Paper interpretation system based on large language model

Abstract

The invention discloses a thesis interpretation system based on a large language model, and relates to the technical field of language models. The method comprises a text analysis large language model and a database; the paper analysis large language model comprises a paper self-interpretation analysis model and a paper comparison interpretation analysis model; the paper self-interpretation analysis model comprises a paper interpretation module, a paper feature extraction module, a paper analysis evaluation module and a paper analysis result generation module; and the paper contrastive interpretation analysis model comprises a network search module, a paper arrangement module, a paper contrastive analysis evaluation module and a paper correlation architecture module. According to the invention, the paper self-interpretation analysis model and the paper comparison interpretation analysis model are established to carry out framework construction on the papers and the related papers, so that the same or different parts of the papers and the related papers can be fully and quickly found; different parts and identical parts of research contents and related papers can be quickly known and compared, and valuable places of the research can be quickly determined.

Description

Technical Field

[0001] The present invention belongs to the technical field of paper query and interpretation, and in particular relates to a paper interpretation system based on a large language model. Background Art

[0002] Research frontiers are the most advanced, up-to-date, and promising research topics or areas in scientific and technological research. They represent the challenges, hot topics, and development trends in scientific and technological development. Currently, scholars both domestically and internationally have conducted extensive research on the evaluation of cutting-edge technologies, constructing various evaluation methods and models. Among them, peer review and bibliometrics are two widely used methods in assessing academic competitiveness. The evaluation of a scientific frontier is generally based primarily on peer review, supplemented by quantitative metrics such as awards, number of published papers, citations, and the H-index.

[0003] When evaluating and interpreting papers, direct comparative analysis is typically performed against databases (major paper repositories) to understand the value of the paper's research areas and modules. However, this often focuses solely on checking for duplicate results, failing to analyze the differences and similarities between the research direction and modules of the paper and those of related papers. This inability to generate a framework diagram that facilitates quick results, effective comparison, and comprehensive interpretation of the paper. Consequently, users must directly review the relevant data from related papers after the paper is interpreted to understand or conduct comparative analysis. This makes it impossible to identify relevant components and generate a framework diagram using the interpretation model. Summary of the Invention

[0004] The purpose of the present invention is to provide a paper interpretation system based on a large language model. By analyzing and interpreting the paper with a large language model and searching for related papers for comparative analysis, the present invention solves the problems of incomplete interpretation of existing papers, unclear analysis results, and the need for users to compare and view the differences between related papers and the present paper.

[0005] To solve the above technical problems, the present invention is achieved through the following technical solutions:

[0006] The present invention is a paper interpretation system based on a large language model, comprising a paper analysis large language model and a database; the paper analysis large language model comprises a paper self-interpretation analysis model and a paper comparative interpretation analysis model;

[0007] The paper self-interpretation analysis model includes a paper interpretation module, a paper feature extraction module, a paper analysis and evaluation module, and a paper analysis result generation module. The paper is input into the paper interpretation module and the interpretation is started. The paper interpretation module reads the paper and analyzes the paper theme, research direction, and research module to generate an interpretation paper framework diagram. The paper feature extraction module extracts the key research and analysis reports in the research direction and research module in the interpretation paper framework diagram. After extraction, the paper analysis and evaluation module evaluates and analyzes the key research and analysis reports. The paper analysis result generation module structures the evaluation and analysis results into the interpretation paper framework diagram to generate an interpretation paper analysis result framework diagram. The original paper and the interpretation paper analysis result framework diagram of the paper will be stored in the database.

[0008] The paper comparative interpretation and analysis model includes a network search module, a paper sorting module, a paper comparative analysis and evaluation module and a paper relevance architecture module. The network search module first reads the interpretation paper analysis result framework diagram from the database. The paper subject, research direction and research module in the interpretation paper analysis result framework diagram send HTTP requests through the network and search for relevant papers from major paper repositories. The paper sorting module sorts and classifies the searched relevant papers and analyzes the relevant papers to generate a relevant paper analysis result framework diagram. The analysis comparison model in the paper comparative analysis and evaluation module compares and analyzes the interpretation paper analysis result framework diagram with the relevant paper analysis result framework diagram to obtain a comparative analysis result. The paper relevance architecture module structures the comparative analysis result into the interpretation paper analysis result framework diagram to obtain a paper relevance architecture interpretation diagram.

