【摘要】 提出一种新的基于客户购物模型的推荐系统框架,它把推荐过程形式化为客户购物信息的知识表达、知识推理过程。该方法首先对客户的购物历史数据进行学习,得到贝叶斯网客户购物模型,然后结合客户当前的购物行为,提出并实施了一种基于概率推理的推荐算法。实验表明该算法能高效实时地为客户产生个性化的商品推荐集合,且在覆盖率和准确率方面优于某些传统方法。更多还原

【Abstract】 Presents a new recommendation framework based on customer shopping model. This framework formalizes the re￣commending process as knowledge representation of the customer shopping information and uncertainty knowledge inference process. Firstly,this approach builds a customer model of Bayesian network by learning from customer shopping history data, then presents a recommendation algorithm based on probability inference in combination with customer present shopping action. Experimental results demonstrate that this method can effectively and in real time generate an individual recommendation set of commodity, it is better than some traditional methods in rates of coverage and precision.更多还原