Mining sentiments in SMS texts for teaching evaluation

Chee Kian Leong, Yew Haur Lee, Wai Keong Mak

Research output: Journal PublicationArticlepeer-review

97 Citations (Scopus)

Abstract

This paper explores the potential application of sentiment mining for analyzing short message service (SMS) texts in teaching evaluation. Data preparation involves the reading, parsing and categorization of the SMS texts. Three models were developed: the base model, the "corrected" model which adjusts for spelling errors and the "sentiment" model which extends the "corrected" model by performing sentiment mining. An "interestingness" criterion selects the "sentiment" model from which the sentiments of the students towards the lecture are discerned. Two types of incomplete SMS texts are also identified and the implications of their removal for the analysis ascertained.

Original languageEnglish
Pages (from-to)2584-2589
Number of pages6
JournalExpert Systems with Applications
Volume39
Issue number3
DOIs
Publication statusPublished - 15 Feb 2012

Keywords

  • Education
  • Sentiment mining
  • SMS texts

ASJC Scopus subject areas

  • General Engineering
  • Computer Science Applications
  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'Mining sentiments in SMS texts for teaching evaluation'. Together they form a unique fingerprint.

Cite this