[0009] The present invention is further configured such that the database stores an evaluation model and evaluation data for evaluating the framework diagram of the analysis results of the interpretation paper, the evaluation model compares and analyzes the research level of each research direction and research module in the framework diagram of the analysis results of the interpretation paper with the evaluation data, and a scoring unit is provided in the evaluation model, the scoring unit scores the research level of each research direction and research module, and annotates the obtained score into the analysis results of the research direction and research module in the framework diagram of the analysis results of the interpretation paper.

[0010] The present invention is further configured such that the paper organizing module analyzes the searched related papers to generate a score for the research direction and research module within the related paper analysis result framework diagram, and also marks the obtained score into the analysis results of the research direction and research module within the related paper analysis result framework diagram, and compares the score of the research direction and research module within the interpreted paper analysis result framework diagram with the score of similar or identical research directions and research modules in the related papers to determine the height of the research degree of the research direction and research module of the paper that needs to be interpreted in the existing related papers.

[0011] The present invention is further configured such that a duplicate checking module is provided in the evaluation model, and the duplicate checking module first compares the research objects, research methods, research data and research results of the research direction and research modules of the paper to be interpreted with the relevant papers to determine the overlap rate between the research direction and research modules of the paper to be interpreted and the research objects, research methods, research data and research results of the existing relevant papers.

[0012] The present invention is further configured to establish a comparison of overlapping parts of research objects, research methods, research data and research results, and analyze the degree of overlap and the differences between the overlapping parts.

[0013] The present invention is further configured to establish an overlap analysis model, and establish an analysis template for the research object, research method, research data and research results. The analysis templates of the overlap analysis models for different fields and directions are different. After determining the field and direction, the overlap analysis model of the corresponding field and direction will be called to analyze the overlap rate of the paper.

[0014] The present invention is further configured such that the framework diagram for interpreting the analysis results of the paper is provided with hyperlinks at the corresponding analysis results of the research direction and the research module, and the location and content of the corresponding research object, research method, research data and research results in the original paper can be viewed by clicking the hyperlinks;

[0015] The framework diagram of the analysis results of the relevant papers also has hyperlinks in the corresponding analysis results of the research directions and research modules. Click the hyperlink to view the corresponding paper name, paper library where the paper is located, and the chapter position in the relevant paper.

[0016] The present invention is further configured such that, when interpreting a paper, the user can choose to analyze the paper using the paper self-interpretation analysis model or to analyze the paper using the paper self-interpretation analysis model and the paper comparative interpretation analysis model. After the analysis results come out, the user can choose whether to download the interpretation results or automatically send them to a designated mailbox.

[0017] The present invention is further configured such that after each paper is interpreted, new fields, new research directions, new research methods and corresponding research results are saved in a database, and these new fields, new research directions, new research methods and corresponding research results are regularly analyzed by professionals to improve the paper self-interpretation analysis model.

[0018] The present invention has the following beneficial effects:

[0019] 1. The present invention establishes two models: a paper self-interpretation analysis model and a paper comparative interpretation analysis model, and constructs a framework for the paper and related papers, which is convenient for quickly understanding the paper framework, paper research content and research results, etc. It can help the interpretation process to quickly understand the paper and facilitate the establishment of a foundation for reading content of the subsequent paper comparative interpretation analysis model. The paper comparative interpretation analysis model can analyze the relevant papers and construct a related paper research framework diagram. Users can see and interpret the research objects, research methods, research data and research results of the related papers, so as to fully and quickly find the similarities or differences between this paper and related papers, and be able to interpret this paper well, solving the problem that traditional interpretation only checks the duplication rate and gives the source of related papers, but cannot know the detailed differences and similarities between related papers.

[0020] 2. The present invention also sets up a scoring module for the research and integrates the scoring structure into the framework diagram, so that users can quickly understand the differences and similarities between the research content and related papers, and quickly determine the valuable aspects of the research of the present invention and the areas with high overlap, so as to provide direction for subsequent modification of the paper.

[0021] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 This is a schematic diagram of the overall interpretation principle of a paper interpretation system based on a large language model.

