Exploiting relevance, coverage, and novelty for query-focused multi-document summarization
Exploiting relevance, coverage, and novelty for query-focused multi-document summarization
复制标题
利用相关性、覆盖范围和新颖性进行以查询为中心的多文档摘要
DOI:
10.1016/j.knosys.2013.02.015
复制
发表时间:
2013-07
影响因子:
8.8
通讯作者:
Shi, Zhongzhi
中科院分区:
文献类型:
--
作者:
Luo, Wenjuan;Zhuang, Fuzhen;He, Qing;Shi, Zhongzhi
Summarization plays an increasingly important role with the exponential document growth on the Web. Specifically, for query-focused summarization, there exist three challenges: (1) how to retrieve query relevant sentences; (2) how to concisely cover the main aspects (i.e., topics) in the document; and (3) how to balance these two requests. Specially for the issue relevance, many traditional summarization techniques assume that there is independent relevance between sentences, which may not hold in reality. In this paper, we go beyond this assumption and propose a novel Probabilistic-modeling Relevance, Coverage, and Novelty (PRCN) framework, which exploits a reference topic model incorporating user query for dependent relevance measurement. Along this line, topic coverage is also modeled under our framework. To further address the issues above, various sentence features regarding relevance and novelty are constructed as features, while moderate topic coverage are maintained through a greedy algorithm for topic balance. Finally, experiments on DUC2005 and DUC2006 datasets validate the effectiveness of the proposed method.
登录
查看更多内容
DOI:
--
发表时间:
--
期刊:
--
影响因子:
--
作者:
N. Metropolis
通讯作者:
N. Metropolis
影响因子:
--
作者:
SHANNON, CE
通讯作者:
SHANNON, CE
DOI:
10.1145/3130348.3130369
发表时间:
1998-08
期刊:
ACM SIGIR Forum
影响因子:
--
作者:
J. Carbonell;Jade Goldstein-Stewart
通讯作者:
J. Carbonell;Jade Goldstein-Stewart
DOI:
10.1007/978-94-017-0171-6_1
发表时间:
2003
期刊:
--
影响因子:
--
作者:
J. Lafferty;Chengxiang Zhai
通讯作者:
J. Lafferty;Chengxiang Zhai
DOI:
10.1145/1390334.1390384
发表时间:
2008-07
期刊:
--
影响因子:
--
作者:
Furu Wei;Wenjie Li;Q. Lu;Yanxiang He
通讯作者:
Furu Wei;Wenjie Li;Q. Lu;Yanxiang He