Query-focused multi-document summarization using hypergraph-based ranking

•We propose a novel hybrid method to capture group relation of sentences.•We cluster sentences with a KL-divergence based on word-topic distribution.•We proposed a vertex reinforcement random walk process in a hypergraph model.•The process simultaneously consider the query similarity, the centrality and the diversity of sentences.•We implement our framework and verify improvement over appropriate baselines.

General graph random walk has been successfully applied in multi-document summarization, but it has some limitations to process documents by this way. In this paper, we propose a novel hypergraph based vertex-reinforced random walk framework for multi-document summarization. The framework first exploits the Hierarchical Dirichlet Process (HDP) topic model to learn a word-topic probability distribution in sentences. Then the hypergraph is used to capture both cluster relationship based on the word-topic probability distribution and pairwise similarity among sentences. Finally, a time-variant random walk algorithm for hypergraphs is developed to rank sentences which ensures sentence diversity by vertex-reinforcement in summaries. Experimental results on the public available dataset demonstrate the effectiveness of our framework.