3 months ago

300-sparsans at SemEval-2018 Task 9: Hypernymy as interaction of sparse attributes

{P{\'e}ter F{\o}ldi{\'a}k M{\'a}rton Makrai G{\'a}bor Berend}

Abstract

This paper describes 300-sparsians{'}s participation in SemEval-2018 Task 9: Hypernym Discovery, with a system based on sparse coding and a formal concept hierarchy obtained from word embeddings. Our system took first place in subtasks (1B) Italian (all and entities), (1C) Spanish entities, and (2B) music entities.

Benchmarks

Benchmark	Methodology	Metrics
hypernym-discovery-on-general	300-sparsans	MAP: 8.95 MRR: 19.44 P@5: 8.63
hypernym-discovery-on-medical-domain	300-sparsans	MAP: 20.75 MRR: 40.60 P@5: 21.43
hypernym-discovery-on-music-domain	300-sparsans	MAP: 29.54 MRR: 46.43 P@5: 28.86

Build AI with AI

From idea to launch — accelerate your AI development with free AI co-coding, out-of-the-box environment and best price of GPUs.

AI Co-coding

Ready-to-use GPUs

Best Pricing

Get Started

Hyper Newsletters

Subscribe to our latest updates

We will deliver the latest updates of the week to your inbox at nine o'clock every Monday morning