Inferring Narrative Causality between Event Pairs in Films
This work addresses the challenge of understanding narrative relations in film scene descriptions, which is incremental as it builds on prior work by focusing on specific causal types.
The paper tackled the problem of learning narrative causality between event pairs in films, showing that their method produces high-quality event pairs judged to have stronger causal relations than those from Rel-grams.
To understand narrative, humans draw inferences about the underlying relations between narrative events. Cognitive theories of narrative understanding define these inferences as four different types of causality, that include pairs of events A, B where A physically causes B (X drop, X break), to pairs of events where A causes emotional state B (Y saw X, Y felt fear). Previous work on learning narrative relations from text has either focused on "strict" physical causality, or has been vague about what relation is being learned. This paper learns pairs of causal events from a corpus of film scene descriptions which are action rich and tend to be told in chronological order. We show that event pairs induced using our methods are of high quality and are judged to have a stronger causal relation than event pairs from Rel-grams.