Keyphrase Extraction from Chinese News Web Pages Based on Semantic Relations

Authors:
Fei Xie;Xindong Wu;Xue-Gang Hu;Fei-Yue Wang
Affiliations:
School of Computer Science and Information Engineering, Hefei University of Technology, Heifei, China 230009 and Department of Computer Science and Technology, Hefei Teachers College, Hefei, China ...;School of Computer Science and Information Engineering, Hefei University of Technology, Heifei, China 230009 and Department of Computer Science, University of Vermont, Burlington, U.S.A. VT 50405;School of Computer Science and Information Engineering, Hefei University of Technology, Heifei, China 230009;Institute of Automation, Chinese Academy of Sciences, Beijing, China
Venue:
PAISI, PACCF and SOCO '08 Proceedings of the IEEE ISI 2008 PAISI, PACCF, and SOCO international workshops on Intelligence and Security Informatics
Year:
2008

Citing 3
Cited 0

KEA: practical automatic keyphrase extraction

Proceedings of the fourth ACM conference on Digital libraries
Lexical cohesion computed by thesaural relations as an indicator of the structure of text

Computational Linguistics
Graph-based ranking algorithms for sentence extraction, applied to text summarization

ACLdemo '04 Proceedings of the ACL 2004 on Interactive poster and demonstration sessions

Quantified Score

Hi-index	0.00

Visualization

Abstract

Keyphrases are very useful for saving time on browsing through the news web pages. A new keyphrase extraction method from Chinese news web pages based on semantic relations is presented in this paper. Semantic relations between phrases are analyzed, and a lexical chain is used to construct a semantic relation graph. Keyphrases are extracted and a semantic link graph is built on the lexical chains. News web pages with core hints are selected from www.163.com to test our method. The experimental results show that the proposed method substantially outperforms the method based on term frequency, especially when the number of keyphrases extracted is 3 - the precision is improved by 26.97 percent, and the recall is improved by 20.93 percent.