Computational Linguistics Colloquium

Thursday, 31 May, 16:15
Conference Room, Building C7 4

Modeling Local Coherence: An Entity-based Approach

This talk proposes a novel framework for representing and measuring local coherence. Central to this approach is the entity grid representation of discourse which captures patterns of entity distribution in texts. I will introduce an algorithm that automatically abstracts a text into a set of entity transition sequences and records, distributional, syntactic and referential information about discourse entities. The framework allows to re-conceptualize coherence assessment as a learning task and is well-suited for ranking-based generation and text classification tasks. Using the proposed representation, I will present results on a variety of tasks, ranging from text ordering to summary coherence evaluation and readability assessment.

(joint work with Regina Barzilay)

If you would like to meet with the speaker, please contact Sebastian Pado.