Rapid increase of the weighted shortest path length in key term concurrence network and its origin

Authors:
Lan Yin;Hua Yang;Donghong Ji;Mingyao Zhang;Hongmiao Wu
Affiliations:
School of Mathematics and Computer Science, Guizhou Normal University, Guiyang, China, Computer School, Wuhan University, Wuhan, China;College of Chinese Language and Literature, Wuhan University, Wuhan, China, School of Mathematics and Computer Science, Guizhou Normal University, Guiyang, China;School of Mathematics and Computer Science, Guizhou Normal University, Guiyang, China, Computer School, Wuhan University, Wuhan, China;College of Chinese Language and Literature, Wuhan University, Wuhan, China, School of Foreign Languages and Literature, Wuhan University, Wuhan, China;School of Foreign Languages and Literature, Wuhan University, Wuhan, China
Venue:
CLSW'12 Proceedings of the 13th Chinese conference on Chinese Lexical Semantics
Year:
2012

Citing 5
Cited 0

PageRank without hyperlinks: structural re-ranking using links induced by language models

Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval
A sentimental education: sentiment analysis using subjectivity summarization based on minimum cuts

ACL '04 Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics
Using random walks for question-focused sentence retrieval

HLT '05 Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing
Language independent extractive summarization

ACLdemo '05 Proceedings of the ACL 2005 on Interactive poster and demonstration sessions
LexRank: graph-based lexical centrality as salience in text summarization

Journal of Artificial Intelligence Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

In previous work, we constructed a Key Term Concurrence Network (KTCN) based on large-scale corpus with an attempt to apply weighted shortest path length to measure semantic relevance between terms. The parameter was tentatively used for query expansion in Information Retrieval task directed to complex user query expressed in natural language. The data obtained from the experiment demonstrated improved performance in the task. However, we also found that as more new expanded terms are appended to the vector of original query, the performance decreases drastically after reaching a peak. This paper respectively explains the causes of this phenomenon from two perspectives: the property of complex network property and corpus linguistics. Based on this conclusion, future work is directed towards how to improve our work.