Based On The Personalized Recommendation System Association Rules Research And Application

With the spread of the Internet, more and more consumer to join online shopping in the team, thus stimulating the electronic commerce system dynamic development. E-commerce for users provide more and more choices, and users also so often lost in a lot of commodity information, there is no way to accurately and rapidly to find their own needs of commodities. In this situation, electronic commerce recommendation system was born, it can directly to interact with users, simulation of the sales staff to users to recommend commodities, and help users find the necessary goods, users successfully completed transactions. Fierce market competition, electronic commerce recommendation system can effectively preserve user, prevent user loss, improve the electronic commerce system of the sales. In recent years, the electronic commerce recommendation system in both theory and practice has had great development. And today is individualized era, personalized service is the philosophical field customer satisfaction, but also the embodiment of modern enterprises to improve the core competitiveness of important ways. How to recommend him to different users interested commodities have belongs to a kind of personalized service. Such electronic commerce recommendation system should also realize the personalized recommendation. From the current situation, the personalized recommendation systems include top-selling products recommended, new recommendation, the related products recommended and with the user group with interest and recommend. This paper mainly discussed the related products recommended, it requires the recommended system from a large amount of historical data in sales study to find some commodity items between related degree, extracts rules, for the later recommended products provide the basis.The main content of this article are as follows:(1) introduces the current mainstream e-commerce personalized recommendation system and related technologies. (2) according to the website of the customer data analysis and processing characteristics, in many of the data mining algorithm chose association rules algorithm. Because of association rules is happened in according to the facts, according to things happen frequency discretion, so calculated results with high recommended the accuracy. (3) has the function of data mining, the personalized recommendation system platform is realized. Based on the comparison of various algorithm for mining association rules, using FP-tree algorithm for customer information and data mining. FP-tree algorithm of data mining technology used for personalized recommendation system.