High-level Features for Learning Subjective Language across Domains.

摘要: In this paper, we propose to study the characteristics for analyzing subjective content in documents. For that purpose, present and evaluate a novel method based on level of abstraction nouns. By comparing state-of-the-art features nouns between three annotated corpora texts downloaded from Wikipedia Web Blogs, show that, building data sets classification opinionated can be done automatically web, at document level. Moreover, accuracy levels within domains 96.5% across 74.5%.