Generative analysis report implementation method based on artificial intelligence algorithm technology

Abstract

The invention discloses a generative analysis report implementation method based on an artificial intelligence algorithm technology, and belongs to the field of software design, and the method comprises the following steps: constructing a B / S architecture system which comprises a front-end server and a rear-end server; uploading a data source through the front-end server, and transmitting the data source to the rear-end server; after receiving the data source, the back-end server preliminarily analyzes the data source and processes the data source into a data sample corresponding to the artificial intelligence large model; the artificial intelligence large model receives the data sample and carries out data processing to generate analysis report content; and the back-end server returns the generated analysis report content to the front-end server, and the front-end server displays the analysis report content in real time according to the analysis report content. According to the invention, by constructing the efficient B / S architecture system and combining the cooperative work of the front end and the rear end, the full-process automation from the uploading, processing and analysis of the data source to the generation and display of the report is successfully realized.

Description

Technical Field

[0001] The present invention relates to the field of software design, and in particular to a method for realizing a generative analysis report based on artificial intelligence algorithm technology. Background Art

[0002] In the modern information society, report generation and data analysis have become important components in many fields, including but not limited to corporate management, financial analysis, business analysis, scientific research reports, etc. However, traditional report generation methods usually rely on manual writing, which has the following major problems:

[0003] 1. Inefficiency: The traditional method requires manual collection, organization, and analysis of data, and the analysis results are expressed in the form of text and charts. This process is time-consuming and labor-intensive, especially when it involves large-scale, multi-dimensional data. Manual processing cannot meet the needs of rapid report generation.

[0004] 2. Accuracy depends on human experience: The quality of manual report writing depends on the professional level and experience of the writer, which is easily affected by subjective bias. At the same time, the analysis of large amounts of complex data may lead to omissions or errors.

[0005] 3. Insufficient personalization and customization: Traditional methods require additional manual input when generating customized reports for specific needs. Different users have different requirements for the content, format, language style, etc. of the report, which further increases the complexity of manual processing.

[0006] 4. High difficulty in data analysis: With the development of big data technology, the amount of data accumulated in various industries has exploded. These data usually have complex structures and diversity, including text, tables, images and videos, etc. It is difficult for traditional manual methods to efficiently extract useful information from them.

[0007] In recent years, with the development of artificial intelligence (AI) and natural language processing (NLP) technology, new technical paths have been provided for automated report generation. In particular, the natural language generation (NLG) model based on deep learning can automatically generate structured and unstructured report texts by analyzing data. In addition, combined with machine learning and data visualization technology, AI can achieve multi-dimensional analysis and visual expression of complex data, thereby significantly improving the efficiency and quality of report generation.

[0008] To this end, those skilled in the art have provided a method for implementing a generative analysis report based on artificial intelligence algorithm technology to solve the problems raised in the above background technology. Summary of the invention

[0009] The purpose of the present invention is to provide a generative analysis report implementation method based on artificial intelligence algorithm technology. By building an efficient B / S architecture system and combining the collaborative work of the front-end and the back-end, the full process automation from data source uploading, processing, analysis to report generation and display is successfully realized. The system makes full use of the powerful text generation capability of the artificial intelligence big model and the intuitive display advantages of data visualization technology, and can generate various types of reports quickly, accurately and personalized, which is particularly suitable for financial analysis, enterprise management, market research and other fields to solve the problems raised in the above background technology.

[0010] To achieve the above object, the present invention provides the following technical solutions:

[0011] A method for realizing a generative analysis report based on artificial intelligence algorithm technology comprises the following steps:

[0012] Build a B / S architecture system, which includes front-end server and back-end server;

[0013] Upload the data source through the front-end server and transfer it to the back-end server;

[0014] After receiving the data source, the backend server performs a preliminary analysis on the data source and processes it into data samples corresponding to the artificial intelligence large model;

[0015] The artificial intelligence model receives the data sample, processes the data, and generates the analysis report content;

[0016] The back-end server returns the generated analysis report content to the front-end server, and the front-end server displays the analysis report content in real time.

