "GPT-3" and uses the following prompt:

[1060] "Please write an abstract of this paper. Include the main points of your conclusions, an outline of your methodology, and any key findings from your results."

[1061] Based on the generated summary, the server creates peer review support information on the paper's novelty, logic, and content consistency, which is displayed on the terminal, allowing users to efficiently proceed with the peer review.

[1062] This system will significantly improve the efficiency of the peer review process, enabling the rapid evaluation of academic research. Furthermore, it is capable of handling multilingual processing and the acceptance or rejection of patent documents, and is expected to be applied in a wide range of fields.

[1063] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1064] Step 1:

[1065] Acquisition of paper data

[1066] The server receives a URL specified by the user as input. It accesses the specified URL, for example, "https: / / example.com / sample_paper.pdf," and downloads the paper data using the HTTP protocol. During this process, the server uses the requests library to send an HTTP request and saves the retrieved data locally. The output is the paper data saved in the server's temporary memory or storage.

[1067] Step 2:

[1068] Tokenization and formatting

[1069] The server receives the paper data obtained in step 1 as input. If the paper data is in PDF format, it extracts the text using a library such as PyMuPDF and tokenizes the data. Specifically, it extracts the text of each page from the PDF file and splits it into a series of words and sentences as text data. The output is text data that can be analyzed.

[1070] Step 3:

[1071] Analysis by generative AI

[1072] The server receives the text data generated in step 2 as input. It uses a generative AI model, such as GPT-3, to analyze the paper data. It provides the generative AI with the prompt, "Write a summary of this paper. Include the main points of the conclusion, an overview of the methodology, and important findings from the results." and performs the analysis. The output is the extracted main points and summary text.

[1073] Step 4:

[1074] Generate a summary

[1075] The server receives as input the analysis results generated in step 3. From the analysis results generated, it creates a summary of the paper. The summary is a concise summary of the main sections, such as conclusions, methodology, and results. The summary is stored in a data structure on the server. The output is a summary of the completed paper.

[1076] Step 5:

[1077] Providing peer review support information

[1078] The server receives the abstract generated in step 4 as input. Based on the abstract, it creates peer review support information regarding novelty, logic, and relevance of the content. This information includes evaluations such as "This paper is innovative compared to previous research," "The logical development is consistent and the conclusion is persuasive," and "The content is highly relevant to the research objectives." The output is specific peer review support information.

[1079] Step 6:

[1080] View peer review support information

[1081] The terminal receives as input the peer review support information generated in step 5. It displays this information to the user through a visually easy-to-understand user interface. For example, it formats the information using HTML and CSS in the browser, providing a display format that is easily accessible and understandable to the user. The output is a screen displaying the peer review support information that the user can view.

[1082] Through these steps, the system will be able to streamline the peer review process and support the rapid evaluation of academic research.

[1083] (Application example 1)

[1084] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1085] Traditional peer review processes are often manual and require a great deal of time and effort. Furthermore, the lack of uniform evaluation criteria makes it difficult to obtain consistent peer review results. Furthermore, when multilingual processing is required, language barriers exist, limiting the access and sharing of information in a field of knowledge. This creates a challenge for researchers and students to quickly grasp the latest research information.

[1086] 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.

[1087] In this invention, the server includes: means for acquiring paper data; means for tokenizing and analyzing the paper data using a generative AI; means for generating summaries from the analysis results of the generative AI; means for providing peer review support information based on the generated summaries; means for generating prompts for evaluating the novelty, logic, and content consistency of the paper based on the summaries in generating the peer review support information; and means for displaying the peer review support information through a user interface. This automates the peer review process, enabling efficient and consistent peer review. Furthermore, multilingual processing is possible, allowing researchers and students to quickly and easily access international research information.

[1088] "Paper data" refers to documents that describe academic research content and are stored electronically.

[1089] "Generative AI" refers to artificial intelligence technology that uses machine learning and deep learning to analyze and generate natural language.

[1090] "Tokenization" refers to the process of breaking down text into smaller units such as words or phrases.

[1091] "Analysis" refers to the process of understanding the acquired data and grasping its meaning and structure.

[1092] "Generative AI" refers to artificial intelligence that has the ability to generate new information or summaries based on data.

