Method for preprocessing abnormal values of e-business sales amounts based on statistical discrimination process

Abstract

The invention discloses a method for preprocessing abnormal values of e-business sales amounts based on a statistical discrimination process. The preprocessing method comprises the following steps: improving data mining technologies and tools; performing preliminary verification on basic data; classifying abnormal values; enhancing comparison and elimination with a false information base, reducing missing data and noise data, and performing zero padding processing on real missing data; performing discrimination and verification on false data; performing verification processing on scalping data; comparing an acquired result with an abnormal database in a data acquisition process; and establishing a basic information base after forming a massive database, and performing batch processing on massive data. Compared with the prior art, the method disclosed by the invention has greater pertinence on abnormal e-business data, and ensures that the acquisition cycle can be shortened and the data accuracy can be greatly increased after the abnormal e-business data is checked; and moreover, the method is simple to operate and ensures that the time for customers to look up information can be saved.

Description

technical field [0001] The invention relates to the technical field of computer network data processing, in particular to a preprocessing method for abnormal values ​​of e-commerce sales based on a statistical discrimination method. Background technique [0002] The current e-commerce database is highly susceptible to noise, missing data and inconsistent data. In reality, e-commerce billing and merchants providing false information are also repeatedly prohibited. Low-quality data will lead to low-quality mining results, and low-quality basic data will directly lead to inability to make high-quality decisions. How to preprocess outliers in e-commerce data, improve data quality, and make efficient statistical decisions are issues that must be paid attention to when doing data analysis. [0003] The current e-commerce data outliers mainly exist in the following ways: [0004] 1) Missing data values, data noise, inconsistent data values, etc. due to omissions in data mining or...

AI Technical Summary

Problems solved by technology

[0004] 1) Missing data values, data noise, inconsistent data values, etc. due to omissions in data mining or other reasons

[0005] 2) Because the merchant provides false information and fabricates false sales records, the sales are too large

[0006] 3) Due to the merchant’s malicious swiping of orders, the sales volume is too large, which eventually leads to the overall sales being too large

Images