Electronic information industry big data platform and data mining method

Abstract

The invention discloses an electronic information industry big data platform and a data mining method in the technical field of electronic information. The big data platform comprises a metadata management module, a data directory management module, a data acquisition management module, a data cleaning and preprocessing module, a data modeling module, a model calculation and optimization module and a result application module. The data mining method comprises the steps of business metadata collection, data cleaning and preprocessing, data model establishment, model training and parameter optimization, and result analysis. According to the scheme, by collecting and updating data in real time, the limitation of data static display in the prior art is overcome, transverse correlation analysis of upstream and downstream of an industrial chain is achieved through the data directory management and data modeling module, the data application depth is improved through the data mining module, the data modeling module achieves comprehensive evaluation of the comprehensive capacity of an enterprise, and the enterprise comprehensive capacity is comprehensively evaluated. According to the scheme, multiple advanced data mining algorithms are adopted, and the accuracy and efficiency of data modeling are improved.

Description

Technical Field

[0001] The present invention relates to the field of electronic information technology, and in particular to a big data platform and a data mining method for the electronic information industry. Background Art

[0002] With the rapid development of the electronic information industry, enterprises are increasingly dependent on data. In existing technologies, data analysis in the electronic information industry mainly focuses on the collection and display of static data. These technologies usually focus on the analysis of internal data of a single enterprise, such as financial statements, product specifications, etc., but lack in-depth understanding and analysis of the upstream and downstream industrial chains of enterprises.

[0003] The problems and limitations of existing technologies mainly include: 1. Existing technologies mainly focus on the static display of data and lack dynamic analysis and prediction capabilities; 2. Existing technologies fail to fully realize horizontal correlation analysis between enterprises, especially upstream and downstream of the industrial chain; 3. Existing technologies usually stay at the surface level in data mining and fail to deeply explore the potential value and correlation behind the data; 4. Existing technologies fail to realize multi-dimensional portrait analysis of enterprise capabilities, which limits the comprehensive assessment of enterprise comprehensive capabilities; 5. Existing technologies are relatively single in data modeling and algorithm application, and fail to make full use of advanced data mining algorithms.

[0004] In view of the limitations of the prior art, the present invention proposes a big data platform and a data mining method for the electronic information industry. Summary of the invention

[0005] The purpose of the present invention is to solve the problems raised in the above-mentioned background technology, and the present invention provides a big data platform and a data mining method for the electronic information industry.

[0006] In order to achieve the above-mentioned purpose, the present invention specifically adopts the following technical solutions:

[0007] A big data platform for the electronic information industry, comprising:

[0008] a. Metadata management module, used to define and attribute data elements in the electronic information industry;

[0009] b. Data directory management module, used to establish a data directory based on the inherent relevance of data;

[0010] c. Data collection management module, used to collect data through registration platforms, government reporting requirements and Internet public channels;

[0011] d. Data cleaning and preprocessing module, used to clean, remove duplicates, process outliers and convert formats of collected data;

[0012] e. Data modeling module, used to extract features such as enterprise information, product specification information, and contract information, and to build data models;

[0013] f. Model calculation and optimization module, used to use at least one data mining algorithm for model training and parameter optimization;

[0014] g. The result application module is used to demonstrate the production division relationship and the degree of cooperation and binding of production collaboration among enterprises in the industrial chain.

[0015] Furthermore, the data collection management module can collect and update data in real time to adapt to the rapid changes in the market environment.

[0016] Furthermore, the data cleaning and preprocessing module includes an automated data quality assessment tool.

[0017] Furthermore, the data modeling module can adopt a variety of data mining algorithms, including but not limited to decision trees, random forests, neural networks, etc.

[0018] Furthermore, the model calculation and optimization module can automatically adjust algorithm parameters according to model performance feedback.

[0019] Furthermore, the result application module can provide user-defined data visualization functions.

[0020] A data mining method for a big data platform in the electronic information industry comprises the following steps:

[0021] S1. Collect business metadata through the data collection management module;

[0022] S2, performing data cleaning and preprocessing through a data cleaning and preprocessing module;

[0023] S3, extracting data features and establishing data models through the data modeling module;

[0024] S4, model training and parameter optimization through model calculation and optimization module;

[0025] S5. Display the analysis results through the result application module.

