摘要
SlopeOne算法是一种快速高效基于项目的协同过滤算法,但它忽略用户之间的相关性与兴趣一致性.因此,提出融合用户特征的加权SlopeOne算法,首先是根据用户特征使用K-means算法将用户聚类并筛选近邻用户集,然后在Weighted SlopeOne算法的基础上将用户年龄特征融入权重改进预测评分,生成推荐列表,最后在MovieLens电影数据集中进行验证.实验表明,其算法一定程度上克服了传统协同过滤算法的数据稀疏性问题,提高了算法的准确性和稳定性.
SlopeOne algorithm is a fast and efficient item-based collaborative filtering algorithm,but it ignores the correlation and interest consistency between users.In the paper a weighted SlopeOne algorithm is proposed that combines users’ features.First,K-means algorithm is used to cluster users according to users’ features and the nearest neighbor user set is selected.Then user age feature is integrated into the weight to improve score prediction based on the weighted SlopeOne algorithm and a list of recommendations is generated.Finally,the algorithm is verified on MovieLens dataset.The experiments show that the algorithm overcomes the problem of data sparseness in traditional collaborative filtering algorithm to a certain extent,and improves the accuracy and stability of the algorithm.
作者
刘文霞
乔秀雯
LIU Wenxia;QIAO Xiuwen(School of Mathematies and Computer Science,Quanzhou Normal University,Fujian 362000,China)
出处
《泉州师范学院学报》
2018年第6期61-64,共4页
Journal of Quanzhou Normal University