[0017] As a further solution of the present invention: the specific process of data processing by the artificial intelligence large model is:

[0018] Step 1: Pre-training model selection and initialization;

[0019] S101: After selecting the pre-trained model, set the system operating environment, including the hardware configuration of the backend server and the model framework;

[0020] S102: Initialize model-related parameters and dependent libraries to prepare for subsequent fine-tuning training;

[0021] Step 2: Model fine-tuning and training;

[0022] S201: Configure LoRA parameters;

[0023] S202: training set generation, i.e. generating tens of thousands of sample data based on the split data modules, and loading them into the SFTTrainer module of Huggingface TRL for training;

[0024] S203: Use high-performance GPU resources to repeatedly iterate the model and save the intermediate training state to ensure recoverability;

[0025] Step 3: Prompt engineering optimization;

[0026] S301: Diversify instructions and design different forms of instructions to improve the adaptability of the model;

[0027] S302: Testing the effect of the instruction, by comparing the model output with the expected content, gradually adjusting the instruction to generate text content that is more in line with human language habits and professional standards;

[0028] Step 4: Report generation and data visualization;

[0029] S401: Text generation: using the fine-tuned model to generate a logically rigorous and fluent report text based on the input data;

[0030] S402: Chart generation, generating bar charts and pie charts through a data visualization library for basic data in the report text;

[0031] S403: Report integration: integrate text and charts to output the final intelligent report.

[0032] As a further solution of the present invention: the specific process of the back-end server performing preliminary analysis on the data source is as follows:

[0033] Data classification: classify data according to its source and nature;

[0034] Format standardization: convert data into a unified format and remove redundant information to ensure data integrity and consistency;

[0035] Sample data generation: Generate training samples from the split basic data modules.

[0036] As a further solution of the present invention: the specific process of the back-end server returning the generated analysis report content to the front-end server is:

[0037] JSON format generation: The backend server packages the generated text and chart data into standard JSON format;

[0038] Return data: Return data to the front-end server through Axios.

[0039] As a further solution of the present invention: the specific process of the front-end server displaying the content of the analysis report is as follows:

[0040] Page rendering: Dynamic rendering is achieved using the VUE framework, so users can view report content in real time;

[0041] Interactive function: supports users to filter, search and export report contents;

[0042] Error handling: The front-end server provides error prompt function.

[0043] As a further solution of the present invention: after the analysis report content is generated, report verification and optimization are performed, specifically:

[0044] Semantic verification: Check the syntax and semantics of the generated report text to ensure that it complies with industry reporting standards;

[0045] Visualization optimization: Adjust the format of the generated charts to make them more in line with human reading habits.

[0046] As a further solution of the present invention: the specific process of the semantic verification is:

[0047] Use professional text checking software based on NLP technology as a tool for semantic verification;

[0048] Configure the tool parameters according to the characteristics of the report text and industry requirements, including the strictness of grammar checking and the context scope of semantic understanding;

[0049] Upload the generated report text content to the tool and wait for the tool to analyze and check it;

[0050] The tool will automatically perform a grammar check on the text, identifying and marking grammatical errors, including spelling errors, punctuation errors, and sentence structure errors;

[0051] Based on the grammatical check, the tool performs semantic check to identify and mark the unclear semantics, ambiguous expressions and those that do not conform to industry standards;

[0052] The tool automatically generates a detailed inspection report that lists all identified errors and recommended corrections;

[0053] Correct the identified errors according to the suggested correction plan and send the revised text to the back-end staff for review.

[0054] As a further solution of the present invention: the format adjustment in the visualization optimization includes adjusting the color contrast and the legend font size.

[0055] As a further solution of the present invention: the front-end server includes a VUE framework and an Axios module, the VUE framework is used to build an interactive interface of the front-end page, and the Axios module is used to configure the data source, upload Excel files and transmit data to the back-end server; through the front-end page, the user can easily upload the data source and view the generated report in real time.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] (1) Efficiency: Automatically generating reports through artificial intelligence models significantly improves the speed of report generation and reduces manual intervention and time costs.

[0058] (2) Accuracy and personalization: The model is fine-tuned to generate highly accurate and personalized reports based on different enterprise or industry data.

[0059] (3) Visualization support: Combined with data visualization technology, the report content not only includes text analysis, but also can intuitively display data through charts, enhancing the readability and comprehensibility of the report.

