改进的个性化推荐算法

协同过滤技术是个性化推荐系统中最早也是最为成功的技术之一。但是随着电子商务系统用户数目和商品数目的日益增加，整个项目空间上用户评分数据极端稀疏，传统的CF（协同过滤）方法均存在各自的不足。本文分析了传统cF算法中存在的问题，对其相似性计算方法进行了改进，提出了一种优化的cF算法。实验结果表明，该算法同传统CF算法相比能显著提高推荐精度。

An adaptive algorithm for personal recommendation

Collaborative filtering is one of the earliest and the most successful technologies in personalizational recommendation systems. But with the development of E - commerce, the amgnitudes of users and rapid growth of commodities, it has resulted in the extreme sparsity of user rating data. Traditional CF methods work poor in this situation. In this paper, the deficiencies in the traditional CF algorithm have been analyzed, similarity calculation has been improved. Then an adaptive CF algorithm has been given out. The experimental results show that this method can noticeably provide better recommendation results than traditional CF algorithms.