Inferring Sentence-internal Temporal Relations

Mirella Lapata, Alex Lascarides

Abstract: In this paper we propose a data intensive approach for inferring sentence-internal temporal relations, which relies on a simple probabilistic model and assumes no manual coding. We explore various combinations of features, and evaluate performance against a goldstandard corpus and human subjects performing the same task. The best model achieves 70.7% accuracy in inferring the temporal relation between two clauses and 97.4% accuracy in ordering them, assuming that the temporal relation is known.