Building domain-oriented sentiment lexicon by improved information bottleneck

Abstract

This paper describes an adapted information bottleneck approach for construction of domain-oriented sentiment lexicon. The basic idea is to use three kinds of relationships (WWinter, WDinter and WDintra,) to infer the semantic orientation of the out-of-domain words. The experimental results demonstrate that proposed method could dramatically improve the accuracy of the baseline approach on the construction of out-of-domain sentiment lexicon.