Customer Online Shopping Feature Extraction based on Data Mining Algorithm

Driven by the climax of the Internet, online shopping has brought people into a new shopping era and has also brought new impacts to enterprises. To improve the market competitiveness of enterprises, enterprises need to continuously mine customer behavior information. In the mining process, due to the high amount of customer behavior characteristics, the existing behavior mining processing has problems such as low acceleration and high error rate. Feature extraction customer behavior mining algorithm, this algorithm estimates the non-customer behavior and customer behavior in online shopping, iterates many times until convergence, and obtains the best mining result corresponding to the regression line and variance feature parameters, and completes the customer behavior mining. The simulation test proves that the proposed algorithm can improve the precision and recall rate, ensure the reliability and stability of customer behavior mining, and has certain use-value in practical applications.