[1093] An "abstract" is a short sentence or paragraph that succinctly summarizes the contents of the entire document.

[1094] "Peer review support information" is specific information provided to assist in the evaluation of a paper, including an assessment of the paper's importance and quality.

[1095] A "prompt sentence" refers to an input sentence that causes a generation AI to generate a specific output.

[1096] "User interface" refers to the interface through which a system and a user communicate with each other.

[1097] In this invention, the server acquires and analyzes paper data using the following means to generate peer review support information.

[1098] 1. Acquisition of paper data

[1099] The server retrieves paper data from a URL or database. The hardware used for this is a cloud server, and the software used is an "HTTP request library" (e.g., Retrofit, OkHttp). The server accesses the URL specified by the user and downloads the paper data. For example, if the user specifies a paper at "https: / / example.com / sample_paper.pdf," the server retrieves the paper data from that URL.

[1100] 2. Analysis of paper data

[1101] The server tokenizes the acquired paper data and analyzes it using generative AI. The hardware used is a cloud server, and the software used is a "natural language processing model" (e.g., GPT-4, BERT). Tokenization breaks the paper data down into small units (words and phrases) that are then analyzed by generative AI.

[1102] 3. Summary Generation

[1103] The server generates a summary from the analysis results of the generative AI. The generative AI model used (e.g., BART, a large-scale language model) is used to summarize the analysis results as short sentences or paragraphs. For example, a BART model is used to generate a summary that concisely summarizes the main points of a paper.

[1104] 4. Generating peer review support information

[1105] Based on the generated abstract, we provide peer review support information, including generating prompts to evaluate the novelty, logic, and relevance of the content of the paper based on the abstract. For example, the generated prompts are as follows:

[1106] "Based on the following paper abstract, please rate the paper's novelty, logic, and relevance to the content. Abstract: [generated abstract content]"

[1107] 5. Display in the user interface

[1108] The device displays the generated peer review support information to the user through a user interface. The hardware used is a smartphone or computer, and the software is a "UI framework" (e.g., Flutter, React Native). Based on this information, the user can review papers efficiently and accurately.

[1109] Program execution example

[1110] For example, if a user reviews a paper located at "https: / / example.com / sample_paper.pdf," the server accesses the URL and retrieves the paper data. Next, it uses GPT-4 to analyze the paper and generates a summary using the BART model. Based on the summary, it inputs a prompt to GPT-3.5-turbo to generate specific review support information, which is then displayed on the smartphone's user interface.

[1111] As a result, this invention automates the peer review process, making it possible to provide fast and consistent peer review. Furthermore, its multilingual capabilities make it easier to access international research information.

[1112] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1113] Step 1:

[1114] The server retrieves the paper data from a URL specified by the user or from a database. The input is a URL provided by the user, and an HTTP request is made based on this to download the paper data. Specifically, the server sends a request using an HTTP request library (e.g., Retrofit, OkHttp) and retrieves PDF or text data as a response. The output is the retrieved paper data.

[1115] Step 2:

[1116] The server tokenizes and analyzes the acquired paper data. The input is the paper data acquired in step 1, which is tokenized using generative artificial intelligence. Specifically, it uses a natural language processing library (e.g., SpaCy, NLTK) to split (tokenize) the paper data into words and phrases. It then analyzes the content using a model such as GPT-4 or BERT. The output is the analyzed data (the tokenized data and its analysis results).

[1117] Step 3:

[1118] The server generates a summary from the analysis results. The input is the analysis result data obtained in step 2, which is input into a summary generation model (e.g., BART, a large-scale language model). The specific operation is to use a summary generation algorithm to extract the main points from the content of the entire paper and generate a concise summary. The output is the generated summary.

[1119] Step 4:

[1120] The server provides peer review support information based on the generated summary. The input is the summary generated in step 3, which is used to generate a prompt and provide evaluation information for peer review. Specifically, it uses a model such as GPT-3.5-turbo to generate a prompt to evaluate the paper's novelty, logic, and content relevance based on the summary, and then generates specific peer review support information based on the prompt. The output is peer review support information. For example, the generated prompt is as follows:

[1121] "Based on the following paper abstract, please rate the paper's novelty, logic, and relevance to the content. Abstract: [generated abstract content]"

[1122] Step 5:

[1123] The terminal displays the generated peer review support information to the user through a user interface. The input is the peer review support information generated in step 4, and a UI framework (e.g., Flutter, React Native) is used to display this information in an easy-to-understand manner for the user. The specific operation is to create a layout for visually displaying the peer review support information and display it on the screen. The output is the peer review support information displayed in a format that the user can view.

