【摘要】 为改进基于项目的协同过滤推荐算法的推荐效果,在项目相似性计算时引入项目类别因素的影响,得出新的推荐算法,即基于项目类别的修正条件概率相似性,并在此基础上提出集成语境信息的多维推荐模型。通过与相关相似性、余弦相似性和修正余弦相似性的数值实验对比,证明在数据比较稀疏的情况下,改进算法所获得的推荐效果有较大提高。更多还原

【Abstract】 To improve recommendation result of the item-based collaborative filtering algorithm,influence of product category in product similarity computation was introduced,and a new recommendation algorithm,i.e.Category-based Adjusted Conditional Probability similarity(CACP),was proposed.Base on this algorithm,multi-dimension recommendation model for integrated contextual information was also presented.Experiment was conducted to compare correlation similarity,cosine similarity and adjusted cosine similarity.Results showed that recommendation result of CACP was greatly improved especially in sparse data environment..更多还原