CN-115599857-B - Intelligent analysis method and device for enterprise data

CN115599857BCN 115599857 BCN115599857 BCN 115599857BCN-115599857-B

Abstract

The invention discloses an intelligent analysis method and device for enterprise data, the method comprises the steps of determining enterprise databases to be analyzed, converging all fields included in all enterprise databases to obtain a candidate field set, executing preprocessing operation on all candidate fields included in the candidate field set to obtain a candidate entity field set, inputting the candidate entity field into a target model for each candidate entity field included in the candidate entity field set based on a preset target model to obtain a field analysis result of the candidate entity field, and generating a target analysis result according to the field analysis result of all the candidate entity fields. Therefore, the method and the device can realize intelligent analysis of the main data of the enterprise, and are beneficial to improving the efficiency of data analysis and improving the accuracy of data analysis.

Inventors

MA HUANAN
SU ZHENWEN
LIANG GUANXIONG
YANG HEMING
LIU JUNFENG
ZHENG XUSHENG
CHEN HUARONG
GUAN XIAOMING
LI BIN
TANG PINGHAI

Assignees

广东省电信规划设计院有限公司

Dates

Publication Date: 20260508
Application Date: 20221026

Claims (8)

1. A method for intelligent analysis of enterprise data, the method comprising: Determining an enterprise database to be analyzed, and converging all fields included in all the enterprise databases to obtain a candidate field set; preprocessing all candidate fields included in the candidate field set to obtain a candidate entity field set; inputting the candidate entity field into a target model aiming at each candidate entity field included in the candidate entity field set based on a preset target model to obtain a field analysis result of the candidate entity field; Generating a target analysis result according to field analysis results of all the candidate entity fields, wherein the target analysis result is used for representing data analysis results of all the enterprise databases; The method further includes, for each candidate entity field included in the candidate entity field set, inputting the candidate entity field to the target model based on a preset target model, and before obtaining a field analysis result of the candidate entity field: determining all target factors and factor weights of each target factor, and determining a target data template, wherein the target factors are factors influencing enterprise database analysis; Constructing a target model based on all the target factors, factor weights of each target factor and the target data template; After generating the target analysis result according to the field analysis results of all the candidate entity fields, the method further comprises: Judging whether the target analysis result meets a preset analysis result condition according to the target analysis result; when the target analysis result is judged to not meet the preset analysis result condition, analyzing a target reason that the target analysis result does not meet the preset analysis result condition; Determining factors to be adjusted of the target model and determining target weights corresponding to the factors to be adjusted based on the target reasons; According to the target weight corresponding to each factor to be adjusted, updating the target model to obtain an updated target model; Wherein, according to the target analysis result, judging whether the target analysis result meets a preset analysis result condition, includes: determining the matching degree between the target analysis result and the target data template according to the target analysis result; When the matching degree is greater than or equal to a preset matching degree threshold value, determining that the target analysis result meets a preset analysis result condition; and when the matching degree is smaller than the preset matching degree threshold value, determining that the target analysis result does not meet the preset analysis result condition.
2. The method for intelligent analysis of enterprise data according to claim 1, wherein the performing a preprocessing operation on all candidate fields included in the candidate field set to obtain a candidate entity field set includes: analyzing each candidate field included in the candidate field set based on a preset field analysis algorithm to obtain a candidate analysis result of the candidate field, wherein the candidate analysis result is used for representing field information of the candidate field; and determining a candidate entity field set according to a candidate analysis result of the candidate field aiming at each candidate field included in the candidate field set.
3. The intelligent analysis method according to claim 2, wherein for each of the candidate fields included in the candidate field set, determining a candidate entity field set according to a candidate analysis result of the candidate field includes: For each candidate field included in the candidate field set, according to candidate analysis results of all the candidate fields, analyzing association information between the candidate field and each residual candidate field except the candidate field to obtain an association information set corresponding to the candidate field, wherein the association information is used for representing field relation between the candidate field and each residual candidate field; Determining a target residual candidate field corresponding to the candidate field according to the association information set corresponding to the candidate field aiming at each candidate field included in the candidate field set, and executing merging operation on the candidate field and the target residual candidate field corresponding to the candidate field to obtain a merging result of the candidate field; and determining a candidate entity field set according to all the merging results.
4. The method for intelligent analysis of enterprise data according to claim 3, wherein for each of the candidate fields included in the candidate field set, determining a target remaining candidate field corresponding to the candidate field according to the association information set corresponding to the candidate field, comprises: And judging whether target association information, of which the association information meets preset association conditions, exists in an association information set corresponding to each candidate field, wherein the candidate fields are included in the candidate field set, and determining all remaining candidate fields corresponding to the target association information as target remaining candidate fields when the target association information exists.
5. The intelligent analysis method of enterprise data according to claim 4, wherein, based on a preset target model, for each candidate entity field included in the candidate entity field set, the candidate entity field is input to the target model, and after obtaining a field analysis result of the candidate entity field, the method further comprises, before generating a target analysis result according to the field analysis results of all the candidate entity fields: sorting the field analysis results of all the candidate entity fields according to preset sorting conditions according to the field scores corresponding to the field analysis results of each candidate entity field included in the target analysis results to obtain a target sequence; generating a target analysis result according to the field analysis results of all the candidate entity fields, including: and generating a target analysis result according to the field analysis results of all the candidate entity fields and the target sequence.
6. An intelligent analysis device for enterprise data, the device comprising: the determining module is used for determining an enterprise database to be analyzed; The aggregation module is used for aggregating all the fields included in the enterprise database to obtain a candidate field set; the processing module is used for executing preprocessing operation on all candidate fields included in the candidate field set to obtain a candidate entity field set; The input module is used for inputting the candidate entity field into the target model aiming at each candidate entity field included in the candidate entity field set based on a preset target model to obtain a field analysis result of the candidate entity field; The generation module is used for generating target analysis results according to field analysis results of all the candidate entity fields, wherein the target analysis results are used for representing data analysis results of all the enterprise databases; The determining module is further configured to determine, when the input module is based on a preset target model, for each candidate entity field included in the candidate entity field set, input the candidate entity field to the target model, and before obtaining a field analysis result of the candidate entity field, determine factor weights of all target factors and each target factor, and determine a target data template, where the target factors are factors that have an influence on enterprise database analysis; The apparatus further comprises: the building module is used for building a target model based on all the target factors, the factor weight of each target factor and the target data template; The apparatus further comprises: The judging module is used for judging whether the target analysis result meets a preset analysis result condition according to the target analysis result after the generating module generates the target analysis result according to the field analysis results of all the candidate entity fields; The analysis module is used for analyzing target reasons that the target analysis result does not meet the preset analysis result conditions when the judgment module judges that the target analysis result does not meet the preset analysis result conditions; the determining module is further configured to determine factors to be adjusted of the target model and determine a target weight corresponding to each factor to be adjusted based on the target reason; The updating module is used for executing updating operation on the target model according to the target weight corresponding to each factor to be adjusted so as to obtain an updated target model; the specific mode of judging whether the target analysis result meets the preset analysis result condition according to the target analysis result by the judging module comprises the following steps: determining the matching degree between the target analysis result and the target data template according to the target analysis result; When the matching degree is greater than or equal to a preset matching degree threshold value, determining that the target analysis result meets a preset analysis result condition; and when the matching degree is smaller than the preset matching degree threshold value, determining that the target analysis result does not meet the preset analysis result condition.
7. An intelligent analysis device for enterprise data, the device comprising: A memory storing executable program code; A processor coupled to the memory; The processor invokes the executable program code stored in the memory to perform the intelligent analysis method of enterprise data as claimed in any one of claims 1-5.
8. A computer storage medium storing computer instructions which, when invoked, are operable to perform the method of intelligent analysis of enterprise data as claimed in any one of claims 1 to 5.