[1124] 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.

[1125] The present invention relates to a system that automates the peer review of papers and provides efficient peer review support. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide a more personalized peer review experience. The following describes embodiments of the present invention.

[1126] System configuration

[1127] 1. Data Acquisition Module: The server uses this module to acquire paper data from the Internet or databases. It downloads the paper data from the URL specified by the user and converts the format if necessary.

[1128] 2. Generative AI module: The server uses generative AI to tokenize and analyze the acquired paper data. The generative AI uses a natural language processing model to understand the content of the paper and extract key points.

[1129] 3. Summary generation module: The server generates a summary of the paper based on the analysis results of the AI. The summary is a concise summary of the main points of the paper, allowing reviewers to quickly grasp the overall picture.

[1130] 4. Peer review support module: The server provides specific peer review support information based on the generated abstract, including an evaluation of the paper's novelty, logic, and content consistency.

[1131] 5. Emotion Engine: The emotion engine installed on the server analyzes the user's facial expressions, voice, keyboard input, etc. to recognize the user's emotional state. Based on this information, the peer review support information is dynamically adjusted.

[1132] 6. User interface module: The terminal displays peer review support information to the user. The user interface is designed to be visually easy to understand, allowing users to easily access and understand the information.

[1133] Program processing flow

[1134] 1. Acquisition of paper data:

[1135] The user enters the URL of the paper they wish to review and sends it to the server via their device. The server then retrieves the paper data from the specified URL.

[1136] 2. Analysis of paper data:

[1137] The server tokenizes the acquired paper data and analyzes it using a generative AI, which analyzes each part of the paper to understand its main points and meaning.

[1138] 3. Generate a summary:

[1139] The server generates a summary from the analysis results of the AI. The summary provides important information to reviewers in a concise manner, streamlining the peer review process.

[1140] 4. Providing peer review support information:

[1141] Based on the server-generated abstract, the system provides specific peer review support information, including an assessment of the paper's novelty, logic, and content relevance.

[1142] 5. Emotion Recognition with Emotion Engine:

[1143] The server recognizes the user's emotions from their facial expressions, voice, keyboard input, etc. The emotion engine analyzes this data and understands the user's current emotional state.

[1144] 6. Dynamic adjustment according to emotions:

[1145] The server adjusts the presentation and design of peer review support information according to the user's emotional state. For example, if the user is feeling stressed, the server will simplify the presentation of information.

[1146] 7. View peer review support information:

[1147] The terminal displays the adjusted peer review support information to the user, who then performs the peer review efficiently based on the provided information.

[1148] Usage example

[1149] For example, if a user reviews a paper located at "https: / / example.com / sample_paper.pdf," the server accesses the URL to retrieve the paper data. It then uses generative artificial intelligence to analyze the paper and generate a summary. It then creates review support information based on the generated summary and uses an emotion engine to recognize the user's emotional state. Finally, it adjusts the information according to the emotion and provides it to the user via their device. The user can use the provided support information to quickly and accurately review the paper.

[1150] This allows our system to significantly streamline the review process, and when combined with emotion recognition, it can provide a more personalized review experience.

[1151] The processing flow will be explained below.

[1152] The present invention relates to a system that automates the peer review of papers and provides efficient peer review support. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to provide a more personalized peer review experience. Below, an embodiment of the present invention will be described using specific processing steps.

[1153] Step 1:

[1154] The user enters the URL of the paper they wish to review and sends it to the server via their device. The user then enters the URL into a dedicated interface and issues a command to retrieve the paper data.

[1155] Step 2:

[1156] The server retrieves the paper data from the specified URL. The server sends an HTTP request, receives the paper data as a response, and temporarily stores the data in memory. If the retrieval fails, an error message is generated to notify the user.

