Coreference Resolution on Blogs and Commented News

Publication type
P1
Publication status
Published
Authors
Hendrickx, I., & Hoste, V.
Editor
S. Lalitha Devi, A. Branco, and R. Mitkov
Series
Anaphora Processing and Applications, Lecture Notes in Artificial Intelligence
Volume
5847
Pagination
43-53
Publisher
Springer - Verlag (Heidelberg)
Download
(.pdf)
Project
DuOMAn

Abstract

We focus on automatic coreference resolution for blogs and news articles with user comments as part of a project on opinion mining. We aim to study the effect of the genre shift from edited structured news- paper text to unedited, unstructured blog data. We compare our coref- erence resolution system on three data sets: newspaper articles, mixed newspaper articles and reader comments, and blog data. As can be ex- pected the performance of the automatic coreference resolution system drops drastically when tested on unedited text. We describe the char- acteristics of the different data sets and we examine the typical errors made by the resolution system.