Description

Intelligent analysis method and device for enterprise data Technical Field The invention relates to the technical field of data analysis, in particular to an intelligent analysis method and device for enterprise data. Background In real life, management of enterprises is not separated from data processing. At present, the analysis and the processing of enterprise data require that the enterprise data is firstly gathered and traced, then manually analyzed and scored, and the enterprise data is cooperated with other business departments in the enterprise to carry out multi-round communication and confirmation. However, the large amount of data in enterprises, the existing method of manually analyzing the data is inefficient and easy to cause low accuracy of data analysis. It is important to provide a new method for analyzing enterprise data to improve the data analysis efficiency. Disclosure of Invention The technical problem to be solved by the invention is to provide the intelligent analysis method and the intelligent analysis device for the enterprise data, which can realize intelligent analysis of the enterprise center data, are beneficial to improving the efficiency of data analysis and are beneficial to improving the accuracy of data analysis. In order to solve the technical problem, the first aspect of the present invention discloses an intelligent analysis method for enterprise data, which comprises: Determining an enterprise database to be analyzed, and converging all fields included in all the enterprise databases to obtain a candidate field set; preprocessing all candidate fields included in the candidate field set to obtain a candidate entity field set; inputting the candidate entity field into a target model aiming at each candidate entity field included in the candidate entity field set based on a preset target model to obtain a field analysis result of the candidate entity field; And generating a target analysis result according to the field analysis results of all the candidate entity fields, wherein the target analysis result is used for representing the data analysis results of all the enterprise databases. In an optional implementation manner, in the first aspect of the present invention, the performing a preprocessing operation on all candidate fields included in the candidate field set to obtain a candidate entity field set includes: analyzing each candidate field included in the candidate field set based on a preset field analysis algorithm to obtain a candidate analysis result of the candidate field, wherein the candidate analysis result is used for representing field information of the candidate field; and determining a candidate entity field set according to a candidate analysis result of the candidate field aiming at each candidate field included in the candidate field set. As an optional implementation manner, in the first aspect of the present invention, the determining, for each of the candidate fields included in the candidate field set, a candidate entity field set according to a candidate analysis result of the candidate field includes: For each candidate field included in the candidate field set, according to candidate analysis results of all the candidate fields, analyzing association information between the candidate field and each residual candidate field except the candidate field to obtain an association information set corresponding to the candidate field, wherein the association information is used for representing field similarity relationship between the candidate field and each residual candidate field; Determining a target residual candidate field corresponding to the candidate field according to the association information set corresponding to the candidate field aiming at each candidate field included in the candidate field set, and executing merging operation on the candidate field and the target residual candidate field corresponding to the candidate field to obtain a merging result of the candidate field; and determining a candidate entity field set according to all the merging results. In an optional implementation manner, in a first aspect of the present invention, for each candidate field included in the candidate field set, determining, according to an association information set corresponding to the candidate field, a target remaining candidate field corresponding to the candidate field includes: And judging whether target association information, of which the association information meets preset association conditions, exists in an association information set corresponding to each candidate field, wherein the candidate fields are included in the candidate field set, and determining all remaining candidate fields corresponding to the target association information as target remaining candidate fields when the target association information exists. As an optional implementation manner, in the first aspect of the present invention, based on a preset target m