摘要
为了解决用户评分数据稀疏性和用户最近邻寻找的准确性问题,提出了一种基于项目分类的协同过滤推荐改进算法。该算法首先利用项目分类信息为类内未评分项目预测评分值;然后通过计算类内用户间的相似度得到目标用户的最近邻居;最后进行推荐。实验结果表明,该算法可以准确地获取用户兴趣最近邻,有效地解决数据稀疏性问题;同时,该算法还极大地提高了系统的工作效率及可扩展性。
To overcome the drawbacks caused by the data sparseness and inaccurate of the user neighbors,this paper came up with an improved collaborative filtering recommendation algorithm,basing on the technique of item classification.The algorithm first rated the unrated items by applying the item classification,and then calculated the user similarity within classes for nearest-neighbors,after which it could recommend the items based on the final prediction.Experimental results show that this algorithm can not only improve the accuracy of nearest neighbor search,but also increase the efficiency and scalability of the system.
出处
《计算机应用研究》
CSCD
北大核心
2012年第2期493-496,共4页
Application Research of Computers
基金
中央高校研究生科技创新基金资助项目(CDJXS11180012)
关键词
项目分类
协同过滤
评分预测
兴趣最近邻
推荐系统
item classification
collaborative filtering
rating predication
interest nearest neighbors
recommendation systems
