Identifying disputed topics in the news

Publication type
C1
Publication status
Published
Authors
De Clercq, O., Hertling, S., Hoste, V., Ponzetto, S.P., & Paulheim, H.
Series
Proceedings of the LD4KD Workshop at ECML/PKDD2014
Pagination
37-48
Publisher
CEUR
Conference
Linked Data for Knowledge Discovery (LD4KD) (Nancy, France)
Download
(.pdf)
Project
PARIS
View in Biblio
(externe link)

Abstract

News articles often reflect an opinion or point of view, with
certain topics evoking more diverse opinions than others. For analyzing and better understanding public discourses, identifying such contested topics constitutes an interesting research question. In this paper, we describe an approach that combines NLP techniques and background knowledge from DBpedia for fi nding disputed topics in news sites. To identify these topics, we annotate each article with DBpedia concepts, extract their categories, and compute a sentiment score in order to identify those categories revealing signifi cant deviations in polarity across diff erent media. We illustrate our approach in a qualitative evaluation on a sample of six popular British and American news sites.