Methods for unsupervised hypernym detection may broadly be categorized according to two paradigms: pattern-based and distributional methods. In this paper, we study the performance of both approaches on several hypernymy tasks and find that simple pattern-based methods consistently outperform distributional methods on common benchmark datasets. Our results show that pattern-based models provide important contextual constraints which are not yet captured in distributional methods.
@article{arxiv.1806.03191,
title = {Hearst Patterns Revisited: Automatic Hypernym Detection from Large Text Corpora},
author = {Stephen Roller and Douwe Kiela and Maximilian Nickel},
journal= {arXiv preprint arXiv:1806.03191},
year = {2018}
}