稀疏数据下基于用户偏好的协同过滤算法

协同过滤(collaborative filtering,CF)是推荐系统中最常用和最成功的推荐技术之一.现实中的数据往往比较稀疏,用户之间缺少共同评定项目,使一些传统的相似性度量无法进行计算;此外,传统的协同过滤算法忽视了用户偏好问题,这样会造成推荐精度的下降.针对这些问题,从用户全局项目和地方评级信息分析影响用户兴趣偏好的因素,通过计算用户评级信息在全局的概率分布和使用海明贴近度计算用户的兴趣偏好度,利用Jeffries-Matusita距离得出关于用户偏好的相似度算法,将相似度算法与加权的Jaccard相似度算法有效结合,提出了一种在稀疏数据下基于用户偏好的协同过滤算法模型.实验结果表明,提出的模型性能优于传统协同过滤算法,并且在更为稀疏的数据集上也有很高的准确率.

Collaborative filtering algorithm based on user preference in sparse data

Collaborative Filtering(CF) is one of the most popular and most successful recommendation techniques used in recommendation systems. However, Sparse data in reality will result in lack of common assessment items between users, which makes some traditional similarity measures unable to calculate; in addition, traditional collaborative filtering algorithm ignores the problem of user preference, which will lead to the decline in recommendation accuracy. To deal with these problems, this article analyzes the factors affecting user interest preferences from the users'' global items and local rating information.By calculating the global probability distribution of users'' rating information and using Hamming closeness to calculate users'' interest preference, the similarity algorithm about users'' preference is obtained by using Jeffries matusita distance. Combining the similarity algorithm with weighted Jaccard similarity algorithm, we propose a collaborative filtering algorithm model based on user preference in sparse data. The experimental results show that the performance of the proposed model is better than that of the traditional collaborative filtering algorithm, and it also has high accuracy on more sparse data sets.