no code implementations • WS 2019 • B{\'a}lint D{\"o}br{\"o}ssy, M{\'a}rton Makrai, Bal{\'a}zs Tarj{\'a}n, Gy{\"o}rgy Szasz{\'a}k
For morphologically rich languages, word embeddings provide less consistent semantic representations due to higher variance in word forms.
1 code implementation • WS 2019 • Bal{\'a}zs Indig, B{\'a}lint Sass, Eszter Simon, Iv{\'a}n Mittelholcz, No{\'e}mi Vad{\'a}sz, M{\'a}rton Makrai
We present a more efficient version of the e-magyar NLP pipeline for Hungarian called emtsv.
no code implementations • SEMEVAL 2018 • G{\'a}bor Berend, M{\'a}rton Makrai, P{\'e}ter F{\"o}ldi{\'a}k
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.
Ranked #3 on Hypernym Discovery on Medical domain
no code implementations • LREC 2016 • M{\'a}rton Makrai
The former is exemplified by Wiktionary, a crowd-sourced dictionary with editions in many languages.