[0060] (4) Flexibility and scalability: This method can adapt to different types of data sources and can generate customized reports based on different needs, with high scalability and versatility.

[0061] (5) Fusion report generation for multimodal data: Supports extracting information from multimodal data sources (text, images, tables, audio, etc.) and generating reports, and achieves cross-modal data fusion and unified output through deep learning. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 The present invention is a flowchart of a method for implementing a generative analysis report based on artificial intelligence algorithm technology. DETAILED DESCRIPTION

[0063] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0064] As mentioned in the background technology of this application, research has found that existing report generation methods usually rely on manual writing, which has the defects of low efficiency, accuracy relying on manual experience, insufficient personalization and customization, and high difficulty in data analysis.

[0065] In order to solve the above-mentioned defects, the present application discloses a generative analysis report implementation method based on artificial intelligence algorithm technology. By constructing an efficient B / S architecture system and combining the collaborative work of the front-end and the back-end, the full process automation from data source uploading, processing, analysis to report generation and display is successfully realized. The system makes full use of the powerful text generation capability of the artificial intelligence big model and the intuitive display advantages of data visualization technology, and can quickly, accurately and personalized generate various types of reports, which is particularly suitable for financial analysis, corporate management, market research and other fields.

[0066] The following will describe in detail how the solution of the present application solves the above technical problems in conjunction with the accompanying drawings.

[0067] See also Figure 1 In an embodiment of the present invention, a method for realizing a generative analysis report based on artificial intelligence algorithm technology includes the following steps: constructing a B / S architecture system, the system including a front-end server and a back-end server; uploading a data source through the front-end server, and transmitting the data source to the back-end server; after receiving the data source, the back-end server performs a preliminary analysis on the data source and processes it into a data sample corresponding to the artificial intelligence large model; the artificial intelligence large model receives the data sample and performs data processing to generate the analysis report content; the back-end server returns the generated analysis report content to the front-end server, and the front-end server performs real-time display according to the analysis report content. This application successfully realizes the full process automation from uploading, processing, analysis to report generation and display of data sources by constructing an efficient B / S architecture system and combining the collaborative work of the front-end and the back-end. The system makes full use of the powerful text generation ability of the artificial intelligence large model and the intuitive display advantage of data visualization technology, and can quickly, accurately and personalized generate various types of reports, especially suitable for financial analysis, enterprise management, market research and other fields.

[0068] In this embodiment, the specific process of data processing by the artificial intelligence big model is as follows:

[0069] Step 1: Pre-training model selection and initialization;

[0070] S101: After selecting the pre-trained model, set up the system operating environment, including the hardware configuration of the backend server (such as GPU server support) and the model framework (such as Huggingface Transformers). For deep learning models, especially large-scale pre-trained models, GPU server support is essential. GPU (graphics processing unit) can accelerate the training and reasoning process of the model and significantly improve computing efficiency. Huggingface Transformers is a popular open source library that provides a large number of pre-trained models and supports a variety of natural language processing tasks;

[0071] S102: Initialize model-related parameters and dependent libraries to prepare for subsequent fine-tuning training; the present invention preferably uses Llama-3-8b-bnb-4bit as a pre-trained model. The model has been trained based on a large amount of general data and has good language understanding and generation capabilities. Through further fine-tuning, it is adapted to report generation tasks, especially vertical field tasks such as generative analysis reports.

[0072] Step 2: Model fine-tuning and training;

[0073] S201: Configure LoRA parameters;

[0074] r (rank): set to 8 to optimize fine-tuning cost and effect.

[0075] target_modules: Specifies the modules in the model that need special attention (such as self-attention layers).

[0076] lora_alpha (scaling factor): Set to 16 to ensure the effectiveness of weight adjustment.

[0077] lora_dropout: Set to 0.1 to increase the generalization ability of the model.

[0078] S202: training set generation, i.e., generating tens of thousands of sample data based on the split data modules, and loading them into the SFTTrainer module of Huggingface TRL for training. The SFTTrainer module of Huggingface TRL is a trainer designed for specific tasks such as Soft Prompt Tuning, and is built based on the Huggingface Transformers library;

[0079] S203: Use high-performance GPU resources to repeatedly iterate the model and save the intermediate state of training to ensure recoverability; fine-tuning is a key step to optimize the model for domain tasks. The present invention adopts LoRA (Low-Rank Adaptation of Large Language Models) technology, which has the characteristics of high efficiency and low cost.

