Topic Detection with Hypergraph Partition algorithm

Xinyue Liu, Fenglong Ma, Hongfei Lin

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

An algorithm named SMHP (Similarity Matrix based Hypergraph Partition) algorithm 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 Vector Space Model (VSM) is constructed according to terms' weight, and the dimension is reduced by combining H-TOPN and Principle Component Analysis (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.

Original languageEnglish (US)
Pages (from-to)2407-2415
Number of pages9
JournalJournal of Software
Volume6
Issue number12 SPEC. ISSUE
DOIs
StatePublished - Dec 19 2011

Fingerprint

Vector spaces
Experiments

All Science Journal Classification (ASJC) codes

  • Software
  • Human-Computer Interaction
  • Artificial Intelligence

Cite this

Liu, Xinyue ; Ma, Fenglong ; Lin, Hongfei. / Topic Detection with Hypergraph Partition algorithm. In: Journal of Software. 2011 ; Vol. 6, No. 12 SPEC. ISSUE. pp. 2407-2415.
@article{40dedd8ccda84b98af783415de908b56,
title = "Topic Detection with Hypergraph Partition algorithm",
abstract = "An algorithm named SMHP (Similarity Matrix based Hypergraph Partition) algorithm 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 Vector Space Model (VSM) is constructed according to terms' weight, and the dimension is reduced by combining H-TOPN and Principle Component Analysis (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.",
author = "Xinyue Liu and Fenglong Ma and Hongfei Lin",
year = "2011",
month = "12",
day = "19",
doi = "10.4304/jsw.6.12.2407-2415",
language = "English (US)",
volume = "6",
pages = "2407--2415",
journal = "Journal of Software",
issn = "1796-217X",
publisher = "Academy Publisher",
number = "12 SPEC. ISSUE",

}

Topic Detection with Hypergraph Partition algorithm. / Liu, Xinyue; Ma, Fenglong; Lin, Hongfei.

In: Journal of Software, Vol. 6, No. 12 SPEC. ISSUE, 19.12.2011, p. 2407-2415.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Topic Detection with Hypergraph Partition algorithm

AU - Liu, Xinyue

AU - Ma, Fenglong

AU - Lin, Hongfei

PY - 2011/12/19

Y1 - 2011/12/19

N2 - An algorithm named SMHP (Similarity Matrix based Hypergraph Partition) algorithm 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 Vector Space Model (VSM) is constructed according to terms' weight, and the dimension is reduced by combining H-TOPN and Principle Component Analysis (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.

AB - An algorithm named SMHP (Similarity Matrix based Hypergraph Partition) algorithm 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 Vector Space Model (VSM) is constructed according to terms' weight, and the dimension is reduced by combining H-TOPN and Principle Component Analysis (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.

UR - http://www.scopus.com/inward/record.url?scp=83455225119&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=83455225119&partnerID=8YFLogxK

U2 - 10.4304/jsw.6.12.2407-2415

DO - 10.4304/jsw.6.12.2407-2415

M3 - Article

AN - SCOPUS:83455225119

VL - 6

SP - 2407

EP - 2415

JO - Journal of Software

JF - Journal of Software

SN - 1796-217X

IS - 12 SPEC. ISSUE

ER -