Distantly Supervised Entity Relation Extraction with Adapted Manual Annotations

  • Changzhi Sun East China Normal University
  • Yuanbin Wu East China Normal University

Abstract

We investigate the task of distantly supervised joint entity relation extraction. It’s known that training with distant supervision will suffer from noisy samples. To tackle the problem, we propose to adapt a small manually labelled dataset to the large automatically generated dataset. By developing a novel adaptation algorithm, we are able to transfer the high quality but heterogeneous entity relation annotations in a robust and consistent way. Experiments on the benchmark NYT dataset show that our approach significantly outperforms state-ofthe-art methods.

Published
2019-07-17
Section
AAAI Technical Track: Natural Language Processing