CN-122022548-A - Enterprise data analysis method and device based on artificial intelligence

CN122022548ACN 122022548 ACN122022548 ACN 122022548ACN-122022548-A

Abstract

The embodiment of the application provides an enterprise data analysis method and device based on artificial intelligence, which are used for realizing effective integration of data by innovatively designing a data processing system and performing quality control and standardized conversion. And constructing a knowledge analysis mechanism, and combining the atlas construction and feature extraction to establish a reliable analysis strategy. Report generation is introduced, and the comprehensiveness of analysis is ensured through state characterization and trend prediction. The method effectively solves the defects of the traditional technology in the aspects of data processing, knowledge analysis, report generation and the like, and provides technical support for enterprise data analysis.

Inventors

SHI YANYING
LIU GUOQIANG

Assignees

紫金诚征信有限公司

Dates

Publication Date: 20260512
Application Date: 20251230

Claims (10)

1. An artificial intelligence based enterprise data analysis method, the method comprising: Establishing an enterprise data access system comprising data acquisition scheduling and quality evaluation, executing internal database access and external platform data capture according to the access system, constructing a data quality monitoring and exception handling method, performing missing value complementation and exception correction by using the handling method to form a normalized multi-source data set, establishing a preprocessing conversion model, and generating a structured enterprise data matrix according to the conversion model; Establishing a large model analysis method for enhancing enterprise knowledge based on the data matrix, expanding model knowledge reserve through an enterprise knowledge graph construction and update mechanism, establishing a query processing framework comprising semantic analysis and multidimensional retrieval, extracting target data dimension according to the processing framework, constructing a feature calculation and association analysis method, and generating an enterprise index feature network by utilizing the analysis method; And establishing an enterprise analysis report generating system based on the characteristic network, establishing an enterprise operation condition characterization method comprising quantitative evaluation and attribution analysis, calculating index time sequence characteristics and associated parameters according to the characterization method, establishing an index interpretation and trend prediction model, and generating an analysis report comprising multidimensional operation state data and development trend data by using the prediction model.
2. The artificial intelligence based enterprise data analysis method of claim 1, wherein the establishing an enterprise data access hierarchy including data acquisition scheduling and quality assessment, performing internal database access and external platform data crawling according to the access hierarchy, and constructing a data quality monitoring and exception handling method, comprises: establishing a connection configuration library based on a data source access interface, grouping and summarizing a database access certificate and a platform interface token according to data types, establishing a timing execution plan and concurrency control rule, and generating a scheduling parameter table according to access frequency and data magnitude to form a data acquisition rule set containing access rights and scheduling strategies; And establishing a data acquisition task queue according to the data acquisition rule set, applying the scheduling parameter table to acquisition task priority sequencing, establishing a data acquisition executor, reading a target data source according to the executor, completing data extraction, and analyzing and storing an acquisition result according to a preset data structure template.
3. The artificial intelligence based enterprise data analysis method of claim 1, wherein the performing missing value completion and anomaly correction using the processing method forms a normalized multi-source dataset, builds a pre-processing transformation model, and generates a structured enterprise data matrix based on the transformation model, comprising: Establishing a data anomaly detection rule base, marking an anomaly data item according to a field value range and a business constraint condition, constructing a missing value complement strategy, calculating complement parameters according to time sequence correlation and a business rule, and using the complement parameters to generate a data correction table to form a data quality evaluation report containing anomaly marks and correction suggestions; and performing data cleaning based on the data quality evaluation report, replacing the target field with the complement value in the data correction table, establishing a data standardization conversion function, performing format unification and coding conversion on the data according to the field mapping relation, and generating a structured data table according to service classification.
4. The method for analyzing enterprise data based on artificial intelligence according to claim 1, wherein the method for analyzing the large model based on the data matrix to build the enterprise knowledge enhancement, expanding model knowledge reserves through an enterprise knowledge graph building and updating mechanism, and building a query processing framework comprising semantic analysis and multidimensional retrieval comprises: Establishing an enterprise domain dictionary and a business rule knowledge base, constructing an entity mapping table according to field association in a data matrix, organizing business indexes and entity attributes into knowledge triples, generating an initial knowledge graph based on the knowledge triples, establishing a knowledge graph updating mechanism, and dynamically expanding the knowledge graph according to newly-added data and rule change; And constructing a semantic understanding model based on the knowledge graph, marking the query sentence by word segmentation according to the business theme and the keywords, establishing a multi-dimensional retrieval rule, and performing deep traversal on the knowledge graph according to the retrieval rule to generate a query instruction set containing entity relations and attribute constraints.
