Hypergraph Partition with Harmonic Average Top-N and PCA for Topic Detection

Author

Liu, Xinyue ; Ma, Fenglong ; Lin, Hongfei ; Shen, Hong

Author_Institution

Sch. of Comput. Sci. & Technol., Dalian Univ. of Technol., Dalian, China

fYear

2010

fDate

18-20 Dec. 2010

Firstpage

269

Lastpage

276

Abstract

An algorithm named SMHP is proposed, which aims at improving the efficiency of Topic Detection. In SMHP, a T-MI-TFIDF model is designed by introducing mutual information (MI) and enhancing the weight of terms in the title. Then VSM is constructed according to terms´ weight, and the dimension is reduced by combining H-TOPN and PCA. Then topics are grouped based on SMHP. Experiment results show the proposed methods are more suitable for clustering topics. SMHP with novel approaches can effectively solve the relationship of multiple stories problem and improve the accuracy of cluster results.

Keywords

graph theory; information retrieval; pattern clustering; principal component analysis; text analysis; H-TOPN; PCA; SMHP algorithm; T-MI-TFIDF model; VSM; clustering topics; harmonic average Top-N; hypergraph partition; multiple stories problem; mutual information; topic detection; Algorithm design and analysis; Clustering algorithms; Harmonic analysis; Mutual information; Noise; Partitioning algorithms; Principal component analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Parallel Architectures, Algorithms and Programming (PAAP), 2010 Third International Symposium on

Conference_Location

Dalian

Print_ISBN

978-1-4244-9482-8

Type

conf

DOI

10.1109/PAAP.2010.38

Filename

5715093