CRIM at SemEval-2018 Task 9: A Hybrid Approach to Hypernym Discovery
TL
TL;DR
This system exploits a combination of supervised projection learning and unsupervised pattern-based hypernym discovery that was ranked first on the 3 sub-tasks for which it submitted results.
Abstract
This report describes the system developed by the CRIM team for the hypernym discovery task at SemEval 2018. This system exploits a combination of supervised projection learning and unsupervised pattern-based hypernym discovery. It was ranked first on the 3 sub-tasks for which we submitted results.
References22 items
1
SemEval-2018 Task 9: Hypernym Discovery
2
Negative Sampling Improves Hypernymy Extraction Based on Projection Learning
3
Why we have switched from building full-fledged taxonomies to simply detecting hypernymy relations
4
Distributional Hypernym Generation by Jointly Learning Clusters and Projections
5