Abstract

In this paper, we build a generic opinion-fact classifier to detect opinions and facts from online news articles and social media datasets such as Youtube comments and idiom hashtags. We further use this classification model to compare opinionatedness of various news article sections. The proposed classifier produces better results than the existing methods over four different datasets, and the opinion fraction of various sections of news articles provides very interesting patterns.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call