UTD's Event Nugget Detection and Coreference System at KBP 2016

Jing Lu and Vincent Ng.
Proceedings of the 2016 Text Analysis Conference, 2016.

Click here for the PDF version.

Abstract

We describe UTD's participating system in the event nugget detection and coreference task at TAC-KBP 2016. We designed and implemented a pipeline system that consists of three components: event nugget identification and subtyping, REALIS value identification, and event coreference resolution. We proposed using an ensemble of 1-nearest-neighbor clasifiers for event nugget identification and subtyping, an SVM classifier for REALIS value identification, and a learning-based multi-pass sieve approach consisting of 1-nearest-neighbor classifiers for event coreference resolution. Though conceptually simple, our system compares favorably with other participating systems, achieving F1 scores of 46.99, 39.78, and 30.08 on these three tasks respectively on the English datset, and F1 socres of 40.01, 33.68, and 26.43 on the Chinese dataset. In particular, it ranked first on English event nugget detection as well as on English and Chinese event nugget coreference.

BibTeX entry

@InProceedings{Lu+Ng:16b,
  author = {Jing Lu and Vincent Ng},
  title = {UTD's Event Nugget Detection and Coreference System at KBP 2016},
  booktitle = {Proceedings of 2016 Text Analysis Conference},
  year = 2016}