5. The artificial intelligence based enterprise data analysis method of claim 1, wherein the extracting the target data dimension according to the processing framework, constructing a feature computation and association analysis method, and generating an enterprise index feature network using the analysis method, comprises: Establishing a multi-dimensional data extraction rule, positioning target data fields according to a query instruction set, constructing a dimension combination strategy, grouping service indexes according to time granularity and an organization level, calculating feature weights based on field correlation, and generating a feature extraction template comprising dimension mapping and weight parameters; And constructing an index calculation function based on the feature extraction template, carrying out aggregation operation on data according to dimension groups, establishing an index association measurement method, and generating a feature network model reflecting the index hierarchical structure and association relation according to the association strength between the time sequence correlation and the service logic calculation index.
6. The artificial intelligence based enterprise data analysis method of claim 1, wherein the establishing an enterprise analysis report generation system based on the feature network, constructing an enterprise business situation characterization method including quantitative evaluation and attribution analysis, comprises: Establishing an enterprise operation index evaluation rule base, constructing an evaluation dimension according to index association relations in a feature network, classifying key indexes according to service topics and importance degrees, calculating a reference threshold value based on index historical data, and generating an index evaluation matrix containing the evaluation dimension and threshold value conditions; And establishing an attribution analysis method based on the index evaluation matrix, reconstructing index change data according to time sequence, establishing an index influence path tracing rule, and carrying out upstream and downstream decomposition on index fluctuation according to the tracing rule to generate attribution analysis results comprising key index change sources and influence degrees.
7. The artificial intelligence based enterprise data analysis method of claim 1, wherein the calculating the time sequence characteristics and the associated parameters according to the characterization method, establishing an index interpretation and trend prediction model, and generating an analysis report including multidimensional business status data and development trend data using the prediction model, comprises: Establishing an index time sequence analysis rule base, constructing a time window sequence according to index historical data, slicing the index data according to service periods, calculating trend characteristic parameters based on sliding windows, carrying out quantitative calculation on time sequence correlation among indexes, and generating a time sequence characteristic matrix containing period characteristics and correlation coefficients; And constructing a trend prediction function based on the time sequence feature matrix, carrying out feature recombination on index data according to a prediction period, establishing index prediction constraint conditions, generating prediction parameters according to a historical fluctuation rule and a business boundary, and forming an analysis report template containing operation condition analysis results and trend prediction data.
8. An artificial intelligence based enterprise data analysis device, the device comprising: The data acquisition module is used for establishing an enterprise data access system comprising data acquisition scheduling and quality evaluation, executing internal database access and external platform data capture according to the access system, constructing a data quality monitoring and exception handling method, performing missing value complementation and exception correction by using the handling method to form a normalized multi-source data set, establishing a preprocessing conversion model, and generating a structured enterprise data matrix according to the conversion model; The characteristic processing module is used for establishing a large model analysis method for enhancing enterprise knowledge based on the data matrix, expanding model knowledge reserve through an enterprise knowledge graph construction and update mechanism, establishing a query processing framework comprising semantic analysis and multidimensional retrieval, extracting target data dimension according to the processing framework, constructing a characteristic calculation and association analysis method, and generating an enterprise index characteristic network by utilizing the analysis method; The data analysis module is used for establishing an enterprise analysis report generating system based on the characteristic network, establishing an enterprise operation condition characterization method comprising quantitative evaluation and attribution analysis, calculating index time sequence characteristics and associated parameters according to the characterization method, establishing an index interpretation and trend prediction model, and generating an analysis report comprising multidimensional operation state data and development trend data by utilizing the prediction model.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor performs the steps of the artificial intelligence based enterprise data analysis method of any one of claims 1 to 7 when the program is executed.
10. A computer readable storage medium having stored thereon a computer program, which when executed by a processor, implements the steps of the artificial intelligence based enterprise data analysis method of any one of claims 1 to 7.

