Temporal Event Knowledge Acquisition via Identifying Narratives
This work addresses the challenge of extracting structured temporal knowledge from unstructured text for applications in natural language processing, though it is incremental as it builds on existing narratology principles.
The paper tackled the problem of acquiring temporal event knowledge from narrative texts by identifying narrative paragraphs and extracting 'before/after' relations, resulting in 287k narrative paragraphs identified and improved performance on temporal relation classification and narrative cloze tasks.
Inspired by the double temporality characteristic of narrative texts, we propose a novel approach for acquiring rich temporal "before/after" event knowledge across sentences in narrative stories. The double temporality states that a narrative story often describes a sequence of events following the chronological order and therefore, the temporal order of events matches with their textual order. We explored narratology principles and built a weakly supervised approach that identifies 287k narrative paragraphs from three large text corpora. We then extracted rich temporal event knowledge from these narrative paragraphs. Such event knowledge is shown useful to improve temporal relation classification and outperform several recent neural network models on the narrative cloze task.