[0080] Step 3: Prompt engineering optimization;

[0081] S301: Diversify instructions and design different forms of instructions to improve the adaptability of the model; for example:

[0082] "Generate an annual financial analysis report for a company."

[0083] "Generate a quarterly financial summary for the following data."

[0084] S302: Test the effect of the instruction, and gradually adjust the instruction by comparing the model output with the expected content, so that the generated text content is more in line with human language habits and professional standards; in order to further improve the quality of report generation, the present invention introduces prompt engineering to study and design the instruction template that is most suitable for generating reports.

[0085] Step 4: Report generation and data visualization;

[0086] S401: Text generation: using the fine-tuned model to generate a logically rigorous and fluent report text based on the input data;

[0087] S402: Chart generation, for basic data in the report text (such as balance sheet, income statement, etc.), generate bar charts, pie charts and dynamic charts through data visualization libraries (such as D3.js, Echarts). Among them, D3.js (Data-Driven Documents) is a powerful JavaScript library for manipulating documents through data. It allows you to bind arbitrary data to DOM and then apply data-driven transformations to documents. Echarts is an open source visualization library implemented in JavaScript, which provides simple, intuitive and powerful charts and visualization components.

[0088] S403: Report integration, formatting and integrating text and charts, and outputting the final intelligent report. Based on the generative text report, the present invention further combines data visualization technology to add intuitive chart display to the report to improve readability.

[0089] In this embodiment, the specific process of the back-end server performing preliminary analysis on the data source is as follows:

[0090] Data classification: Classify data according to its source and nature (such as financial statements, corporate indicators, market data, etc.);

[0091] Format standardization: Convert data into a unified format (such as CSV, JSON), remove redundant information, and ensure data integrity and consistency. CSV and JSON are two commonly used data formats. CSV is suitable for storing structured data and is easy to open and edit in spreadsheet software. However, it does not support nested data structures or complex data types. JSON is suitable for storing semi-structured or unstructured data, supports nested data structures, and is easy to use in Web applications.

[0092] Sample data generation: Generate training samples from the split basic data modules; the format is as follows:

[0093]

[0094] The data source is the basis for report generation. One of the innovations of the present invention is to split the complex original data source into multiple basic data modules to facilitate subsequent model training and generation.

[0095] In this embodiment, the specific process of the back-end server returning the generated analysis report content to the front-end server is as follows:

[0096] JSON format generation: The backend server packages the generated text and chart data into standard JSON format; for example:

[0097]

[0098]

[0099] Return data: Return data to the front-end server through Axios. This setting enables the back-end server to format the report generated by the model and return it to the front-end server in JSON format.

[0100] In this embodiment, the specific process of the front-end server displaying the analysis report content is as follows:

[0101] Page rendering: Dynamic rendering is achieved using the VUE framework, so users can view report content in real time;

[0102] Interactive function: supports users to filter, search and export report contents, such as downloading as PDF file or sending to email;

[0103] Error handling: The front-end server provides an error prompt function. For example, when data upload is incomplete or model generation fails, it prompts users with feasible solutions.

[0104] In this embodiment, after the analysis report content is generated, report verification and optimization are performed, specifically:

[0105] Semantic verification: Check the syntax and semantics of the generated report text to ensure that it complies with industry reporting standards;

[0106] Visualization optimization: Adjust the format of the generated charts to make them more in line with human reading habits, such as adjusting color contrast, legend font size, etc.

[0107] In this embodiment, the specific process of the semantic verification is:

[0108] Use professional text checking software based on NLP (natural language processing) technology as a tool for semantic verification;

[0109] Configure the tool parameters according to the characteristics of the report text and industry requirements, including the strictness of grammar checking and the context scope of semantic understanding;

[0110] Upload the generated report text content to the tool and wait for the tool to analyze and check;

[0111] The tool will automatically perform a grammar check on the text, identifying and marking grammatical errors, including spelling errors, punctuation errors, and sentence structure errors;

[0112] Based on the grammatical check, the tool performs semantic check to identify and mark the unclear semantics, ambiguous expressions and those that do not conform to industry standards;

[0113] The tool automatically generates a detailed inspection report that lists all identified errors and recommended corrections;

[0114] Correct the identified errors according to the suggested correction plan and send the revised text to the back-end staff for review.

