Personalized text summarization using NMF and cluster refinement

As accessing text information on the Internet has become popular, the needs for automatic personalized document summarization have increased. In this paper, a personalized document summarization method that uses Non-negative Matrix Factorization (NMF) and cluster refinement is proposed. The proposed method uses NMF with cluster refinement to summarize generic summary so that it can extract sentences covering the major topics of the document. In addition, the method can improve the quality of personalized summaries because the inherent semantics of the documents are well reflected with respect to user interest. The experimental results demonstrate that the proposed method achieves better performance the than other methods.