Based On Association Rules Of The E-commerce Recommendation Intelligence System Research

With the popularity of the Internet, people from the traditional shopping methodshave gradually transferred to Internet shopping. Patrons makes more and more users in thecase, e-commerce systems become way more user choice. However, as the web sitestructure has become increasingly complex, as well as more and more information aboutcommodity types is also an important issue. Users will often be lost in the mass ofinformation in space, there will be no way to identify the goods they need. Therefore, aquality and efficient e-commerce recommendation systems how to improve thecompetitiveness of the e-commerce website and sales capabilities, effectively retainingcustomers, has gradually become a study on the key technology of e-commerce Web site,a vertical focus of researchers.All major e-commerce site through the recommendation system to provide users withdiverse referral service, recommended technology makes good benefits gained in practicalapplication. At the same time as the exponential growth in the number of users and sitecontent expanding the size of the structure, also brings to the recommendation systemsuch as the recommended efficiency recommended accuracy challenges. In this article thefollowing points are recommended for E-Commerce the main problems encountered andthe techniques involved in analysis and study.This article begins with an analysis of e-commerce and data mining research at homeand abroad, and elaborated on the significance of this topic. Second detail the technologiesinvolved in e-commerce recommender system including data mining and Web mining.Data mining techniques described in detail, and describes the function of data mining, themining process, common algorithms and basic tasks. Web data mining in the role ofe-commerce recommendation systems. Through a simple e-commerce recommendersystem model describes the workflow for its specific and concrete application of thetechnology. Finally, is also key to this article, by subparagraph set of Apriori algorithmand based on an improved algorithm based on Boolean matrix. And improved algorithmsare applied in Recommender Systems so that it can increase the system’s recommendedquality, increase business sales.Apriori algorithm deficiencies still exist in this article, while providing real-timerecommendation service, for how efficiently to improve recommendation systems ofquality also recommended further research needs to be done. How this article can be usedwith improved algorithm combined with other recommendation algorithm, allows theadvantages of their respective algorithms to better play in Recommender Systems researchdirection is the next step.