[0115] In this embodiment, the front-end server includes a VUE framework and an Axios module. The VUE framework is used to build an interactive interface for the front-end page, and the Axios module is used to configure the data source, upload the Excel file, and transmit the data to the back-end server; through the front-end page, the user can easily upload the data source and view the generated report in real time. Among them, the VUE framework is a progressive JavaScript framework for building user interfaces. It is easy to get started and can be easily integrated with other libraries or existing projects. The core library of the VUE framework only focuses on the view layer, and is very easy to get started, and is also easy to integrate with third-party libraries or existing projects. Axios is a Promise-based HTTP client for browsers and node.js. It can send asynchronous HTTP requests to the back-end server and process the response. In this embodiment, when a user uploads an Excel file through a page created by the VUE framework, the Axios module captures this action and sends the file data to the back-end server for processing. At the same time, the Axios module can also be used to obtain the generated report data from the back-end server and display it to the user on the front-end page.

[0116] The present invention provides a method for realizing generative analysis reports based on artificial intelligence algorithm technology. By building an efficient B / S architecture system and combining the collaborative work of the front end and the back end, the whole process from uploading, processing, and analysis of data sources to report generation and display is successfully realized. The system makes full use of the powerful text generation ability of the artificial intelligence big model and the intuitive display advantages of data visualization technology, and can quickly, accurately, and personalized generate various types of reports, which are particularly suitable for financial analysis, enterprise management, market research and other fields.

[0117] The main advantages of the present invention are embodied in the following aspects:

[0118] 1. Automation and efficiency: Through the application of artificial intelligence big models, the system can automatically process and analyze large amounts of data and generate reports that meet industry standards, greatly improving the speed and efficiency of report generation and reducing manual intervention and time consumption.

[0119] 2. Intelligence and accuracy: Through fine-tuning of large models and refined processing of data samples, the system can generate high-quality, professional report content according to specific needs. The model is not only flexible and adaptable, but can also be optimized according to different business scenarios to ensure the accuracy and rationality of the report content.

[0120] 3. Visualization and interactivity: Combined with data visualization technology, the system not only provides text descriptions, but also generates various charts, making the report content more intuitive and easy to understand. At the same time, the front-end display part provides flexible interactive functions, and users can customize report content, filter data, and export reports according to their needs, which greatly improves the user experience.

[0121] 4. Flexibility and scalability: The system is designed to be highly flexible and can adapt to different data source formats and industry requirements. Whether it is financial statements, market data or internal enterprise data, the system can process them efficiently and generate highly targeted reports. In addition, the system also has good scalability, and new data processing modules and report generation templates can be easily added according to future business development.

[0122] 5. Easy to deploy and maintain: The technical solution of the present invention adopts a standard B / S architecture, and the deployment and maintenance of the system are very convenient. The front-end interface is friendly, and users can easily upload data and view reports through the browser; the back-end server adopts a modular design, which is easy to expand or upgrade according to business needs.

[0123] In summary, the present invention provides an efficient, intelligent, accurate and flexible solution for report generation by combining artificial intelligence, big data processing and visualization technology. Its application in the fields of corporate finance, market analysis, scientific research reports, etc. has broad prospects, can effectively reduce labor costs, improve the quality and real-time performance of reports, and help various industries achieve data-driven intelligent decision-making.

[0124] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

[0125] What is described above 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. A method for realizing a generative analysis report based on artificial intelligence algorithm technology, characterized in that: The following steps are involved: Build a B / S architecture system, which includes front-end server and back-end server; Upload the data source through the front-end server and transfer it to the back-end server; After receiving the data source, the backend server performs a preliminary analysis on the data source and processes it into data samples corresponding to the artificial intelligence large model; The AI ​​big model receives the data sample, processes the data and generates the analysis report content; The back-end server returns the generated analysis report content to the front-end server, and the front-end server displays the analysis report content in real time.

