Research On Algorithm Of Association Rules And Its Application In The Automobile Sale Website

With the rapid development of Internet, it has become an important way for people to receive information. The users' demands of that whether the website's design can according with users' interest leads to "personality". During the design of a website, it can meet users' needs if Personalized Recommendation could be implemented based on users' interests. According to Data Mining ideas and Web Data Mining technology, finding out users' browsing behavior patterns could implement personalized recommendation effectively.Based on the research of association rules' theory, the paper focuses on analyzing and comparing the Apriori algorithm and the FP-Growth algorithm. FP-growth algorithm has good adaptability with different length of rules, while it has a huge enhancement in the efficiency than Apriori algorithm. But the FP-Growth algorithm mines frequent itemsets by gradually producing the Conditional Pattern Base and Conditional Frequent Pattern Tree, thus affecting the frequency of mining. Considering the characteristics of the association rules for Personalized Recommendation, on the basis of the initial FP-tree structure, the method use IFreq-Set-Tree structure to store frequent itemsets. It generates anterior item as i-size(l