基于关联规则的电子商务个性化推荐研究-管理科学与工程专业论文.docxVIP

基于关联规则的电子商务个性化推荐研究 RESEARCH ON PERSONALIZED RECOMMENDATION SYSTEM FOR E-COMMERCE BASED ON ASSOCIATION RULES ABSTRACT In recent years, e-commerce developed rapidly, with the rapid growth of commercial web sites, the overloaded information has brought a heavy processing burden for the users to choose their favorite products. In order to solve the nuisance caused by overloaded information, many e-commerce personalization recommendation systems committed to meet the different preferences of different users came into being. As one of the most widely used technology in personalized recommendation system, association rules have attracted more researchers’ attention. In this paper, based on the survey of mining association rules algorithm and mining fuzzy association rules algorithm, we point out that the association rule mining technology has achieved fruitful results, especially the fuzzy theory can express the real situation of the real world more accurately, but compared with the actual needs，there are still some shortcomings, including improving the quality of the rules, the dynamic association rule mining, the organization and handling of the association rules , the new extended association rules. For the current association rule mining algorithm does not take the profits of business issues into account. This paper presents a new type of extended association rule mining algorithm - weighted fuzzy-level association rule mining algorithm WFGAR. The algorithm takes the profit of the product as the weight of the project, takes the product sales volume into account in the excavation process. The business can recommend higher profitable commodity to users by the excavated association rules. This algorithm is more in line with the actual situation. In the artificial data set, this paper carries out simulated tests for the WFGAR. By varying the minimum support, transaction volume, fan-out and project number in the experiment, the performance of the WFGAR algorithm is tested.