[0026] Furthermore, the data collection step includes synchronously collecting data from multiple data sources.

[0027] Furthermore, the data cleaning and preprocessing step includes using a machine learning algorithm to identify and correct data anomalies.

[0028] Furthermore, the result application step includes customizing data reports and visual displays according to user needs.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] 1. This solution overcomes the limitations of static data display in existing technologies by collecting and updating data in real time, and provides dynamic analysis and prediction capabilities.

[0031] 2. Through data catalog management and data modeling modules, this solution can realize horizontal correlation analysis between enterprises, especially upstream and downstream of the industrial chain, thereby providing a more comprehensive perspective of the industrial chain.

[0032] 3. The data mining module of this solution is not limited to surface-level data aggregation, but deeply explores the potential value and relevance behind the data, thereby improving the depth of data application.

[0033] 4. This solution extracts multi-dimensional features such as enterprise information, product specification information and contract information through the data modeling module. This solution can achieve a comprehensive assessment of the comprehensive capabilities of the enterprise and provide richer information for decision-making.

[0034] 5. This solution uses a variety of advanced data mining algorithms including decision trees, random forests, and neural networks to improve the accuracy and efficiency of data modeling.

[0035] 6. The data cleaning and preprocessing module in this solution includes automated data quality assessment tools, which improves the efficiency and accuracy of data processing.

[0036] 7. The model calculation and optimization module in this solution can automatically adjust the algorithm parameters according to the model performance feedback to ensure that the model maintains optimal performance over time.

[0037] 8. The result application module in this solution provides user-defined data visualization functions, allowing users to customize data reports and visualization displays according to their own needs, improving user experience and data accessibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a schematic diagram of the structure of a big data platform for the electronic information industry in the present invention;

[0039] Figure 2 The present invention is a schematic diagram of the steps of a data mining method for a big data platform in the electronic information industry. DETAILED DESCRIPTION

[0040] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0041] The electronic information industry big data platform and data mining method provided in this embodiment are mainly used to solve the problems existing in the prior art, mainly including: 1. The prior art mainly focuses on the static display of data and lacks dynamic analysis and prediction capabilities; 2. The prior art fails to fully realize the horizontal correlation analysis between enterprises, especially the upstream and downstream of the industrial chain; 3. The prior art usually stays at the surface level in data mining and fails to deeply explore the potential value and correlation behind the data; 4. The prior art fails to realize the multi-dimensional portrait analysis of enterprise capabilities, which limits the comprehensive evaluation of the comprehensive capabilities of enterprises; 5. The prior art is relatively single in data modeling and algorithm application, and fails to make full use of advanced data mining algorithms; The following will be combined with Figure 1 and Figure 2 Make a detailed description:

[0042] Embodiment 1:

[0043] A big data platform for the electronic information industry, comprising:

[0044] a. Metadata management module, used to define and attribute data elements in the electronic information industry, ensuring the basic structure and standardized processing of data;

[0045] b. Data catalog management module, used to establish a data catalog based on the inherent relevance of the data, which helps to organize and index the data for subsequent analysis and retrieval;

[0046] c. Data collection management module, which is used to collect data through registration platforms, government reporting requirements and Internet public channels, realizes multi-source integration and real-time update of data, and improves the comprehensiveness and timeliness of data;

[0047] d. Data cleaning and preprocessing module, used to clean, remove duplicates, process outliers and convert formats of collected data, ensuring data quality and consistency;

[0048] e. Data modeling module, which is used to extract features such as enterprise information, product specification information and contract information, and establish data models, which provides a basis for in-depth analysis of the industrial chain;

[0049] f. Model calculation and optimization module, which is used to use at least one data mining algorithm to perform model training and parameter optimization, which not only improves the prediction accuracy of the model, but also achieves continuous optimization of model performance by automatically adjusting algorithm parameters;

[0050] g. The result application module is used to display the production division relationship and the degree of cooperation and binding of production collaboration among enterprises in the industrial chain. It provides user-defined data visualization functions, enabling users to customize data reports and visualization displays according to their own needs.

[0051] The big data platform has achieved the following beneficial effects: enhanced the real-time and dynamic analysis capabilities of data, improved data quality and analysis depth, realized comprehensive and multi-dimensional analysis of the industrial chain, and provided flexible data reporting and visualization, thereby providing strong data support and insights for decision-making in the electronic information industry.

