Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Unsupervised Induction of Contingent Event Pairs from Film Scenes

Aug 30, 2017
Zhichao Hu, Elahe Rahimtoroghi, Larissa Munishkina, Reid Swanson, Marilyn A. Walker

Share this with someone who'll enjoy it:

Human engagement in narrative is partially driven by reasoning about discourse relations between narrative events, and the expectations about what is likely to happen next that results from such reasoning. Researchers in NLP have tackled modeling such expectations from a range of perspectives, including treating it as the inference of the contingent discourse relation, or as a type of common-sense causal reasoning. Our approach is to model likelihood between events by drawing on several of these lines of previous work. We implement and evaluate different unsupervised methods for learning event pairs that are likely to be contingent on one another. We refine event pairs that we learn from a corpus of film scene descriptions utilizing web search counts, and evaluate our results by collecting human judgments of contingency. Our results indicate that the use of web search counts increases the average accuracy of our best method to 85.64% over a baseline of 50%, as compared to an average accuracy of 75.15% without web search.

* In Proceedings of Conference on Empirical Methods in Natural Language Processing (EMNLP 2013) 

   Access Paper Source

Share this with someone who'll enjoy it: