Intelligent scientific research literature analysis system based on AI large model

Abstract

The invention discloses an intelligent scientific research literature analysis system based on an AI large model, and the system comprises an information input module which is used for obtaining literature analysis demand information of a user; the literature analysis module is used for screening a plurality of finance literatures in corresponding fields according to the literature analysis demand information, inputting a single literature into a pre-trained AI large model to obtain a single analysis result, summarizing all the single analysis results, and performing secondary analysis to obtain a preliminary analysis result; the result feedback module is used for acquiring evaluation feedback of the user on the preliminary analysis result; and the data storage module is used for storing the analysis result output by the literature analysis module and the evaluation feedback corresponding to the analysis result. Compared with the prior art, the problem that scientific research literature analysis efficiency is low in a traditional research mode in the specific finance subdivision field is solved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an intelligent scientific research document analysis system based on an AI large model. Background Art

[0002] A large AI model refers to a machine learning model with ultra-large-scale parameters (usually more than one billion) and complex computing structures. It is usually able to process massive amounts of data and complete various complex tasks, such as natural language processing and image recognition.

[0003] In order to analyze and study a certain technical sub-field of finance, researchers need to read and understand the scientific research literature in this field and conduct in-depth research. The traditional analysis method of manually studying and analyzing scientific research literature, which only relies on the expert knowledge of researchers, is inefficient in the analysis and research of literature. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent scientific research literature analysis system based on an AI large model to solve the problem of low efficiency of scientific research literature analysis in traditional research methods in specific financial sub-fields.

[0005] In order to achieve the above purpose, the present invention adopts the following technical solution: an intelligent scientific research document analysis system based on AI large model, comprising:

[0006] An information input module, which is used to obtain the user's literature analysis demand information;

[0007] The literature analysis module is used to screen several financial literatures in the corresponding field according to the literature analysis demand information. After a single literature is input into the pre-trained AI large model, a single analysis result is obtained. After summarizing all the single analysis results, a secondary analysis is performed to obtain the preliminary analysis result.

[0008] A result feedback module, which is used to obtain user evaluation feedback on the preliminary analysis results;

[0009] The data storage module is used to store the analysis results output by the document analysis module and the evaluation feedback corresponding to the analysis results.

[0010] As a further description of the above technical solution:

[0011] The information input module includes an information acquisition unit and a keyword extraction unit. The information acquisition unit is used to perform preliminary processing on the input document analysis demand information, and the keyword extraction unit is used to extract characteristic words related to finance in the document analysis demand information.

[0012] As a further description of the above technical solution:

[0013] The literature analysis module includes a literature screening unit, a preliminary analysis unit and a secondary analysis unit. The literature screening unit is used to screen financial literature in the corresponding sub-field according to the extracted characteristic vocabulary. The preliminary analysis unit is used to input a single document into the pre-trained AI large model to obtain a single analysis result. The single analysis result includes author information, literature research topic and research conclusion. The secondary analysis unit is used to summarize all single analysis results and then conduct a secondary analysis to obtain a preliminary analysis result. The analysis indicators of the preliminary analysis result include research trends in sub-fields, distribution of research directions in sub-fields, hot research directions in sub-fields, distribution of author institutions, research trends of institutions and author research trends.

[0014] As a further description of the above technical solution:

[0015] The result feedback module includes a content identification unit and a satisfaction evaluation unit. The content identification unit is used for users to identify unsatisfactory analysis indicators in the preliminary analysis results, and the satisfaction evaluation unit is used for users to feedback satisfaction evaluation of the preliminary analysis results.

[0016] As a further description of the above technical solution:

[0017] The content identification unit also includes an inefficient identification subunit and a content refinement subunit. The inefficient identification subunit is used to identify analysis indicators with low importance in the user's preliminary analysis results, and the content refinement subunit is used to identify analysis indicators with brief content and insufficient description in the user's preliminary analysis results.

[0018] As a further description of the above technical solution:

[0019] The document analysis module also includes a content optimization unit, which is used to optimize the preliminary analysis results to obtain the final analysis results according to the identification of the analysis indicators in the preliminary analysis results by the content identification unit.

[0020] As a further description of the above technical solution:

[0021] The content optimization unit includes a sorting subunit and a ratio adjustment subunit. The sorting subunit adjusts the sorting of each analysis indicator in the final analysis result according to the identification of the analysis indicator in the preliminary analysis result by the inefficiency identification subunit. The ratio adjustment subunit adjusts the ratio of the number of words in a single analysis indicator to the total number of words in the final analysis result according to the identification of the inefficiency identification subunit and the identification of the content refinement subunit.

[0022] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0023] 1. In the present invention, when researchers need to conduct research on a certain sub-field of finance, they input analysis demand information through the information input module. The literature analysis module screens a number of corresponding financial literature according to the literature analysis demand information, and obtains a single analysis result after a single literature is input into the pre-trained AI large model. After summarizing all the single analysis results, a secondary analysis is performed to obtain a preliminary analysis result, which quickly realizes the analysis and summary of a large number of literature in the sub-field, and effectively improves the efficiency of researchers' scientific research literature analysis.

[0024] 2. In the present invention, for the preliminary analysis results output by the document analysis module, researchers can evaluate and provide feedback on the preliminary analysis results through the result feedback module. The document analysis module optimizes the preliminary analysis results based on the evaluation feedback to obtain the final analysis results, and outputs analysis results that meet the current needs of researchers, thereby effectively improving the file analysis quality of the scientific research document analysis system.

[0025] 3. In the present invention, based on the general big model, financial literature and financial experts' summaries of literature are put into the AI ​​big model for training to obtain a pre-trained AI big model. Subsequently, the powerful computing and processing capabilities of the AI ​​big model are used to complete the analysis and summary of a single financial literature. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0027] Figure 1 This is a system architecture diagram of an intelligent scientific research literature analysis system based on AI large model. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0029] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0031] In the description of the embodiments of the present invention, it should be noted that the terms "upper", "inner", etc. indicate directions or positional relationships based on the directions or positional relationships shown in the accompanying drawings, or are directions or positional relationships in which the inventive product is usually placed when in use. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as a limitation on the present invention.

[0032] In the description of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal communication of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0033] Example 1

[0034] See also Figure 1 The present invention provides a technical solution: an intelligent scientific research document analysis system based on an AI large model, comprising:

[0035] An information input module, which is used to obtain the user's literature analysis demand information;

[0036] The literature analysis module is used to screen several financial literatures in the corresponding field according to the literature analysis demand information. After a single literature is input into the pre-trained AI large model, a single analysis result is obtained. After summarizing all the single analysis results, a secondary analysis is performed to obtain the preliminary analysis result.

[0037] A result feedback module, which is used to obtain user evaluation feedback on the preliminary analysis results;

[0038] The data storage module is used to store the analysis results output by the document analysis module and the evaluation feedback corresponding to the analysis results.

[0039] The information input module includes an information acquisition unit and a keyword extraction unit. The information acquisition unit is used to perform preliminary processing on the input document analysis demand information and remove redundant spaces and punctuation marks. The keyword extraction unit is used to extract characteristic words related to finance in the document analysis demand information. After extracting the words from the document analysis demand information, the keyword extraction unit searches for standard words in a pre-set finance vocabulary library to obtain characteristic words with corresponding standard words.

[0040] The literature analysis module includes a literature screening unit, a preliminary analysis unit and a secondary analysis unit. The literature screening unit is used to screen financial literature in the corresponding sub-field according to the extracted characteristic vocabulary. The preliminary analysis unit is used to input a single document into the pre-trained AI large model to obtain a single analysis result. The single analysis result includes author information, literature research topic and research conclusion. The author information specifically includes author name, school, publication time and other information. The secondary analysis unit is used to summarize all single analysis results and then conduct secondary analysis to obtain preliminary analysis results. The analysis indicators of the preliminary analysis results include research trends in sub-fields, distribution of research directions in sub-fields, hot research directions in sub-fields, distribution of author schools, research trends of schools and author research trends.

[0041] For the retrieved relevant literature in the subdivided field, the single analysis result obtained by summarizing a single document according to the AI ​​big model is combined with the research topic of the literature and the publication time of the article to obtain the research trend of the subdivided field, the distribution of research directions in the subdivided field, and the hot research direction in the subdivided field. Combined with the research topic of the literature, the publication time of the article, and the author information, the distribution of the author's school, the research trend of the school, and the research trend of the author are obtained to realize the literature analysis of the subdivided field. Based on the general big model, the financial literature and the summary of the literature by financial experts are put into the AI ​​big model for training to obtain the pre-trained AI big model. Subsequently, the powerful computing power and processing power of the AI ​​big model are used to complete the analysis and summary of a single financial literature.

[0042] Working principle: When researchers need to conduct research on a certain sub-field of finance, they can input analysis demand information through the information input module. The document analysis demand information includes requirements such as specific fields, publication time, and publishing institutions. The document analysis module selects several corresponding financial documents based on the document analysis demand information. After a single document is input into the pre-trained AI large model, a single analysis result is obtained. After summarizing all single analysis results, a secondary analysis is performed to obtain preliminary analysis results, which can quickly realize the analysis and summary of a large number of documents in this sub-field, effectively improving the efficiency of researchers' scientific research document analysis.

[0043] For the preliminary analysis results output by the literature analysis module, researchers can evaluate and provide feedback on the preliminary analysis results through the result feedback module. The literature analysis module optimizes the preliminary analysis results based on the evaluation feedback to obtain the final analysis results, and outputs analysis results that meet the current needs of researchers, effectively improving the file analysis quality of the scientific research literature analysis system.

[0044] Example 2

[0045] Based on the above embodiments, this embodiment further makes the following improved technical solutions: the result feedback module includes a content identification unit and a satisfaction assessment unit, the content identification unit is used for users to identify unsatisfactory analysis indicators in the preliminary analysis results, and the satisfaction assessment unit is used for users to feedback satisfaction evaluation of the preliminary analysis results.

[0046] The user can mark the analysis indicators that are not satisfactory in the preliminary analysis results through the result feedback module. The overall satisfaction of the user with the preliminary analysis results can be evaluated from low to high, from level 1 to 5.

[0047] When evaluating the satisfaction of the preliminary analysis results of user feedback, when the proportion of evaluations below level 4 in all evaluation feedback reaches the set threshold, the secondary analysis unit adjusts the preliminary analysis results based on the statistics of the unsatisfactory analysis indicators fed back by the content identification unit.

[0048] Example 3

[0049] Based on the above embodiments, this embodiment further makes the following improved technical solutions: the content identification unit also includes an inefficient identification sub-unit and a content refinement sub-unit, the inefficient identification sub-unit is used to identify analysis indicators with low importance in the preliminary analysis results of users, and the content refinement sub-unit is used to identify analysis indicators with brief content and insufficient description in the preliminary analysis results of users.

[0050] Users can use content identification units to specifically classify and identify unimportant analysis indicators and analysis indicators with insufficient descriptions, clarify their own literature analysis needs, and obtain high-quality analysis results from the analysis system.

[0051] Example 4

[0052] This embodiment further makes the following improved technical solution on the basis of the above embodiment: the document analysis module also includes a content optimization unit, which is used to optimize the preliminary analysis results to obtain the final analysis results according to the identification of the analysis indicators in the preliminary analysis results by the content identification unit.

[0053] The content optimization unit includes a sorting subunit and a ratio adjustment subunit. The sorting subunit adjusts the sorting of each analysis indicator in the final analysis result according to the identification of the analysis indicator in the preliminary analysis result by the inefficiency identification subunit. The ratio adjustment subunit adjusts the ratio of the number of words in a single analysis indicator to the total number of words in the final analysis result according to the identification of the inefficiency identification subunit and the identification of the content refinement subunit.

