Research On The Application Of Data Mining In E - Commerce Recommendation System

Electronic commerce industry face how to realize one-to-one personalized recommendation service that helps businesses and consumers to improve the quality of decision-making has become the research direction of the industry concern. The basic principle, the paper from the data mining and collaborative filtering algorithm and industry characteristics, in-depth analysis of the implementation process and effect evaluation of recommender systems, especially for innovative exploration in the recommendation to enhance the ability of the long tail of goods and avoid Matthew effect. Recommendation service excavate interest from consumer behavior data and associated data information of potential purchase intention, inductive preferences and characteristics of customer, to predict consumer buying behavior, online shopping decision, upgrade the shopping experience and optimization of human-machine interface, precision marketing, cross selling, improve the conversion rate.This paper first discusses the relevant theory of data association rules mining, clustering and collaborative filtering are systematically sorted and summarized, the electronic commerce industry as the breakthrough point, uses the data mining study and system detailed analysis of recommendation system theory perspective. Through the analysis of industry data characteristics, recommendation algorithm, application cases and recommend effect, trying to explain the e-commerce recommendation system of mining algorithm, the system architecture, solving scheme and the influence factors, put forward the assessment strategies and measures for improvement and optimization suggestions, realize the personalized service of high efficiency and accuracy in order to.