作者: Qin Lu , Wenjie Li , Yanxiang He , Furu Wei
关键词: Moral graph 、 Computer science 、 Information retrieval 、 Iterative method 、 Text graph 、 Graph (abstract data type) 、 Multi-document summarization 、 Cluster (physics) 、 Automatic summarization 、 Sentence
摘要: In this paper, we develop a novel cluster-sensitive graph model for query-oriented multi-document summarization. Upon it, an iterative algorithm, namely QoCsR, is built. As there existence of natural clusters in the case that document comprises collection sentences, suggest distinguishing intra- and inter-document sentence relations order to take into consideration influence cluster (i.e. document) global information on local evaluation. our model, five kinds are involved among three objects, i.e. document, query. Three them new normally ignored previous graph-based models. All these then appropriately formulated QoCsR algorithm though different ways. ROUGE evaluations shows can outperform best DUC 2005 participating systems.