CN111553550A - An evaluation method for data quality of electric power big data based on user behavior analysis - Google Patents

Abstract

The invention discloses a method for evaluating the data quality of electric power big data for user behavior analysis, comprising the following steps: S1: collecting historical network data of several users through a data acquisition module, and integrating the historical network data through a data integration module; S2: stratify the characteristic data; S3: use the simple random sampling method to sample each layer of data, and obtain multi-group stratified sampling data, summarize the multi-group stratified sampling data, and obtain data samples; S4: according to the central The preset rules of the processing module evaluate the data samples in multiple dimensions, obtain the index evaluation result corresponding to each evaluation index, and then comprehensively evaluate the multiple evaluation results according to the weight; The accuracy of the data evaluation is improved; then the multiple evaluation results are comprehensively evaluated according to the weight, which is beneficial to improve the accuracy of the evaluation results.

Description

An evaluation method for data quality of electric power big data based on user behavior analysis

technical field

The invention belongs to the technical field of data quality evaluation of electric power big data, and in particular relates to a data quality evaluation method of electric power big data aiming at user behavior analysis.

Background technique

With the progress and development of society, the use of electricity has become more and more extensive, and the problem of electricity shortage in various places needs to analyze the electricity consumption behavior of customers, and then use the analysis results to control the power supply and formulate scientific, Reasonable and personalized electricity guiding strategy. All kinds of electricity consumption data are collected to form big data. Once the quality of big data is unqualified or inaccurate, it is difficult to obtain accurate analysis results. For this reason, we propose an assessment of the data quality of electricity big data based on user behavior analysis. method to solve the problems mentioned in the above background art.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a method for evaluating the data quality of electric power big data for user behavior analysis, so as to solve the problems raised in the above background art.

In order to achieve the above purpose, the present invention provides the following technical solutions: a method for evaluating the data quality of electric power big data for user behavior analysis, comprising the following steps:

S1: Collect historical network data of several users through the data collection module, and integrate the historical network data through the data integration module;

S2: in the integrated and classified data, retrieve feature data according to the data features preset by the central processing module, and stratify the feature data;

S3: Sampling each layer of data using a simple random sampling method, and obtain multi-group stratified sampling data, summarize the multi-group stratified sampling data, and obtain a data sample;

S4: Evaluate the data samples in multiple dimensions according to the rules preset by the central processing module, obtain an index evaluation result corresponding to each evaluation index, and then comprehensively evaluate the multiple evaluation results according to the weight;

S5: Display the comprehensive evaluation results through the visualization module.

Preferably, the data integration module in the step S1 is used to filter the historical network data, the filtering includes removing abnormal data, classifying the data after the removal, and assigning weights to the classified data according to the categories.

Preferably, the removing of abnormal data includes removing data that does not have sample significance, removing inaccurate data, and removing data with large fluctuations before and after.

Preferably, the feature data in the step S2 includes the historical network data corresponding to a plurality of feature parameters.

Preferably, in the step S4, multiple dimensions include data access, accuracy, completeness, consistency and timeliness, and the accuracy includes data syntax accuracy, data semantic accuracy, and data accuracy measurement coverage , metadata accuracy, data range accuracy, and data value precision.

Preferably, the visualization module can conveniently view the comprehensive evaluation results, and is suitable for data evaluation business personnel who have no in-depth understanding of algorithms and interfaces to view the comprehensive evaluation results.

Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention provides a method for evaluating the data quality of electric power big data for user behavior analysis. The characteristic data is retrieved, and the characteristic data is stratified. Since the classified data may have different weights according to different categories, the accuracy of data evaluation is improved through the allocation of weights.

Sampling each layer of data using a simple random sampling method, and obtain multi-group stratified sampling data, summarize the multi-group stratified sampling data, obtain data samples, and perform multiple dimensions on the data samples according to the rules preset by the central processing module. The evaluation results of each evaluation index corresponding to each evaluation index are obtained, and then the multiple evaluation results are comprehensively evaluated according to the weight, which is conducive to improving the accuracy of the evaluation results;

The visualization module can conveniently view the comprehensive evaluation results, which is suitable for data evaluation business personnel who have no in-depth understanding of algorithms and interfaces to view the comprehensive evaluation results.

