Why we have switched from building full-fledged taxonomies to simply\n detecting hypernymy relations

2017/03/12 by José Camacho-Collados, Camacho-Collados, Jose · 2 citations
Computer Science · #Natural Language Processing Techniques #Topic Modeling #Semantic Web and Ontologies

paper · pdf · doi:10.48550/arxiv.1703.04178

Abstract

The study of taxonomies and hypernymy relations has been extensive on the\nNatural Language Processing (NLP) literature. However, the evaluation of\ntaxonomy learning approaches has been traditionally troublesome, as it mainly\nrelies on ad-hoc experiments which are hardly reproducible and manually\nexpensive. Partly because of this, current research has been lately focusing on\nthe hypernymy detection task. In this paper we reflect on this trend, analyzing\nissues related to current evaluation procedures. Finally, we propose three\npotential avenues for future work so that is-a relations and resources based on\nthem play a more important role in downstream NLP applications.\n

Cited by

• RUSSE'2020: Findings of the First Taxonomy Enrichment Task for the\n Russian language
• Concurrent Learning of Semantic Relations

Related

• Hypernyms under Siege: Linguistically-motivated Artillery for Hypernymy Detection
• Data Augmentation for Hypernymy Detection
• Improving Hypernymy Detection with an Integrated Path-based and Distributional Method
• When Hearst Is not Enough: Improving Hypernymy Detection from Corpus with Distributional Models
• Improving Hypernymy Extraction with Distributional Semantic Classes