Name
Abstract | ||
|---|---|---|
With the exponential growth of the web documents and the requirement of limited bandwidth for mobile devices, it becomes more and more difficult for users to get information they look forward to from the vast amount of information. Query-focused summarization gets more attention from both the research and engineering area in recent years. However, existing query-focused summarization methods don't consider the conceptual relation and the concept importance that make up the sentences, a concept is the title of a wikipedia article and can express an entity or action. In this article. We propose a novel method called Query-focused Multi-document Summarization based on Concept Importance (QMSCI). We first map sentence to concepts and get ranked weighted concepts by reinforcement between the concepts of sentences and concepts of the query in a bipartite graph, then we use the ranked weighted concepts to help to rank the sentences in a hyper-graph model, sentences that contain important concepts, related with the query and also central among sentences are ranked higher and comprise the summary. We experiment on the DUC datasets, the experimental result demonstrates the effectiveness of our proposed method compared to the state-of-art methods. |
Year | DOI | Venue |
|---|---|---|
2016 | 10.1007/978-3-319-31750-2_35 | ADVANCES IN KNOWLEDGE DISCOVERY AND DATA MINING, PAKDD 2016, PT II |
Keywords | Field | DocType |
Query-focused summarization,Bipartite-graph model,Hyper-graph model | Data mining,Computer science,Artificial intelligence,Natural language processing,Multi-document summarization,Automatic summarization,Ranking,Information retrieval,Bipartite graph,Bandwidth (signal processing),Mobile device,Sentence,Machine learning | Conference |
Volume | ISSN | ISBN |
9652 | 0302-9743 | 978-3-319-31750-2; 978-3-319-31749-6 |
Citations | PageRank | References |
0 | 0.34 | 12 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Zheng Hai-Tao | 1 | 142 | 24.39 |
| Ji-Min Guo | 2 | 0 | 0.34 |
| Jiang Yong | 3 | 156 | 41.60 |
| Xia Shu-Tao | 4 | 342 | 75.29 |