Description of drawings

FIG. 1 is a schematic flowchart of a method for evaluating data quality of electric power big data for user behavior analysis according to the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

Example 1

The present invention provides a method for evaluating the data quality of electric power big data for user behavior analysis as shown in FIG. 1, including the following steps:

S1: Collect historical network data of several users through the data collection module, and integrate the historical network data through the data integration module;

S2: in the integrated and classified data, retrieve feature data according to the data features preset by the central processing module, and stratify the feature data;

S3: Sampling each layer of data using a simple random sampling method, and obtain multi-group stratified sampling data, summarize the multi-group and stratified sampling data, and obtain a data sample;

S4: Evaluate the data samples in multiple dimensions according to the rules preset by the central processing module, obtain an index evaluation result corresponding to each evaluation index, and then comprehensively evaluate the multiple evaluation results according to the weight;

S5: Display the comprehensive evaluation results through the visualization module.

Specifically, the data integration module in the step S1 is used to filter the historical network data, and the filtering includes the elimination of abnormal data, the classification of the eliminated data, and the weight distribution of the classified data according to the categories. The data is classified to facilitate subsequent retrieval, and since the classified data may have different weights according to different categories, the accuracy of data evaluation is improved through the allocation of weights.

Specifically, the removal of abnormal data includes removal of data that does not have sample significance, removal of inaccurate data, and removal of data with large fluctuations before and after, which is conducive to improving the accuracy of the data and improving the subsequent evaluation results. accuracy.

Specifically, the feature data in the step S2 includes the historical network data corresponding to a plurality of feature parameters.

Specifically, the multiple dimensions in step S4 include data access, accuracy, completeness, consistency, and timeliness, and the accuracy includes data syntax accuracy, data semantic accuracy, and data accuracy measurement coverage , metadata accuracy, data range accuracy, and data value precision.

Specifically, the visualization module can conveniently view the comprehensive evaluation results, and is suitable for data evaluation business personnel who have no in-depth understanding of algorithms and interfaces to view the comprehensive evaluation results.

To sum up, compared with the prior art, the present invention improves the accuracy of data processing by integrating historical network data, retrieving characteristic data according to preset data characteristics, and layering the characteristic data. ;

Sampling each layer of data using a simple random sampling method, and obtain multi-group stratified sampling data, summarize the multi-group stratified sampling data, obtain data samples, and perform multiple dimensions on the data samples according to the rules preset by the central processing module. To obtain the evaluation results corresponding to each evaluation index, and then conduct a comprehensive evaluation of multiple evaluation results according to the weight, which is conducive to improving the accuracy of the evaluation results;

The visualization module can easily view the comprehensive evaluation results, which is suitable for data evaluation business personnel who have no in-depth understanding of algorithms and interfaces to view the comprehensive evaluation results.

Finally, it should be noted that the above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, the The technical solutions recorded in the foregoing embodiments can be modified, or some technical features thereof can be equivalently replaced, and any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included. within the protection scope of the present invention.

Claims (6)

1. a method for evaluating the quality of power big data data for user behavior analysis, is characterized in that: comprise the steps: S1: Collect historical network data of several users through the data collection module, and integrate the historical network data through the data integration module; S2: in the integrated and classified data, retrieve feature data according to the data features preset by the central processing module, and stratify the feature data; S3: Sampling each layer of data using a simple random sampling method, and obtain multi-group stratified sampling data, summarize the multi-group stratified sampling data, and obtain a data sample; S4: Evaluate the data samples in multiple dimensions according to the rules preset by the central processing module, obtain an index evaluation result corresponding to each evaluation index, and then comprehensively evaluate the multiple evaluation results according to the weight; S5: Display the comprehensive evaluation results through the visualization module. 2. A method for evaluating the quality of power big data data for user behavior analysis according to claim 1, wherein the data integration module in the step S1 is used to filter historical network data, and the filtering includes Elimination of abnormal data, classification of eliminated data, and weight distribution of classified data according to categories. 3. The method for evaluating the data quality of electric power big data for user behavior analysis according to claim 2, wherein the removing of abnormal data comprises removing data that does not have sample significance, removing data that does not Accurate data, remove data with large fluctuations before and after. 4 . The method for evaluating the data quality of electric power big data for user behavior analysis according to claim 1 , wherein the characteristic data in the step S2 includes the historical network data corresponding to a plurality of characteristic parameters. 5 . 5. A method for evaluating data quality of electric power big data for user behavior analysis according to claim 1, characterized in that: in the step S4, multiple dimensions include data access status, accuracy, completeness, consistency The accuracy includes data syntactic accuracy, data semantic accuracy, data accuracy measurement coverage, metadata accuracy, data range accuracy, and data value accuracy, and timeliness. 6. a kind of evaluation method for the power big data data quality of user behavior analysis according to claim 1, it is characterized in that: described visualization module is convenient to view comprehensive evaluation result, is suitable for algorithm and interface without in-depth understanding The data evaluation business personnel's view of the comprehensive evaluation results.

Images