Description

Enterprise data analysis method and device based on artificial intelligence Technical Field The application relates to the field of data processing, in particular to an enterprise data analysis method and device based on artificial intelligence. Background The existing enterprise data analysis method has obvious defects. The traditional system has poor performance in terms of data acquisition and quality processing, and cannot effectively integrate multi-source data, so that an analysis basis is affected. Furthermore, the prior art has bottlenecks in knowledge modeling and feature extraction. Most systems lack sophisticated knowledge enhancement mechanisms and analysis strategies, resulting in inadequate understanding. Existing systems have technology shortboards in report generation. The lack of in-depth analysis on enterprise operation is difficult to realize efficient trend prediction through model optimization, and decision support is affected. The resolution of these problems is of great importance to the enhancement of enterprise data analysis capabilities. Disclosure of Invention Aiming at the problems in the prior art, the application provides an enterprise data analysis method and device based on artificial intelligence, which can effectively solve the defects of the traditional technology in the aspects of data processing, knowledge analysis, report generation and the like and provide technical support for enterprise data analysis. In order to solve at least one of the problems, the application provides the following technical scheme: In a first aspect, the present application provides an artificial intelligence based enterprise data analysis method, including: Establishing an enterprise data access system comprising data acquisition scheduling and quality evaluation, executing internal database access and external platform data capture according to the access system, constructing a data quality monitoring and exception handling method, performing missing value complementation and exception correction by using the handling method to form a normalized multi-source data set, establishing a preprocessing conversion model, and generating a structured enterprise data matrix according to the conversion model; Establishing a large model analysis method for enhancing enterprise knowledge based on the data matrix, expanding model knowledge reserve through an enterprise knowledge graph construction and update mechanism, establishing a query processing framework comprising semantic analysis and multidimensional retrieval, extracting target data dimension according to the processing framework, constructing a feature calculation and association analysis method, and generating an enterprise index feature network by utilizing the analysis method; And establishing an enterprise analysis report generating system based on the characteristic network, establishing an enterprise operation condition characterization method comprising quantitative evaluation and attribution analysis, calculating index time sequence characteristics and associated parameters according to the characterization method, establishing an index interpretation and trend prediction model, and generating an analysis report comprising multidimensional operation state data and development trend data by using the prediction model. Further, a connection configuration library is established based on the data source access interface, the database access credentials and the platform interface token are grouped and summarized according to data types, a timing execution plan and concurrency control rule are established, a scheduling parameter table is generated according to the access frequency and the data magnitude, and a data acquisition rule set containing access rights and scheduling strategies is formed; And establishing a data acquisition task queue according to the data acquisition rule set, applying the scheduling parameter table to acquisition task priority sequencing, establishing a data acquisition executor, reading a target data source according to the executor, completing data extraction, and analyzing and storing an acquisition result according to a preset data structure template. Further, the method further comprises the steps of establishing a data anomaly detection rule base, marking an anomaly data item according to a field value range and a business constraint condition, constructing a missing value complement strategy, calculating complement parameters according to time sequence correlation and a business rule, and using the complement parameters to generate a data correction table to form a data quality evaluation report containing anomaly marks and correction suggestions; and performing data cleaning based on the data quality evaluation report, replacing the target field with the complement value in the data correction table, establishing a data standardization conversion function, performing format unification and coding conversion o