Abduction for Discourse Interpretation: A Probabilistic Framework
Ekaterina Ovchinnikova, Andrew S. Gordon, Jerry R. Hobbs · 2013
Abduction allows us to model interpre-tation of discourse as the explanation of observables, given additional knowledge about the world. In an abductive frame-work, many explanations can be con-structed for the same observation, requir-ing an approach to estimate the likelihood of these alternative explanations. We show that, for discourse interpretation, weighted abduction has advantages over alternative approaches to estimating the likelihood of hypotheses. However, weighted ab-duction has no probabilistic interpretation, which makes the estimation and learning of weights difficult. To address this, we propose a formal probabilistic abductive framework that captures the advantages weighted abduction when applied to dis-course interpretation. 1