[0024] Figure 2 This is a schematic diagram of the principle of a paper self-interpretation analysis model of a paper interpretation system based on a large language model.

[0025] Figure 3 This is a schematic diagram of the principle of a paper comparative interpretation and analysis model for a paper interpretation system based on a large language model. DETAILED DESCRIPTION

[0026] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0027] See also Figure 1-3 The present invention is a paper interpretation system based on a large language model, comprising a paper analysis large language model and a database; the paper analysis large language model comprises a paper self-interpretation analysis model and a paper comparative interpretation analysis model;

[0028] The paper self-interpretation analysis model includes a paper interpretation module, a paper feature extraction module, a paper analysis and evaluation module, and a paper analysis result generation module. The paper is input into the paper interpretation module and the interpretation is started. The paper interpretation module reads the paper and analyzes the paper theme, research direction, and research module to generate an interpretation paper framework diagram. The paper feature extraction module extracts the key research and analysis reports in the research direction and research module in the interpretation paper framework diagram. After extraction, the paper analysis and evaluation module evaluates and analyzes the key research and analysis reports. The paper analysis result generation module structures the evaluation and analysis results into the interpretation paper framework diagram to generate an interpretation paper analysis result framework diagram. The original paper and the interpretation paper analysis result framework diagram of the paper will be stored in the database.

[0029] The paper comparative interpretation and analysis model includes a network search module, a paper sorting module, a paper comparative analysis and evaluation module and a paper relevance architecture module. The network search module first reads the interpretation paper analysis result framework diagram from the database. The paper subject, research direction and research module in the interpretation paper analysis result framework diagram send HTTP requests through the network and search for relevant papers from major paper repositories. The paper sorting module sorts and classifies the searched relevant papers and analyzes the relevant papers to generate a relevant paper analysis result framework diagram. The analysis comparison model in the paper comparative analysis and evaluation module compares and analyzes the interpretation paper analysis result framework diagram with the relevant paper analysis result framework diagram to obtain a comparative analysis result. The paper relevance architecture module structures the comparative analysis result into the interpretation paper analysis result framework diagram to obtain a paper relevance architecture interpretation diagram.

[0030] The interpretation of the paper is divided into two parts: constructing a framework diagram for the paper, interpreting the research direction and content of the paper, and generating a framework diagram for interpreting the paper. The key research and analysis reports are extracted to generate a framework diagram for interpreting the analysis results of the paper. The framework diagram for interpreting the analysis results of the paper can facilitate a quick understanding of the content of the paper and facilitate the subsequent comparative interpretation of the analysis model to extract relevant content.

[0031] The paper comparative interpretation and analysis model can extract the content of the interpretation paper analysis result framework diagram, send HTTP requests through the network, and search for relevant papers from major paper repositories. The paper sorting module organizes and classifies the searched relevant papers and analyzes the relevant papers to generate a relevant paper analysis result framework diagram. The relevant paper analysis result framework diagram can easily show the degree of relevance to the research of this paper and the number of papers studying this area. It is convenient for users to know how many research areas in their research areas are being studied and published. Then, a framework diagram of related papers in similar or identical research modules is generated for comparison with this paper. The analysis and comparison model in the paper comparative analysis and evaluation module compares and analyzes the interpretation paper analysis result framework diagram with the relevant paper analysis result framework diagram to obtain a comparative analysis result. The analysis and comparison result can facilitate users to see the comparison result at a glance. The result structure is then added to the interpretation paper analysis result framework diagram to obtain a paper relevance architecture interpretation diagram to save the interpretation data.

[0032] The database stores an evaluation model and evaluation data for evaluating the framework diagram of the analysis results of the interpretation paper. The evaluation model compares and analyzes the research level of each research direction and research module in the framework diagram of the analysis results of the interpretation paper with the evaluation data. A scoring unit is provided in the evaluation model. The scoring unit scores the research level of each research direction and research module, and annotates the obtained score into the analysis results of the research direction and research module in the framework diagram of the analysis results of the interpretation paper.

[0033] The evaluation scores can more intuitively show the value of your own research, as well as the similarities and differences in research. Through the scores, you can also see the value of the research in each corresponding position, which can make it more convenient to fully interpret the value of the paper for each part.

