基于矩阵分解模型的微博好友推荐算法

微博作为一种实时的信息传播和分享的社交网络平台,对人们日常生活的影响越来越大.在微博中,用户可以通过关注关系,添加自己感兴趣的好友,扩大自己的交际圈.但如何推荐高质量的关注好友,一直是个性化服务的难点之一.针对此种情况,提出一种微博好友推荐算法,旨在为用户推荐高质量的关注用户.该算法是对基于Seeker-Source矩阵分解模型的一种改进算法.文中分析了微博用户的多种数据源信息,并给出了相应的特征提出方法,最后将这些特征引入到Seeker-Source矩阵分解模型中,通过对模型的优化求解,得到最佳的参数因子矩阵,从而完成好友推荐.在真实的微博数据集上的实验表明,本文所提出的算法取得了良好的效果.

Algorithm for Micro-blog User's Followee Recommendation Based on Matrix Factorization

Micro-blog is a social network platform that provides us a new communication and information sharing service. It has become more and more important in our daily life. An user can follow his interested friends to expand his social circle throw following relationship. But how to recommend high quality following users is always a difficulty of personalized service. For the issue, a Seeker-Source matrix factorization model based on micro-blog features is proposed in this paper. The algorithm is an improved algorithm which is based on "Seeker-Source". We extracted the characteristics of user's interest from each data source, and then introduced into the matrix factorization model which is suitable for recommending followee friends. Finally, we optimize the model and get the best factor parameter matrix to recommend followee friends. The experimental results carried on real data sets show that the proposed method performs better than the traditional matrix factorization model.