Research and Application of Improved Apriori Algorithm to Electronic Commerce

In order to analyze the shopping habits of consumers and more accurately mine the characteristics of consumption, we hereby have proposed and proved Theorem 1-3 to improve the classical Apriori algorithm, resulting in the reduction of database access. We improved the efficiency in the frequent-item sets-based establishment of strong association rule. With this new design we fulfilled the timely recommendation of related products to customers, reflecting the principle of personalized service in e-shopping.