Centering: a framework for modeling the local coherence of discourse
Computational Linguistics
Learning semantic links from a corpus of parallel temporal and causal relations
HLT-Short '08 Proceedings of the 46th Annual Meeting of the Association for Computational Linguistics on Human Language Technologies: Short Papers
Predicting unknown time arguments based on cross-event propagation
ACLShort '09 Proceedings of the ACL-IJCNLP 2009 Conference Short Papers
Utility evaluation of cross-document information extraction
HLT '10 Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics
Hi-index | 0.00 |
Traditional Information Extraction (IE) systems identify many unconnected facts. The objective of this paper is to define a new cross-document information extraction task and demonstrate a system which can extract, rank and track events in two dimensions: temporal and spatial. The system can automatically label the person entities involved in significant events as 'centroid arguments', and then present the events involving the same centroid on a time line and on a geographical map.