[1157] Step 3:

[1158] The server tokenizes the acquired paper data and converts it into an input format for the generation AI. Tokenization is the process of dividing text data into analyzable units.

[1159] Step 4:

[1160] The server analyzes the tokenized paper data using a generative AI, which uses a natural language processing model to understand the content of the paper and extract key points.

[1161] Step 5:

[1162] The server generates a summary from the analysis results of the AI. Based on the analysis results, a summary that concisely summarizes the main points of the paper is created. The summary is important information that allows reviewers to quickly grasp the overall picture of the paper.

[1163] Step 6:

[1164] The server generates abstracts and provides peer review support information, including an assessment of the paper's novelty, logic, and content consistency.

[1165] Step 7:

[1166] The emotion engine installed on the server analyzes the user's facial expressions, voice, keyboard input, etc. to recognize the user's emotional state. The emotion engine uses this data to understand the user's current emotional state.

[1167] Step 8:

[1168] The server dynamically adjusts the presentation and design of peer review support information according to the user's perceived emotional state: for example, if the user is feeling stressed, it will simplify the information or add an encouraging message.

[1169] Step 9:

[1170] The terminal displays tailored peer review support information to the user. The user interface is designed to be visually intuitive, allowing users to easily access and understand the information.

[1171] Step 10:

[1172] Users can check the peer review support information provided through the device interface and evaluate papers based on that information, allowing them to efficiently carry out their peer review work.

[1173] Step 11:

[1174] The user enters the peer review results and sends them to the server via their terminal. The final peer review results are stored in the system and used to make decisions on whether to accept or reject the paper.

[1175] Through this series of processes, the system of the present invention significantly streamlines the peer review process and supports rapid evaluation of academic research. Furthermore, by combining it with an emotion engine, it is possible to provide users with a more appropriate and personalized peer review experience.

[1176] Example 2

[1177] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1178] The traditional paper peer review process is time-consuming and places a heavy burden on reviewers. This burden is particularly pronounced when the paper is written in multiple languages ​​or when the reviewer's emotional state affects their work efficiency. Furthermore, traditional systems provide static peer review support information and are not sufficiently personalized to meet the needs of individual users. There is a need for a system that can solve these problems and provide efficient and personalized paper peer review support.

[1179] 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.

[1180] In this invention, the server includes means for acquiring paper data, means for tokenizing and analyzing the paper data using a generating AI, means for generating a summary from the analysis result of the generating AI, means for providing peer review support information based on the generated summary, means for recognizing a user's emotional state, means for dynamically adjusting the peer review support information based on the recognized emotional state, and means for displaying the peer review support information through a user interface, thereby enabling efficient and personalized paper peer review support.

[1181] "Means for obtaining paper data" refers to a device or function for downloading paper data from a specified URL and converting the format as necessary.

[1182] "Means for tokenizing and analyzing paper data using generative artificial intelligence" refers to a device or function that uses a natural language processing model to divide paper data into individual tokens and analyze their contents.

[1183] "Means for generating summaries from the analysis results of generative artificial intelligence" refers to a device or function that generates a summary that succinctly summarizes the main points of a paper based on the analyzed data.

[1184] "Means for providing peer review support information based on the generated abstract" refers to a device or function that provides peer review support information such as the novelty, logic, and content consistency of a paper based on the generated abstract.

[1185] The "means for recognizing the emotional state of the user" refers to a device or function that analyzes emotions from the user's facial expressions, voice, keyboard input, etc., and recognizes that emotional state.

[1186] "Means for dynamically adjusting peer review support information based on a recognized emotional state" refers to a device or function that dynamically adjusts the display method and content of peer review support information in accordance with the recognized emotional state of the user.

[1187] A "means for displaying peer review support information through a user interface" is an interface designed to visually display the tailored peer review support information and to allow easy user access.

[1188] The present invention relates to a system for automating the peer review of papers and providing efficient and personalized peer review support to users. The system functions as follows through interactions between a server, a terminal, and a user.

[1189] The system includes the following functions: acquiring paper data, analyzing it using generative artificial intelligence, generating summaries, providing peer review support information, recognizing the user's emotional state and dynamically adjusting information, and displaying information through a user interface.

