Feature selection for improving opinion identification from web authors' posts

Athanasia Koumpouri, Iosif Mporas, Vasileios Megalooikonomou

Research output: Chapter in Book/Report/Conference proceedingConference contribution


In the present article, we address the problem of automatic opinion identification of web users from movie reviews. Specifically, relying on six well-known machine learning algorithms, we investigate the effectiveness of feature selection in the improvement of the accuracy of opinion identification. The feature ranking is performed over a set of statistical, part-ofspeech tagging and language model based features. In the experiments, we employed classification models based on decision trees, support vector machines and lazy-learning algorithms. The experimental evaluation performed on the publicly available Polarity Dataset v2.0 demonstrated that feature selection significantly improves the accuracy of opinion identification regardless of the type of machine learning algorithm used.

Original languageEnglish
Title of host publicationProceedings - 19th Panhellenic Conference on Informatics, PCI 2015
EditorsDemosthenes Akoumianakis, Nikitas N. Karanikolas, Mara Nikolaidou, Michalis Xenos, Dimitris Vergados
PublisherACM Press
Number of pages6
ISBN (Electronic)9781450335515
Publication statusPublished - 1 Oct 2015
Externally publishedYes
Event19th Panhellenic Conference on Informatics, PCI 2015 - Athens, Greece
Duration: 1 Oct 20153 Oct 2015

Publication series

NameACM International Conference Proceeding Series


Conference19th Panhellenic Conference on Informatics, PCI 2015


  • Feature ranking
  • Feature selection
  • Opinion classification
  • Opinion mining
  • Sentiment analysis


Dive into the research topics of 'Feature selection for improving opinion identification from web authors' posts'. Together they form a unique fingerprint.

Cite this