Merging Verb Senses of Hindi WordNet using Word Embeddings.

Sudha Bhingardive; Ratish Puduppully; Dhirendra Singh; Pushpak Bhattacharyya

Merging Verb Senses of Hindi WordNet using Word Embeddings.

Sudha Bhingardive, Ratish Puduppully, Dhirendra Singh, Pushpak Bhattacharyya

Industrial Design Centre at the Indian Institute of technology, Mumbai, India.

Publikation: Konference artikel i Proceeding eller bog/rapport kapitel › Konferencebidrag i proceedings › Forskning › peer review

Abstract

In this paper, we present an approach for merging fine-grained verb senses of Hindi WordNet. Senses are merged based on gloss similarity score. We explore the use of word embeddings for gloss similarity computation and compare with various WordNet based gloss similarity measures.

Our results indicate that word embeddings show significant improvement over WordNet based measures. Consequently, we observe an increase in accuracy on merging fine-grained senses. Gold standard data constructed for our experiments is made available.

Originalsprog	Udefineret/Ukendt
Titel	Proceedings of the 11th International Conference on Natural Language Processing
Publikationsdato	2014
Sider	344-352
Status	Udgivet - 2014
Udgivet eksternt	Ja

Adgang til dokumentet

Citationsformater

@inproceedings{a51fbe9de41b45ab91cac17e54208026,

title = "Merging Verb Senses of Hindi WordNet using Word Embeddings.",

abstract = "In this paper, we present an approach for merging fine-grained verb senses of Hindi WordNet. Senses are merged based on gloss similarity score. We explore the use of word embeddings for gloss similarity computation and compare with various WordNet based gloss similarity measures.Our results indicate that word embeddings show significant improvement over WordNet based measures. Consequently, we observe an increase in accuracy on merging fine-grained senses. Gold standard data constructed for our experiments is made available.",

author = "Sudha Bhingardive and Ratish Puduppully and Dhirendra Singh and Pushpak Bhattacharyya",

note = "DBLP's bibliographic metadata records provided through http://dblp.org/search/publ/api are distributed under a Creative Commons CC0 1.0 Universal Public Domain Dedication. Although the bibliographic metadata records are provided consistent with CC0 1.0 Dedication, the content described by the metadata records is not. Content may be subject to copyright, rights of privacy, rights of publicity and other restrictions.",

year = "2014",

language = "Udefineret/Ukendt",

pages = "344--352",

booktitle = "Proceedings of the 11th International Conference on Natural Language Processing",

}

TY - GEN

T1 - Merging Verb Senses of Hindi WordNet using Word Embeddings.

AU - Bhingardive, Sudha

AU - Puduppully, Ratish

AU - Singh, Dhirendra

AU - Bhattacharyya, Pushpak

N1 - DBLP's bibliographic metadata records provided through http://dblp.org/search/publ/api are distributed under a Creative Commons CC0 1.0 Universal Public Domain Dedication. Although the bibliographic metadata records are provided consistent with CC0 1.0 Dedication, the content described by the metadata records is not. Content may be subject to copyright, rights of privacy, rights of publicity and other restrictions.

PY - 2014

Y1 - 2014

N2 - In this paper, we present an approach for merging fine-grained verb senses of Hindi WordNet. Senses are merged based on gloss similarity score. We explore the use of word embeddings for gloss similarity computation and compare with various WordNet based gloss similarity measures.Our results indicate that word embeddings show significant improvement over WordNet based measures. Consequently, we observe an increase in accuracy on merging fine-grained senses. Gold standard data constructed for our experiments is made available.

AB - In this paper, we present an approach for merging fine-grained verb senses of Hindi WordNet. Senses are merged based on gloss similarity score. We explore the use of word embeddings for gloss similarity computation and compare with various WordNet based gloss similarity measures.Our results indicate that word embeddings show significant improvement over WordNet based measures. Consequently, we observe an increase in accuracy on merging fine-grained senses. Gold standard data constructed for our experiments is made available.

M3 - Konferencebidrag i proceedings

SP - 344

EP - 352

BT - Proceedings of the 11th International Conference on Natural Language Processing

ER -