Aiming at the deficiency of conventional recommendation method in short text message processing,this paper pro-posed a microblog recommendation method based on user interest model and conversation extraction.First,it applied a Ncut_weighted non-negative matrix factorization (Ncut_weighted NMF)to obtain user-interest matrix.And then used Single-Pass clustering based on frequency and correlation for conversation extraction to obtain microblog conversation.Experiments show that this method can effectively cluster micro-blogs and support micro-blog recommendation.%针对传统推荐方法在短文本处理方面的不足，提出一种基于用户兴趣模型与会话抽取算法的微博推荐方法。该方法应用基于归一化割加权NMF的微博用户兴趣模型获取用户—主题矩阵，产生用户感兴趣的微博主题，结合基于Single-Pass 聚类模型的会话在线抽取算法SPFC（single-pass based on frequency and correlation）获取微博的会话队列，并与用户感兴趣的微博主题进行相似度计算，最后得到实时的微博推荐结果。实验表明，此方法能有效地进行微博推荐。
展开▼