[0052] The data collection management module in the first embodiment operates by collecting and updating data in real time to ensure the timeliness and accuracy of platform information. This real-time performance is achieved by continuously monitoring and integrating data from different sources, including activity data of the user registration platform, official data required by the government, and third-party data from public channels on the Internet. This continuous data flow provides the platform with a dynamic and constantly updated data pool, so that it can quickly respond to rapid changes in the market environment.

[0053] In this way, the data collection management module not only enhances the platform's sensitivity to market dynamics, but also improves the accuracy and reliability of data analysis. This real-time update capability is crucial for predicting market trends, assessing risks, and formulating rapid response strategies. In addition, real-time data collection also means that the platform can capture short-term market fluctuations and changes in consumer behavior, providing companies with a more accurate and timely decision-making support tool.

[0054] The data cleaning and preprocessing module in the first embodiment operates by integrating automated data quality assessment tools. These tools are started immediately after the data enters the platform and perform a series of quality checks and preprocessing operations on the data. These operations include data cleaning, deduplication, outlier detection, data standardization and format conversion to ensure data consistency and accuracy.

[0055] Automated data quality assessment tools automatically identify inconsistencies, missing values, errors or abnormal data points in the data through preset rules and algorithms, and take corresponding measures to correct or mark them. For example, it can identify and correct format errors, fill or delete missing values, and smooth or remove outliers, thereby improving the reliability of the data;

[0056] This automated processing not only improves the efficiency of data processing and reduces the need for manual intervention, but also ensures the effectiveness of data analysis results through continuous quality monitoring. The application of automated tools also means that data problems can be quickly identified and resolved, reducing the risk of analytical bias or wrong decisions caused by data quality issues.

[0057] The data modeling module in the first embodiment operates by using a variety of advanced data mining algorithms to process and analyze data. These algorithms include but are not limited to decision trees, random forests, neural networks, etc. The application of these algorithms enables the data modeling module to extract key features from the cleaned and preprocessed data and construct a data model that reflects multi-dimensional features such as enterprise information, product specifications, and contract information;

[0058] By using these algorithms, the data modeling module can identify patterns, trends and correlations in the data, thereby building more accurate and reliable models. For example, the decision tree algorithm can help understand the impact of data features on the outcome variable; the random forest algorithm improves the stability and accuracy of the model by building multiple decision trees and voting; the neural network algorithm can capture complex nonlinear relationships;

[0059] The application of multiple data mining algorithms improves the accuracy and reliability of model predictions; the diversity of algorithms enables the model to capture complex relationships and deep features in the data; cross-validation of different algorithms enhances the model's ability to generalize to new data; accurate data models provide decision makers with deeper data insights, thereby optimizing the decision-making process; the module can select appropriate algorithms based on different data characteristics to improve the model's adaptability and flexibility; using advanced algorithms, companies can discover new business opportunities and solutions and drive business innovation.

[0060] This data modeling method that integrates multiple algorithms not only improves the depth and breadth of data analysis, but also provides companies with richer and more flexible data analysis tools, helping them to better understand market and business conditions and thus stay ahead in the highly competitive electronic information industry.

[0061] The model calculation and optimization module mentioned in the first embodiment operates by using real-time feedback of model performance to automatically adjust the parameters of the data mining algorithm. This process is achieved through a closed-loop optimization mechanism, in which the performance of the model is continuously monitored, and according to preset performance indicators, such as accuracy, recall rate or F1 score, the system automatically adjusts the algorithm parameters to optimize the model;

[0062] This automated adjustment mechanism not only saves time and effort in manually adjusting parameters, but also can more accurately find the optimal parameter combination through rapid iteration, thereby improving the overall performance of the model. In addition, this dynamic adjustment capability enables the model to adapt to changes in data and maintain its long-term effectiveness and accuracy;

[0063] This module includes the following benefits: Automated parameter adjustment reduces the manual intervention and time required for model optimization; continuous performance monitoring and parameter adjustment ensure that the model can achieve higher accuracy and efficiency; the model can automatically adapt to changes in data to maintain its predictive ability over time; by continuously adjusting parameters, the risk of the model overfitting the training data is reduced; a self-optimizing model can provide more reliable results and enhance user confidence in the model output; the model calculation and optimization module supports continuous learning and improvement, ensuring that the model becomes more accurate over time.

