ePattern discovery of topics in network forum based on wavelet analysis and neural network

Hong Zhang, Hua Zhong*, Wei Cheng

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Network forum, one of main sources of the public opinions on the Internet, attracts more and more researchers to focus on this new area. The pattern discovery of topics will help the researchers understand more about user behavioral. The method proposed in this paper is the combined wavelet multi-resolution analysis and the BP neural network algorithm to discover patterns of topic trend which can be used to predict. The method has been tested in experiment by real-world dataset. The results derived show that it has a better accuracy than the methods used before.

Original languageEnglish
Title of host publicationProceedings of the IASTED International Conference on Advances in Computer Science and Engineering, ACSE 2009
Pages27-31
Number of pages5
Publication statusPublished - 2009
Externally publishedYes
EventIASTED International Conference on Advances in Computer Science and Engineering, ACSE 2009 - Phuket, Thailand
Duration: 16 Mar 200918 Mar 2009

Publication series

NameProceedings of the IASTED International Conference on Advances in Computer Science and Engineering, ACSE 2009

Conference

ConferenceIASTED International Conference on Advances in Computer Science and Engineering, ACSE 2009
Country/TerritoryThailand
CityPhuket
Period16/03/0918/03/09

Keywords

  • Neural networks
  • Pattern discovery
  • Wavelet analysis

Fingerprint

Dive into the research topics of 'ePattern discovery of topics in network forum based on wavelet analysis and neural network'. Together they form a unique fingerprint.

Cite this