|Keyphrase extraction based on topic relevance and term association|
|Authors:||Decong Li, Sujian Li, Wenjie Li, Congyun Gu, Yun Li|
|Citation:||Journal of Information and Computational Science 7 (1): 293-299. 2010.|
|Publication type:||Journal article|
|Google Scholar cites:||Not available|
|Added by Wikilit team:||Yes|
|Article:||Google Scholar BASE PubMed|
|Other scholarly wikis:||AcaWiki Brede Wiki WikiPapers|
|Web search:||Bing Google Yahoo! — Google PDF|
Keyphrases are concise representation of documents and usually are extracted directly from the original text. This paper proposes a novel approach to extract keyphrases. This method proposes two metrics, named topic relevance and term association respectively, for determining whether a term is a keyphrase. Using Wikipedia knowledge and betweenness computation, we compute these two metrics and combine them to extract important phrases from the text. Experimental results show the effectiveness of the proposed approach for keyphrases extaction.
"In this paper, with the help of Wikipedia knowledge,we construct a semantic graph, based on which topic relevance and term associationq qre combined to extract keyphrases."
|Theory type:||Design and action|
|Wikipedia coverage:||Sample data|
|Research design:||Mathematical modeling, Statistical analysis|
|Data source:||Wikipedia pages|
|Collected data time dimension:||Cross-sectional|
|Unit of analysis:||Article|
|Wikipedia data extraction:||Live Wikipedia|
|Wikipedia page type:||Article|
"The keyphrase extraction approach proposed in this paper performs better in comparison to three other approaches."
"Wikipedia pages; documents
The Wikipedia language seems to be "English", but it seems not to be specified, so it should be "not specified"?
The study made an computer-based experiment to evaluate the performance of their keyphrase extraction method. if "computer-based experiment" is regarded as an "experiment" we should label this as an "experiment."