摘要
为了提高个性化推荐效果及预测准确度,特别是针对传统算法中评分矩阵过于稀疏等问题提出一种新颖的协同过滤算法。该算法首先利用RFM模型合理地筛选用户信息,其次通过黏性客户的消费记录稠密化用户—项目评分矩阵,并改进了传统相似度计算公式。通过仿真实验证实了算法的准确性,最后将其应用于一套具有个性化商品推荐功能的系统原型中,证明了该推荐算法的有效性及实用性。
In order to improve the accuracy of recommendation, especially the matrix score of personalized recommendation technology is too spars, a new recommendation algorithm was proposed. The advantages of this algorithm were mainly embodied in the following aspects. Firstly, the improved algorithm with RFM model was used to select the original customer in some condition, making the recommended source of data more accurate and efficient. Secondly, in the improved algorithm the customer consumption history records were filled to the matrix to improve the consistency of the matrix of score. Thirdly, the traditional Pearson similarity calculation formula was improved to make the search of target users of similar neighbor more accurate. Then the simulation experiment was carried on by using the improved algorithm. It can be proved that the improved algorithm is better than the traditional one in accuracy. At last, the improved algorithm was applied to a recommendation system with personalized recommendation function. It was shown that the recommendation algorithm was efficient and valid.
出处
《电信科学》
北大核心
2015年第9期103-111,共9页
Telecommunications Science
基金
北京市大学生研究训练项目(No.14010321029)~~
关键词
个性化推荐
协同过滤
评分矩阵
personalized recommendation, collaborative filtering, score matrix