[0064] Through this intelligent model calculation and optimization method, the big data platform can provide more powerful and flexible analysis capabilities, helping companies make more timely and accurate decisions in a rapidly changing market environment.

[0065] The result application module in the first embodiment operates by providing a user-defined data visualization function, allowing the user to display and analyze data according to their needs and preferences. This module enables the user to select different data views, chart types and display parameters through an interactive interface and visualization tools, so as to more intuitively understand the data and gain insights;

[0066] Users can choose appropriate visualization forms, such as bar charts, line charts, scatter plots, heat maps, or maps, according to their analysis objectives, and customize elements such as colors, labels, and scales to enhance the expressiveness and readability of visualizations. In addition, the result application module may also include advanced functions, such as data filtering, drilling, and time series analysis, to support more complex data analysis needs;

[0067] The benefits of using this module include: user-defined visualization functions improve the flexibility and satisfaction of users' interaction with data; through intuitive visualization, users can identify data patterns, trends and anomalies more quickly; clear data presentation helps decision makers understand analysis results more accurately and make more informed decisions; different users can customize data views according to their specific needs to meet personalized analysis perspectives; effective data visualization can quickly convey complex information and improve the efficiency and effectiveness of information transmission; users can explore data in depth through interactive tools to discover deeper insights.

[0068] Through this user-driven result application module, the big data platform not only provides powerful data analysis capabilities, but also greatly enhances users' understanding and utilization of data through intuitive and personalized data presentation, thereby achieving more efficient information management and decision support in the electronic information industry.

[0069] The specific usage of the big data platform for the electronic information industry can be summarized into the following consecutive steps:

[0070] Step 1: First, users use the data collection management module to synchronously collect business metadata from multiple data sources, including internal enterprise data, market data, user behavior data, etc., and use automated tools or API interfaces to achieve real-time data acquisition;

[0071] Step 2: The collected raw data will be sent to the data cleaning and preprocessing module, where the data will go through steps such as cleaning, deduplication, outlier processing, and format standardization. Machine learning algorithms will be used to help identify and correct anomalies in the data to ensure data quality.

[0072] Step 3: After data cleaning, the data modeling module will extract key data features, such as company information, product specifications, contract information, etc., and build a data model based on these features. Users can choose different data mining algorithms, such as decision trees, random forests, or neural networks, to build the model;

[0073] Step 4: The model calculation and optimization module will train and verify the established model. The system will automatically adjust the algorithm parameters according to the model performance feedback and optimize the model to improve the prediction accuracy and generalization ability;

[0074] Step 5: Finally, users obtain the analysis results through the result application module, which provides user-defined data visualization functions, allowing users to customize data reports and visualization displays according to their needs to intuitively understand the analysis results;

[0075] Step 6: Users can interact with the visualization results, such as clicking, dragging, or entering parameters to explore the data in depth, drill down, or filter data to obtain more detailed information;

[0076] Step 7: Users can generate customized data reports based on the analysis results. These reports can be used for internal decision support or as business presentation materials. The reports can be distributed to team members or stakeholders through the sharing function of the platform;

[0077] Step 8: As the market environment and data change, users can return to the data collection step and start a new round of analysis process to ensure the timeliness and accuracy of the model and analysis results.

[0078] Through this series of steps, companies in the electronic information industry can make full use of the big data platform to achieve full-process management from data collection to analysis and decision support, and enhance business insight and market competitiveness.

[0079] The main business process steps of the electronic information industry big data platform can be summarized as follows:

[0080] 1. Through the data collection management module, business metadata is collected from enterprise user registration platforms, government reporting requirements, and Internet public channels to achieve multi-source data integration;

[0081] 2. In the data cleaning and preprocessing module, the collected data is cleaned, deduplicated, processed for outliers, and format converted to ensure the accuracy and completeness of the data;

[0082] 3. The data modeling module extracts key data features based on core business data such as enterprises, products, and contracts, laying the foundation for modeling;

[0083] 4. Select appropriate modeling algorithms, such as decision trees, random forests, neural networks, etc., use classification algorithms, clustering algorithms, time series models, deep learning models, etc. to build corresponding relationships between the upstream and downstream industrial chains of enterprises;