[0054] After the user marks the unimportant analysis indicators through the inefficient identification subunit, the sorting subunit adjusts the sorting of the unimportant analysis indicators in the final analysis results to the back, so that researchers can quickly obtain effective information. On the other hand, after the user marks the analysis indicators through the inefficient identification subunit and the content refinement subunit, the sorting subunit increases the number of words of the important analysis indicators according to the importance and current number of words of the individual analysis indicators in the preliminary analysis results and the needs of the researchers, and adjusts the proportion of the word count of the analysis indicator to the total word count of the final analysis results.

[0055] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An intelligent scientific research literature analysis system based on AI large model, characterized in that: include: An information input module, which is used to obtain the user's literature analysis demand information; The literature analysis module is used to screen several financial literatures in the corresponding field according to the literature analysis demand information. After a single literature is input into the pre-trained AI large model, a single analysis result is obtained. After summarizing all the single analysis results, a secondary analysis is performed to obtain the preliminary analysis result. A result feedback module, which is used to obtain user evaluation feedback on the preliminary analysis results; The data storage module is used to store the analysis results output by the document analysis module and the evaluation feedback corresponding to the analysis results.

2. According to claim 1, the intelligent scientific research document analysis system based on AI large model is characterized in that: The information input module includes an information acquisition unit and a keyword extraction unit. The information acquisition unit is used to perform preliminary processing on the input document analysis demand information, and the keyword extraction unit is used to extract characteristic words related to finance in the document analysis demand information.

3. According to claim 2, the intelligent scientific research document analysis system based on AI large model is characterized in that: The literature analysis module includes a literature screening unit, a preliminary analysis unit and a secondary analysis unit. The literature screening unit is used to screen financial literature in corresponding subdivided fields according to the extracted characteristic vocabulary. The preliminary analysis unit is used to input a single document into a pre-trained AI large model to obtain a single analysis result. The single analysis result includes author information, literature research topics and research conclusions. The secondary analysis unit is used to summarize all single analysis results and then perform secondary analysis to obtain preliminary analysis results. The analysis indicators of the preliminary analysis results include research trends in subdivided fields, distribution of research directions in subdivided fields, hot research directions in subdivided fields, distribution of author institutions, research trends of institutions and author research trends.

4. According to claim 1, the intelligent scientific research document analysis system based on AI large model is characterized in that: The result feedback module includes a content identification unit and a satisfaction evaluation unit. The content identification unit is used for users to identify unsatisfactory analysis indicators in the preliminary analysis results, and the satisfaction evaluation unit is used for users to feedback satisfaction evaluation of the preliminary analysis results.

5. According to claim 4, the intelligent scientific research document analysis system based on AI large model is characterized in that: The content identification unit also includes an inefficient identification subunit and a content refinement subunit. The inefficient identification subunit is used to identify analysis indicators with low importance in the user's preliminary analysis results, and the content refinement subunit is used to identify analysis indicators with brief content and insufficient description in the user's preliminary analysis results.

6. According to claim 5, the intelligent scientific research document analysis system based on AI large model is characterized in that: The document analysis module further comprises a content optimization unit, which is used to optimize the preliminary analysis result to obtain the final analysis result according to the identification of the analysis index in the preliminary analysis result by the content identification unit.

7. According to claim 6, the intelligent scientific research document analysis system based on AI large model is characterized in that: The content optimization unit includes a sorting subunit and a ratio adjustment subunit. The sorting subunit adjusts the sorting of each analysis indicator in the final analysis result according to the identification of the analysis indicator in the preliminary analysis result by the inefficiency identification subunit. The ratio adjustment subunit adjusts the ratio of the number of words in a single analysis indicator to the total number of words in the final analysis result according to the identification of the inefficiency identification subunit and the identification of the content refinement subunit.

Images