Skip to main navigation Skip to search Skip to main content

Emotion monitoring and abnormal warning based on online comments

  • Beihang University

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

Abstract

This article aims to find out and monitor emotion tendency contained in online comments to support decision making. We have constructed an emotion monitoring and abnormal warning model based on sentiment classification. We compared multiple sentiment classification methods for Chinese online comments on training set. Experimental results showed that the performance of information gain feature selection and support vector machine is the best. In the end, we gave an application research on two competitive products of the emotion monitoring and abnormal warning model to identify potential opportunity and risk.

Original languageEnglish
Title of host publicationInformation Science and Management Engineering
Pages1045-1052
Number of pages8
DOIs
StatePublished - 2014
EventWIT Transactions on Information and Communication Technologies - Wuhan, China
Duration: 7 May 20138 May 2013

Publication series

NameWIT Transactions on Information and Communication Technologies
Volume46 VOLUME 2
ISSN (Print)1743-3517

Conference

ConferenceWIT Transactions on Information and Communication Technologies
Country/TerritoryChina
CityWuhan
Period7/05/138/05/13

Keywords

  • Abnormal warning
  • Online comment
  • Sentiment classification
  • Sentimental indicator

Fingerprint

Dive into the research topics of 'Emotion monitoring and abnormal warning based on online comments'. Together they form a unique fingerprint.

Cite this