[0084] 5. The model calculation and optimization module initializes product sales, calculates based on the model, analyzes the revenue changes of each company in the upstream and downstream industrial chains of the product, and continuously optimizes the model configuration parameters;

[0085] 6. The result application module uses model relationships to display the production division relationship of enterprises in the electronic information industry chain and the degree of production collaboration and binding between enterprises, so that users can predict the degree of impact on related enterprises in the industry chain based on changes in market terminal sales;

[0086] 7. Through the business process diagram, the entire process from data collection to result application is clearly displayed to help users understand how the platform operates;

[0087] 8. Users can use the result application module to customize data reports and visualizations according to their needs, and conduct in-depth analysis and decision support;

[0088] 9. The platform provides continuous monitoring and optimization functions to ensure that data and models remain up to date as the market and data change;

[0089] 10. Ultimately, through the entire business process, the platform supports users to discover new business insights, provides decision support, helps companies grasp market trends, and optimizes strategic planning.

[0090] These business process steps reflect the complete chain from the collection of raw data to final decision support, ensuring that enterprises can make accurate data-driven decisions in the electronic information industry.

[0091] Embodiment 2:

[0092] Embodiment 2 proposes a data mining method for a big data platform in the electronic information industry. The operation mode is to perform a series of orderly steps to ensure the efficiency and accuracy of the entire process from data collection to analysis. The following are the specific steps and beneficial effects of the method:

[0093] S1. First, collect business metadata through the data collection management module. This step ensures the originality and comprehensiveness of the data and lays a solid foundation for subsequent analysis;

[0094] S2. Subsequently, the data cleaning and preprocessing module performs necessary cleaning and preprocessing on the collected data, including removing noise, filling missing values, and unifying the format, so as to improve the data quality;

[0095] S3. Next, the data modeling module extracts key features from the data and builds a data model based on these features. This step is the core of data analysis and directly affects the depth and breadth of the analysis results.

[0096] S4. Then, the model calculation and optimization module trains the established model and automatically adjusts the algorithm parameters according to the model performance feedback to optimize the prediction ability and accuracy of the model;

[0097] S5. Finally, the analysis results are displayed through the result application module. This step enables users to intuitively understand the analysis results through customized data visualization functions.

[0098] This method ensures the orderliness and repeatability of the data mining process through clear steps, thereby improving work efficiency; this method improves data availability and the reliability of analysis results through data cleaning and preprocessing; the data modeling step in this method deeply mines the intrinsic characteristics of the data, providing deeper insights for decision-making; the automated model training and parameter optimization mechanism in this method improves the prediction accuracy and adaptability of the model; the user-defined visualization result display in this method enhances the communication effect of information, allowing non-technical users to easily understand complex analysis results; the entire data mining method in this method supports the adjustment of each step according to different business needs, and has high flexibility and adaptability.

[0099] Through this data mining method, companies in the electronic information industry can extract valuable information from large amounts of data more efficiently and accurately, supporting them in making fast and accurate business decisions in a highly competitive market.

[0100] The data collection step in the second embodiment operates by synchronously collecting data from multiple data sources, which is the basis of the entire data mining method. Through synchronous collection, this step ensures the diversity and comprehensiveness of the data, providing rich original materials for subsequent analysis;

[0101] In this step, the data collection management module may involve interfacing with different data sources, monitoring data updates in real time, and using efficient data extraction technology to ensure the real-time and synchronization of data collection. The synchronously collected data may include user behavior data, transaction records, social media information, market research reports, etc. The diversity of these data sources provides a multi-angle perspective for in-depth analysis.

[0102] This data collection step has multiple benefits, including: collecting data from multiple data sources, increasing the dimension and depth of the data, and providing richer information for analysis; synchronous collection ensures the timeliness of the data, so that the analysis results can reflect the latest market and business conditions; synchronous collection of multiple data sources helps maintain data consistency and reduces differences caused by different data update times; the ability to select and integrate different data sources as needed improves the flexibility and scalability of data collection; multi-source data collection reduces the risk of affecting the entire analysis process due to problems with a single data source.

[0103] Through this multi-source synchronous data collection method, the big data platform can more comprehensively capture and reflect the dynamics of the electronic information industry, providing enterprises with a solid and reliable data foundation, thereby supporting more accurate and in-depth data analysis and decision-making.

