Aspectuality Across Genre: A Distributional Semantics Approach
This work addresses the challenge of verb aspect interpretation for natural language processing tasks like textual entailment, but it is incremental as it builds on existing distributional semantics methods.
The paper tackled the problem of modeling lexical aspect of verbs in English using distributional semantics, showing that local context and closed class words are effective indicators, and outperformed previous work on three datasets.
The interpretation of the lexical aspect of verbs in English plays a crucial role for recognizing textual entailment and learning discourse-level inferences. We show that two elementary dimensions of aspectual class, states vs. events, and telic vs. atelic events, can be modelled effectively with distributional semantics. We find that a verb's local context is most indicative of its aspectual class, and demonstrate that closed class words tend to be stronger discriminating contexts than content words. Our approach outperforms previous work on three datasets. Lastly, we contribute a dataset of human--human conversations annotated with lexical aspect and present experiments that show the correlation of telicity with genre and discourse goals.