[1190] Acquisition of paper data

[1191] First, the user enters the URL of the paper to be peer-reviewed. The device sends this URL to the server, and the server accesses the specified URL to retrieve the paper data. The retrieved paper data is in PDF format, for example, and the server converts it into an appropriate format as needed. This format conversion is performed using an OCR tool (e.g., Tesseract).

[1192] Analysis using generative artificial intelligence

[1193] The server tokenizes the acquired paper data using natural language processing libraries (e.g., NLTK, SpaCy). This tokenized data is then fed into a generative AI model (e.g., GPT-4) that analyzes each part of the paper and extracts key points. This analysis process clarifies the paper's subject, purpose, results, etc.

[1194] Generate a summary

[1195] The server uses topic modeling and summary generation algorithms to generate summaries based on the analysis results obtained from the generative AI model. These summaries provide a compact summary of important information for reviewers, allowing for quick understanding.

[1196] Providing peer review support information

[1197] Based on the generated summary, the server generates peer review support information, including an evaluation of the paper's novelty, logic, and content consistency. The server scores the paper based on these criteria and uses a rule-based engine (e.g., if-then rules) to generate specific advice and comments.

[1198] Recognizing the user's emotional state

[1199] The server analyzes the user's facial expressions, voice, keyboard input, etc. collected through the device and uses an emotion engine to recognize the user's emotional state. For example, facial expression analysis and voice emotion recognition technology are used to determine whether the user is feeling stressed.

[1200] Dynamic adjustment of information

[1201] Based on the user's recognized emotional state, the server dynamically adjusts the display method and content of peer review support information. For example, if the user is feeling stressed, the server will adjust the display to show only the most important information in a concise manner.

[1202] Displaying Information Through a User Interface

[1203] Finally, the adjusted peer review support information is displayed on the user interface on the device. The user interface has an intuitive and easy-to-use design, allowing users to easily check the information. Users can utilize the provided support information to review papers quickly and accurately.

[1204] Prompt Sentence Examples

[1205] An example of a prompt that our system could input to the generative AI model would be:

[1206] Generate an abstract for the following paper:

[1207] URL: https: / / example.com / sample_paper.pdf

[1208] Summarize the contents of the paper and state the main points and conclusions.

[1209] This allows our system to significantly streamline the review process and, when combined with emotion recognition, provide a more personalized review experience.

[1210] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1211] Step 1:

[1212] The user enters the URL of the paper they wish to review, and the device sends the URL to the server.

[1213] Input: URL of the paper to be reviewed

[1214] Output: Request for paper data from server

[1215] Specific operation: The user enters a URL into the input field of the terminal and clicks the send button. The terminal sends this URL as an HTTP request to the server.

[1216] Step 2:

[1217] The server accesses the specified URL and retrieves the paper data.

[1218] Input: URL of the paper sent from the device

[1219] Output: Retrieved paper data (PDF format, etc.)

[1220] Specific operation: The server accesses the URL, downloads the paper data in PDF format, and temporarily stores it on the server.

[1221] Step 3:

[1222] The server tokenizes the acquired paper data and analyzes it using generative AI.

[1223] Input: Retrieved paper data

[1224] Output: Tokenized text data and analysis results

[1225] How it works: The server uses a natural language processing library (e.g., NLTK, SpaCy) to extract and tokenize text from the PDF, then feeds the tokenized data into a generative AI model (e.g., GPT-4) to parse out key points.

[1226] Step 4:

[1227] The server generates a summary from the analysis results of the generation AI.

[1228] Input: Analysis results by the generation AI

[1229] Output: Abstract text of the paper

[1230] Specific operation: The server uses topic modeling and summary generation algorithms to generate summaries from the analysis results, which are then stored in an internal database.

[1231] Step 5:

[1232] The server generates peer review support information based on the abstract.

[1233] Input: Generated summary

[1234] Output: Peer review support information (evaluation of novelty, logic, and consistency)

[1235] Specific Actions: The server analyzes the summary, calculates a score based on the evaluation criteria, and uses a rule-based engine to generate specific advice and comments.

[1236] Step 6:

[1237] The server recognizes the user's emotional state.