[0104] The data cleaning and preprocessing step in the second embodiment uses machine learning algorithms to identify and correct data anomalies, such as cluster analysis, outlier detection algorithms, etc., to automatically identify anomalies and inconsistencies in the data, which may include values ​​that deviate from the normal range, data entry errors, missing values, or format errors. Once these anomalies are identified, the system will take corresponding measures, such as using statistical methods to fill missing values, remove or correct outliers, to ensure the consistency and accuracy of the data, thereby reducing human errors and improving analysis efficiency;

[0105] Automated anomaly detection not only speeds up data preprocessing, but also enhances the stability and predictive power of the model, because the cleaned data is cleaner. In addition, this highly adaptable anomaly handling capability can flexibly respond to different data characteristics, reduce the cost of subsequent data processing, ensure the reliability of data analysis results, and provide strong support for decision-making;

[0106] Through this data cleaning and preprocessing method combined with machine learning technology, the big data platform can ensure high data quality standards and provide a solid data analysis foundation for companies in the electronic information industry, making data analysis results more trustworthy and providing strong support for decision-making.

[0107] The result application step of the second embodiment is to customize the data report and visual display according to the user's needs. This step is the key link to convert the data analysis results into information that the user can understand and operate. Through this step, the user can customize the personalized data report and interactive visual display according to his or her specific needs and preferences;

[0108] In this step, the result application module provides a wealth of customization options, including selecting different data views, chart types, color schemes and display parameters, as well as setting conditions such as data filtering, sorting and time series. Users can use these tools to deeply analyze data, discover key trends and patterns, and present these findings in an intuitive way, making insights from complex data easy to understand and share;

[0109] This customized result application not only improves the user experience and makes data analysis closer to the user's actual application scenarios, but also enhances the communication effect of information through visualization, so that non-technical users can easily grasp the core content of data analysis. In addition, customized data reports and visualizations also support users to use data more effectively in presentations, reports or decision-making discussions.

[0110] Through this result application step, the big data platform not only provides in-depth data analysis capabilities, but also greatly enhances users' understanding and application of data through customized data reports and visualization, thereby achieving more efficient information management and decision support in the electronic information industry.

[0111] 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. A big data platform for the electronic information industry, characterized in that: include: a. Metadata management module, used to define and attribute data elements in the electronic information industry; b. Data directory management module, used to establish a data directory based on the inherent relevance of data; c. Data collection management module, used to collect data through registration platforms, government reporting requirements and Internet public channels; d. Data cleaning and preprocessing module, used to clean, remove duplicates, process outliers and convert formats of collected data; e. Data modeling module, used to extract features such as enterprise information, product specification information, and contract information, and establish data models; f. Model calculation and optimization module, used to use at least one data mining algorithm for model training and parameter optimization; g. The result application module is used to demonstrate the production division relationship and the degree of cooperation and binding of production collaboration among enterprises in the industrial chain.

2. The electronic information industry big data platform according to claim 1, characterized in that: The data collection management module can collect and update data in real time to adapt to the rapid changes in the market environment.

3. The electronic information industry big data platform according to claim 1, characterized in that: The data cleaning and preprocessing module includes an automated data quality assessment tool.

4. The electronic information industry big data platform according to claim 1, characterized in that: The data modeling module can adopt a variety of data mining algorithms, including but not limited to decision trees, random forests, neural networks, etc.

5. The electronic information industry big data platform according to claim 1, characterized in that: The model calculation and optimization module can automatically adjust algorithm parameters according to model performance feedback.

6. The electronic information industry big data platform according to claim 1, characterized in that: The result application module can provide user-defined data visualization functions.

7. A data mining method for a big data platform in the electronic information industry according to claims 1-6, characterized in that: The following steps are involved: S1. Collect business metadata through the data collection management module; S2, performing data cleaning and preprocessing through a data cleaning and preprocessing module; S3, extracting data features and establishing data models through the data modeling module; S4, model training and parameter optimization through model calculation and optimization module; S5. Display the analysis results through the result application module.

8. The data mining method according to claim 7, characterized in that: The data collection step includes synchronously collecting data from multiple data sources.

9. The data mining method according to claim 7, characterized in that: The data cleaning and preprocessing steps include using machine learning algorithms to identify and correct data anomalies.

10. The data mining method according to claim 7, characterized in that: The result application step includes customizing data reports and visual displays according to user needs.

Images