Positive, Negative, or Neutral: Learning an Expanded Opinion Lexicon from Emoticon-Annotated Tweets

Abstract

We present a supervised framework for expanding an opinion lexicon for tweets. The lexicon contains part-of-speech (POS) disambiguated entries with a three-dimensional probability distribution for positive, negative, and neutral polarities. To obtain this distribution using machine learning, we propose word-level attributes based on POS tags and information calculated from streams of emoticon-annotated tweets. Our experimental results show that our method outperforms the three-dimensional word-level polarity classification performance obtained by semantic orientation, a state-of-the-art measure for establishing world-level sentiment.

Cite

Text

Bravo-Marquez et al. "Positive, Negative, or Neutral: Learning an Expanded Opinion Lexicon from Emoticon-Annotated Tweets." International Joint Conference on Artificial Intelligence, 2015.

Markdown

[Bravo-Marquez et al. "Positive, Negative, or Neutral: Learning an Expanded Opinion Lexicon from Emoticon-Annotated Tweets." International Joint Conference on Artificial Intelligence, 2015.](https://mlanthology.org/ijcai/2015/bravomarquez2015ijcai-positive/)

BibTeX

@inproceedings{bravomarquez2015ijcai-positive,
  title     = {{Positive, Negative, or Neutral: Learning an Expanded Opinion Lexicon from Emoticon-Annotated Tweets}},
  author    = {Bravo-Marquez, Felipe and Frank, Eibe and Pfahringer, Bernhard},
  booktitle = {International Joint Conference on Artificial Intelligence},
  year      = {2015},
  pages     = {1229-1235},
  url       = {https://mlanthology.org/ijcai/2015/bravomarquez2015ijcai-positive/}
}