[1238] Input: User facial expression, voice, and keyboard input data collected from the device

[1239] Output: User's emotional state data

[1240] Specific operation: The device's camera and microphone collect the user's facial expressions and voice, and send them to the server. The server then uses emotion recognition technology to analyze the user's emotional state.

[1241] Step 7:

[1242] The server dynamically adjusts the peer review support information based on the perceived emotional state.

[1243] Input: User's emotional state data

[1244] Output: Adjusted peer review supporting information

[1245] Specific operation: The server receives the results of the emotion engine and adjusts the display method and content of the information. For example, if the user is feeling stressed, it will simply display only the important information.

[1246] Step 8:

[1247] The terminal displays the adjusted peer review support information on the user interface.

[1248] Input: Coordinated peer review supporting information

[1249] Output: Information displayed in the user interface

[1250] Specific operation: The terminal displays the adjusted peer review support information received from the server on the user interface. The user uses this information to perform the peer review efficiently.

[1251] The above is the specific processing flow of the program of this system.

[1252] (Application example 2)

[1253] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1254] Conventional paper peer review systems make it difficult for reviewers to efficiently review a large number of papers. Furthermore, because information is provided uniformly without considering the reviewer's emotional state, stress and dissatisfaction are likely to occur. To address these issues, a system that recognizes the user's emotions and provides dynamically adjusted peer review support information is needed.

[1255] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring paper data, means for tokenizing and analyzing the paper data using a generation AI, means for generating a summary from the analysis results of the generation AI, means for providing peer review support information based on the generated summary, means for displaying the peer review support information through a user interface, means for recognizing user emotions, and means for dynamically adjusting the peer review support information based on the user emotions. This allows reviewers to review papers efficiently and also allows them to receive personalized information according to their own emotional state.

[1256] "Paper data" refers to scientific or academic documents written by researchers or authors, and is material that presents specific research results or findings.

[1257] "Generative AI" refers to an AI technology that performs advanced processing such as natural language processing, tokenization, analysis, and summary generation based on large datasets.

[1258] "Tokenization" is the process of breaking down text into smaller units such as words and phrases in order to analyze it.

[1259] "Analysis" refers to the process of understanding the content of acquired data and extracting important information.

[1260] A summary is a concise summary of the main points of an entire text.

[1261] "Peer review support information" refers to information to assist reviewers in their work, such as evaluations of the paper's novelty, logic, and content consistency, as well as specific feedback.

[1262] A "user interface" is a means by which a user interacts with a computer system, and refers to a screen on which information is visually displayed and an operating means.

[1263] "User emotions" refer to the psychological states such as stress, excitement, and concentration that users experience when reviewing papers.

[1264] "Dynamic adjustment" refers to the process of changing the content and presentation of information in real time according to the user's emotions and situation.

[1265] The system that realizes this application example is composed of a series of modules that make up the entire system. The detailed operation of each module is shown below.

[1266] Program processing

[1267] Acquisition of paper data

[1268] First, the server retrieves the paper data from the URL specified by the user. This retrieved paper data is temporarily stored for subsequent analysis. Specifically, the server uses a network library (e.g., the requests library) to download the paper data via the Internet.

[1269] Analysis of paper data

[1270] The server then uses a generative AI model to tokenize the acquired paper data and analyze its content. The generative AI uses a natural language processing library (e.g., the Transformers library). Key points and important information are extracted from the analysis results, which then serve as the basis for generating summaries, as described below.

[1271] Generate a summary

[1272] The server generates a summary based on the analyzed data. The generated summary is intended to allow reviewers to quickly grasp the overall picture of the paper, concisely summarizing only the most important points. The summary is generated using the generative AI model mentioned above.

[1273] Emotion recognition by emotion engine

[1274] Furthermore, the server is equipped with an emotion engine that analyzes the user's facial expressions, voice, keyboard input, etc. to recognize their emotional state. This emotion data is then used to dynamically adjust the review support information. The emotion engine uses the dlib library, which provides facial recognition technology, and hardware such as a webcam.

[1275] Dynamic Adjustment

[1276] The server dynamically adjusts the peer review support information it provides based on the user's recognized emotional state. For example, if the user is feeling stressed, it displays information in a simplified manner to reduce the burden on the user. This allows the user to review efficiently in a relaxed state.

