English

When Hearst Is not Enough: Improving Hypernymy Detection from Corpus with Distributional Models

Computation and Language 2020-10-13 v1

Abstract

We address hypernymy detection, i.e., whether an is-a relationship exists between words (x, y), with the help of large textual corpora. Most conventional approaches to this task have been categorized to be either pattern-based or distributional. Recent studies suggest that pattern-based ones are superior, if large-scale Hearst pairs are extracted and fed, with the sparsity of unseen (x, y) pairs relieved. However, they become invalid in some specific sparsity cases, where x or y is not involved in any pattern. For the first time, this paper quantifies the non-negligible existence of those specific cases. We also demonstrate that distributional methods are ideal to make up for pattern-based ones in such cases. We devise a complementary framework, under which a pattern-based and a distributional model collaborate seamlessly in cases which they each prefer. On several benchmark datasets, our framework achieves competitive improvements and the case study shows its better interpretability.

Cite

@article{arxiv.2010.04941,
  title  = {When Hearst Is not Enough: Improving Hypernymy Detection from Corpus with Distributional Models},
  author = {Changlong Yu and Jialong Han and Peifeng Wang and Yangqiu Song and Hongming Zhang and Wilfred Ng and Shuming Shi},
  journal= {arXiv preprint arXiv:2010.04941},
  year   = {2020}
}

Comments

Accepted by EMNLP2020 Main Conference

R2 v1 2026-06-23T19:13:55.768Z