基于浏览偏好挖掘的实时商品推荐方法

Real-time recommendation method based on browsing preferences mining

在分析了当前推荐技术中各种算法的优缺点和及其存在的主要问题的基础上,提出一种浏览偏好挖掘的实时商品推荐方法。该算法通过分析用户Web游览记录,并使用贝叶斯网预测其浏览偏好,然后将用户偏好与商品特征进行匹配计算进而产生商品推荐。实验表明该方法能为用户提供更为精确有效的个性化推荐。

After analyzing the advantages and disadvantages of various algorithms and the main problems of the current recommended technology, this paper put forward a real-time recommendation method based on user＇ s browsing preferences mining. The algorithm first used Bayesian Network （BN） to estimate user＇s preferences by analyzing his Web browsing history, and then generated recommendations by calculating the matching degree of user＇s preferences and the characteristics of goods. The experimental results indicate that this method can provide personal recommendations more accurately and efficiently.