[1277] User Interface Display

[1278] Finally, the adjusted peer review support information is displayed on the user's device. The user interface is designed to be visually easy to understand, allowing users to quickly access the information they need. This display can be done using a web application framework (e.g., Flask or Django).

[1279] Specific examples

[1280] For example, if a user enters a URL such as "https: / / example.com / sample_paper.pdf" to review a paper, the server accesses the URL to retrieve the paper data. The retrieved data is analyzed using a generative AI model to generate a summary. The system then creates peer review support information based on the generated summary and uses an emotion engine to recognize the user's emotional state. Finally, the system adjusts the information according to the user's emotions and provides it to the user through a user interface.

[1281] An example of a prompt could be a question such as "What are the main points?", which allows the generative AI model to extract key points from the paper and support efficient peer review.

[1282] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1283] Step 1:

[1284] The user enters the URL of the paper they wish to review and sends it to the server via their terminal. When the server receives this URL, it uses a network library (for example, the requests library) to retrieve the paper data from the URL. The retrieved paper data is temporarily stored on the server. The input is the "URL of the paper" and the output is the "paper data."

[1285] Step 2:

[1286] The server tokenizes and analyzes the acquired paper data using a generative AI model (e.g., the transformers library). This process understands the structure and content of the entire paper and extracts key points. The input is "paper data" and the output is "analysis results."

[1287] Step 3:

[1288] The server generates a summary based on the analysis results. The generated summary is intended to allow reviewers to quickly grasp the overall picture of the paper. This summary generation is also performed using a generative AI model. The input is the "analysis results" and the output is the "summary."

[1289] Step 4:

[1290] The server generates peer review support information based on the abstract. This information includes an evaluation of the paper's novelty, logic, and content consistency. The input is the "abstract" and the output is the "peer review support information."

[1291] Step 5:

[1292] The server uses an emotion engine to analyze the user's facial expressions, voice, and keyboard input to recognize their emotional state. The emotion engine uses the dlib library, which provides facial recognition technology, and hardware such as a webcam. The input is "user's emotional data (facial expressions, voice, and input)" and the output is "recognized emotional state."

[1293] Step 6:

[1294] The server dynamically adjusts the peer review support information based on the recognized emotional state. For example, if the user is feeling stressed, it simplifies the information and makes the display easier to understand. The inputs are the "recognized emotional state" and "peer review support information," and the output is "adjusted peer review support information."

[1295] Step 7:

[1296] The terminal displays the adjusted peer review support information to the user. The information is provided in a visually easy-to-understand format through the user interface. The input is the adjusted peer review support information, and the output is the display to the user.

[1297] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1298] 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.

[1299] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1300] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1301] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1302] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1303] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1304] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1305] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1306] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1307] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1308] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1309] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1310] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1311] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1312] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1313] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1314] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1315] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1316] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1317] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1318] The following is further disclosed regarding the above embodiment.

[1319] (Claim 1)

[1320] A means of obtaining paper data;

[1321] A means of tokenizing and analyzing paper data using generative artificial intelligence;

[1322] A means for generating a summary from the analysis results of the generating artificial intelligence;

[1323] A means of providing peer review support information based on the generated abstract;

[1324] a means for displaying peer review support information through a user interface;

[1325] A system including:

[1326] (Claim 2)

[1327] 2. The system according to claim 1, further comprising means for processing the acquired paper data in multiple languages.

[1328] (Claim 3)

[1329] The system of claim 1, wherein the peer review support information includes information regarding the novelty, logic, and content consistency of the document being peer reviewed.

[1330] (Claim 4)

[1331] The system according to claim 1, further comprising a means for learning from a database of previous research, and improving the peer review process based on the analysis results.

[1332] (Claim 5)

[1333] 2. The system of claim 1, further comprising means for detecting hallucinations in the generative artificial intelligence and means for suggesting modifications if detected.

[1334] (Claim 6)

[1335] 10. The system of claim 1, wherein the user interface includes means for visually displaying peer review support information.

[1336] (Claim 7)

[1337] The system according to claim 1, comprising means applicable to the acceptance or rejection of not only academic papers but also patent documents.