[0034] The paper sorting module analyzes the searched relevant papers to generate a score for the research direction and research module within the relevant paper analysis result framework diagram, and also marks the obtained score into the analysis results of the research direction and research module within the relevant paper analysis result framework diagram. The score of the research direction and research module within the interpretation paper analysis result framework diagram is compared with the score of similar or identical research directions and research modules in the relevant papers to determine the height of the research degree of the research direction and research module of the paper that needs to be interpreted in the existing relevant papers.

[0035] There will be many relevant papers, and the research value of each relevant paper in the entire relevant field (targeted scoring of the same or different parts of each research module) can be used to obtain the research level of most relevant papers, and then compare them with this application to see how the research level of this paper compares to the existing relevant papers, so that it can be modified or improved later.

[0036] The evaluation model is provided with a duplicate checking module, which first compares the research objects, research methods, research data and research results of the research direction and research modules of the paper to be interpreted with the relevant papers to determine the overlap rate between the research direction and research modules of the paper to be interpreted and the research objects, research methods, research data and research results of the existing relevant papers.

[0037] Many interpretation or query systems have a paper search module, which primarily analyzes the extent and degree of overlap between a paper's research and existing related papers. It also analyzes the overlap rate in more specific areas, such as research subjects, methods, data, and results. This allows researchers to identify areas of similar or identical research content and quickly address these areas.

[0038] Create a comparison chart to compare the overlap between research subjects, methods, data, and results, and analyze the degree of overlap and the differences between the overlaps. Create a table to quickly identify similarities and identify locations where the overlap rate reaches a certain level, and understand the degree of overlap.

[0039] Establish an overlap analysis model, creating an analysis template for each research object, research method, research data, and research results. Different fields and directions use different templates for overlap analysis. Once the field and direction are determined, the overlap analysis model for the corresponding field and direction will be used to analyze the overlap rate of papers. This allows for a more professional analysis of the overlapped parts, reducing the need for users to perform their own analysis or providing a model analysis result when comparing, thereby improving analysis efficiency.

[0040] The framework diagram for interpreting the analysis results of the paper has hyperlinks in the corresponding analysis results of the research direction and research module. Clicking the hyperlinks can view the corresponding research objects, research methods, research data and research results in the original paper.

[0041] The framework diagram of the analysis results of the relevant papers also has hyperlinks in the corresponding analysis results of the research directions and research modules. Click the hyperlink to view the corresponding paper name, paper library and chapter position of the relevant paper.

[0042] The framework diagram only provides a general overview of the research content and analysis. It doesn't provide access to the original research, related paper references, or chapters. By creating hyperlinks on the generated framework diagram, you can quickly view the content or find related paper references and chapters. This makes searching more convenient.

[0043] When interpreting a paper, users can choose to use the self-interpretation analysis model or a combination of the self-interpretation analysis model and the comparative interpretation analysis model. Once the analysis results are available, users can choose whether to download them or automatically send them to a designated email address, making it easy to save interpretation data. Users can also select the level of analysis to tailor the interpretation to their needs.

[0044] After each paper interpretation, new areas, new research directions, new research methods, and corresponding research results are saved in the database. These new areas, new research directions, new research methods, and corresponding research results are regularly analyzed by professionals to improve the paper self-interpretation analysis model. The existing interpretation model can be regularly improved by interpreting relatively less researched papers, thereby improving the accuracy and perspective of interpreting less researched papers.

