Unsupervised extraction of appraisal expressions

  • Authors:
  • Kenneth Bloom;Shlomo Argamon

  • Affiliations:
  • Illinois Institute of Technology, Chicago, IL;Illinois Institute of Technology, Chicago, IL

  • Venue:
  • AI'10 Proceedings of the 23rd Canadian conference on Advances in Artificial Intelligence
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

The goal of sentiment analysis is to characterize texts in terms of the opinions and evaluations they express As such, a wide variety of different tasks have been addressed in the field However, there is not yet a clear consensus on how to formalize the notion of “sentiment” or “subjective language” The most commonly studied kind of subjective language in sentimant analysis is evaluative language, that which gives a positive or negative evaluation of some target (Although positioning language, which relates the position of one opinion holder with respect to those of other opinion holders and intentional language and some aspects of modality have also been included.)