[1338] "Example 1"

[1339] (Claim 1)

[1340] A means of obtaining article data from the internet or databases;

[1341] A means for tokenizing the acquired paper data using generative artificial intelligence and analyzing it using a natural language processing model;

[1342] A means for extracting important points from the analysis results of the generative artificial intelligence and generating a summary of the paper;

[1343] A means of providing peer review support information based on the generated abstract, including an evaluation of the paper's novelty, logic, and content consistency; and

[1344] a means for displaying peer review support information through a user interface;

[1345] A system including:

[1346] (Claim 2)

[1347] 2. The system according to claim 1, further comprising means for processing the acquired paper data in multiple languages.

[1348] (Claim 3)

[1349] The system of claim 1, wherein the peer review support information includes an evaluation of the paper's novelty, logic, and content consistency.

[1350] "Application Example 1"

[1351] (Claim 1)

[1352] A means of obtaining paper data;

[1353] A means of tokenizing and analyzing paper data using generative artificial intelligence;

[1354] A means for generating a summary from the analysis results of the generative AI;

[1355] A means of providing peer review support information based on the generated abstract;

[1356] In generating peer review support information, a means for generating prompt sentences for evaluating the novelty, logic, and content consistency of a paper based on its abstract;

[1357] a means for displaying peer review support information through a user interface;

[1358] A system including:

[1359] (Claim 2)

[1360] 2. The system according to claim 1, further comprising means for processing the acquired paper data in multiple languages.

[1361] (Claim 3)

[1362] 2. The system of claim 1, wherein the peer review support information includes a prompt that provides an evaluation of the novelty, logic, and content consistency of the document being reviewed based on the summary.

[1363] "Example 2: Combining Emotion Engines"

[1364] (Claim 1)

[1365] A means of obtaining paper data;

[1366] A means of tokenizing and analyzing paper data using generative artificial intelligence;

[1367] A means for generating a summary from the analysis results of the generating artificial intelligence;

[1368] A means of providing peer review support information based on the generated abstract;

[1369] means for recognizing the emotional state of a user;

[1370] a means for dynamically adjusting peer review support information based on the perceived emotional state;

[1371] a means for displaying peer review support information through a user interface;

[1372] A system including:

[1373] (Claim 2)

[1374] 2. The system according to claim 1, further comprising means for processing the acquired paper data in multiple languages.

[1375] (Claim 3)

[1376] The system according to claim 1, further comprising means for providing the review support information with information regarding the novelty, logic, and content consistency of the document being reviewed.

[1377] "Application example 2 when combining emotion engines"

[1378] (Claim 1)

[1379] A means of obtaining paper data;

[1380] A means of tokenizing and analyzing paper data using generative artificial intelligence;

[1381] A means for generating a summary from the analysis results of the generating artificial intelligence;

[1382] A means of providing peer review support information based on the generated abstract;

[1383] a means for displaying peer review support information through a user interface;

[1384] means for recognizing a user's emotion;

[1385] a means for dynamically adjusting the peer review support information based on user sentiment;

[1386] A system including:

[1387] (Claim 2)

[1388] 2. The system according to claim 1, further comprising means for processing article data in multiple languages.

[1389] (Claim 3)

[1390] The system of claim 1, wherein the peer review support information includes information regarding the novelty, logic, and content consistency of the document being peer reviewed. [Explanation of symbols]

[1391] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of obtaining paper data; A means of tokenizing and analyzing paper data using generative artificial intelligence; A means for generating a summary from the analysis results of the generative artificial intelligence; A means of providing peer review support information based on the generated abstract; a means for displaying peer review support information through a user interface; A system including:

2. 2. The system according to claim 1, further comprising means for processing the acquired paper data in multiple languages.

3. 2. The system according to claim 1, wherein the peer review support information includes information regarding the novelty, logic, and content consistency of the document to be peer reviewed.

4. The system according to claim 1, further comprising means for learning from a database of previous research, and for improving the peer review process based on the analysis results.

5. 2. The system of claim 1, further comprising means for detecting hallucinations in the generative artificial intelligence and means for suggesting modifications if detected.

6. 10. The system of claim 1, wherein the user interface includes means for visually displaying the peer review support information.

7. 10. The system of claim 1, further comprising means applicable to the acceptance or rejection of patent documents as well as academic papers.

Images