[0045] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0046] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the content of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A paper interpretation system based on a large language model, comprising a large language model for paper analysis and a database; characterized by: The paper analysis language model includes a paper self-interpretation analysis model and a paper comparative interpretation analysis model; The paper self-interpretation analysis model includes a paper interpretation module, a paper feature extraction module, a paper analysis and evaluation module, and a paper analysis result generation module. The paper is input into the paper interpretation module and the interpretation is started. The paper interpretation module reads the paper and analyzes the paper theme, research direction, and research module to generate an interpretation paper framework diagram. The paper feature extraction module extracts the key research and analysis reports in the research direction and research module in the interpretation paper framework diagram. After extraction, the paper analysis and evaluation module evaluates and analyzes the key research and analysis reports. The paper analysis result generation module structures the evaluation and analysis results into the interpretation paper framework diagram to generate an interpretation paper analysis result framework diagram. The original paper and the interpretation paper analysis result framework diagram of the paper will be stored in the database. The paper comparative interpretation and analysis model includes a network search module, a paper sorting module, a paper comparative analysis and evaluation module and a paper relevance architecture module. The network search module first reads the interpretation paper analysis result framework diagram from the database. The paper subject, research direction and research module in the interpretation paper analysis result framework diagram send HTTP requests through the network and search for relevant papers from major paper repositories. The paper sorting module sorts and classifies the searched relevant papers and analyzes the relevant papers to generate a relevant paper analysis result framework diagram. The analysis comparison model in the paper comparative analysis and evaluation module compares and analyzes the interpretation paper analysis result framework diagram with the relevant paper analysis result framework diagram to obtain a comparative analysis result. The paper relevance architecture module structures the comparative analysis result into the interpretation paper analysis result framework diagram to obtain a paper relevance architecture interpretation diagram.

2. A paper interpretation system based on a large language model according to claim 1, characterized in that: The database stores an evaluation model and evaluation data for evaluating the framework diagram of the analysis results of the interpretation paper. The evaluation model compares and analyzes the research level of each research direction and research module in the framework diagram of the analysis results of the interpretation paper with the evaluation data. A scoring unit is provided in the evaluation model. The scoring unit scores the research level of each research direction and research module, and annotates the obtained score into the analysis results of the research direction and research module in the framework diagram of the analysis results of the interpretation paper.

3. A paper interpretation system based on a large language model according to claim 2, characterized in that: The paper sorting module analyzes the searched relevant papers to generate a score for the research direction and research module within the relevant paper analysis result framework diagram, and also marks the obtained score into the analysis results of the research direction and research module within the relevant paper analysis result framework diagram. The score of the research direction and research module within the interpretation paper analysis result framework diagram is compared with the score of similar or identical research directions and research modules in the relevant papers to determine the height of the research degree of the research direction and research module of the paper that needs to be interpreted in the existing relevant papers.

4. A paper interpretation system based on a large language model according to claim 3, characterized in that: The evaluation model is provided with a duplicate checking module, which first compares the research objects, research methods, research data and research results of the research direction and research modules of the paper to be interpreted with the relevant papers to determine the overlap rate between the research direction and research modules of the paper to be interpreted and the research objects, research methods, research data and research results of the existing relevant papers.

5. A paper interpretation system based on a large language model according to claim 4, characterized in that: Establish a comparison of the overlapping parts of research objects, research methods, research data and research results, and analyze the degree of overlap and the differences between the overlapping parts.

6. A paper interpretation system based on a large language model according to claim 1, characterized in that: An overlap analysis model is established, and an analysis template is established for the research object, research method, research data, and research results. The analysis templates of the overlap analysis models for different fields and directions are different. After determining the field and direction, the overlap analysis model of the corresponding field and direction will be called to analyze the overlap rate of the papers.

7. A paper interpretation system based on a large language model according to claim 1, characterized in that: The framework diagram for interpreting the analysis results of the paper has hyperlinks in the corresponding analysis results of the research direction and research module. Clicking the hyperlinks can view the corresponding research objects, research methods, research data and research results in the original paper. The framework diagram of the analysis results of the relevant papers also has hyperlinks in the corresponding analysis results of the research directions and research modules. Click the hyperlink to view the corresponding paper name, paper library where the paper is located, and the chapter position in the relevant paper.

8. A paper interpretation system based on a large language model according to claim 1, characterized in that: When interpreting a paper, users can choose to analyze it using the paper's self-interpretation analysis model or jointly analyze it using the paper's self-interpretation analysis model and the paper's comparative interpretation analysis model. After the analysis results come out, users can choose whether to download the interpretation results or automatically send them to a designated email address.

9. A paper interpretation system based on a large language model according to claim 1, characterized in that: After each paper is interpreted, new fields, new research directions, new research methods and corresponding research results are saved in the database. These new fields, new research directions, new research methods and corresponding research results are regularly analyzed by professionals to improve the paper self-interpretation analysis model.

Images