摘要： 传统二部图投影和排序(BGPR)算法的推荐精度不高。为此，提出一种基于项目属性和项目度的BGPR算法。分析二部图投影和随机游走的特点，引入项目属性和项目度2个影响因子，通过对初始化向量和项目相似性的优化，设计个性化推荐算法。实验结果表明，该算法的推荐精度较高。

关键词: 个性化推荐, 二部图投影, 项目属性, 项目度, 随机游走

Abstract: The recommendation accuracy of traditional Bipartite Graph Projection and Ranking(BGPR) algorithm is not high. To solve this problem, this paper proposes BGPR algorithm based on item attribute and item degree. This paper analyzes the bipartite graph characteristics of projection and the random walk, introduces two influencing factors, such as the project properties and projects degree, and gets the personalized recommendation algorithm through the similarity of initialization vector and project optimization. Experimental results show that this algorithm can efficiently improve the recommendation precision of the algorithm.

Key words: personality recommendation, bipartite graph projection, item attribute, item degree, random walk

中图分类号: 

• TP301.6