2. According to claim 1, a method for realizing a generative analysis report based on artificial intelligence algorithm technology is characterized in that: The specific process of data processing by the artificial intelligence big model is as follows: Step 1: Pre-training model selection and initialization; S101: After selecting the pre-trained model, set the system operating environment, including the hardware configuration of the backend server and the model framework; S102: Initialize model-related parameters and dependent libraries to prepare for subsequent fine-tuning training; Step 2: Model fine-tuning and training; S201: Configure LoRA parameters; S202: training set generation, i.e. generating tens of thousands of sample data based on the split data modules, and loading them into the SFTTrainer module of Huggingface TRL for training; S203: Use high-performance GPU resources to repeatedly iterate the model and save the intermediate training state to ensure recoverability; Step 3: Prompt engineering optimization; S301: Diversify instructions and design different forms of instructions to improve the adaptability of the model; S302: Testing the effect of the instruction, by comparing the model output with the expected content, gradually adjusting the instruction to generate text content that is more in line with human language habits and professional standards; Step 4: Report generation and data visualization; S401: Text generation: using the fine-tuned model to generate a logically rigorous and fluent report text based on the input data; S402: Chart generation, generating bar charts and pie charts through a data visualization library for basic data in the report text; S403: Report integration: integrate text and charts to output the final intelligent report.

3. The method for realizing a generative analysis report based on artificial intelligence algorithm technology according to claim 2 is characterized in that: The specific process of the back-end server performing preliminary analysis on the data source is as follows: Data classification: classify data according to its source and nature; Format standardization: convert data into a unified format and remove redundant information to ensure data integrity and consistency; Sample data generation: Generate training samples from the split basic data modules.

4. The method for realizing a generative analysis report based on artificial intelligence algorithm technology according to claim 3 is characterized in that: The specific process of the back-end server returning the generated analysis report content to the front-end server is as follows: JSON format generation: The backend server packages the generated text and chart data into standard JSON format; Return data: Return data to the front-end server through Axios.

5. The method for realizing a generative analysis report based on artificial intelligence algorithm technology according to claim 4 is characterized in that: The specific process of the front-end server displaying the content of the analysis report is as follows: Page rendering: Use the VUE framework to achieve dynamic rendering, so that users can view the report content in real time; Interactive function: supports users to filter, search and export report contents; Error handling: The front-end server provides error prompt function.

6. The method for realizing a generative analysis report based on artificial intelligence algorithm technology according to claim 5 is characterized in that: After the analysis report is generated, the report is verified and optimized, specifically: Semantic verification: Check the syntax and semantics of the generated report text to ensure that it complies with industry reporting standards; Visualization optimization: Adjust the format of the generated charts to make them more in line with human reading habits.

7. The method for realizing a generative analysis report based on artificial intelligence algorithm technology according to claim 6 is characterized in that: The specific process of the semantic verification is as follows: Use professional text checking software based on NLP technology as a tool for semantic verification; Configure the tool parameters according to the characteristics of the report text and industry requirements, including the strictness of grammar checking and the context scope of semantic understanding; Upload the generated report text content to the tool and wait for the tool to analyze and check; The tool will automatically perform a grammar check on the text, identifying and marking grammatical errors, including spelling errors, punctuation errors, and sentence structure errors; Based on the grammatical check, the tool performs semantic check to identify and mark the unclear semantics, ambiguous expressions and those that do not conform to industry standards; The tool automatically generates a detailed inspection report that lists all identified errors and recommended corrections; Correct the identified errors according to the suggested correction plan and send the revised text to the back-end staff for review.

8. The method for realizing a generative analysis report based on artificial intelligence algorithm technology according to claim 7 is characterized in that: The format adjustment in the visualization optimization includes adjusting the color contrast and the legend font size.

9. The method for realizing a generative analysis report based on artificial intelligence algorithm technology according to claim 8, characterized in that: The front-end server includes a VUE framework and an Axios module. The VUE framework is used to build an interactive interface for the front-end page, and the Axios module is used to configure the data source, upload Excel files, and transmit data to the back-end server. Through the front-end page, users can easily upload data sources and view generated reports in real time.

Images