英文摘要 |
Association rule mining discovers correlations among products from transactional data- bases. It can provide useful information to website operators for designing appropriate marketing activities, e.g., cross-selling. Web traversal pattern mining discovers access patterns among customers from web logs. It can provide useful information to customers for suggesting appropriate navigation paths and website operators for improving website structures. However, website operators consider not only the pure navigation behaviors, but also the purchasing behaviors of customers. In this point of view, this paper proposes a new algorithm for mining web transaction patterns. It can discover the associations among traveling and purchasing behaviors of customers and overcome the disadvantages of traditional methods. By this way, the enterprises can fully grasp the behaviors of customers and increase the income of website. The experimental results show that our proposed method can efficiently discover web transaction patterns. |