Pairwise FastText Classifier for Entity Disambiguation

Cheng Yu; Bing Chu; Rohit Ram; James Aichinger; Lizhen Qu; Hanna Suominen

Pairwise FastText Classifier for Entity Disambiguation

Cheng Yu, Bing Chu, Rohit Ram, James Aichinger, Lizhen Qu, Hanna Suominen

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

For the Australasian Language Technology Association (ALTA) 2016 Shared Task, we devised Pairwise FastText Classifier (PFC), an efficient embedding-based text classifier, and used it for entity disambiguation. Compared with a few baseline algorithms, PFC achieved a higher F1 score at 0.72 (under the team name BCJR). To generalise the model, we also created a method to bootstrap the training set deterministically without human labelling and at no financial cost. By releasing PFC and the dataset augmentation software to the public1, we hope to invite more collaboration.

Original language	English
Title of host publication	Proceedings of Australasian Language Technology Association Workshop 2016 Workshop
Editors	Trevor Cohn
Place of Publication	Pennsylvania, USA
Publisher	Association for Computational Linguistics
Pages	175âˆ’179pp
Edition	Peer Reviewed
ISBN (Print)	9781510833166
Publication status	Published - 2016
Event	Australasian Language Technology Association Workshop (ALTA 2016) - Caulfield, Australia Duration: 1 Jan 2016 → … http://alta2016.alta.asn.au/U16/U16-1.pdf

Conference

Conference	Australasian Language Technology Association Workshop (ALTA 2016)
Period	1/01/16 → …
Other	December 5â€“7 2016
Internet address	http://alta2016.alta.asn.au/U16/U16-1.pdf

Cite this

@inproceedings{32fb4ced3791434a9557dc87a9369c39,

title = "Pairwise FastText Classifier for Entity Disambiguation",

abstract = "For the Australasian Language Technology Association (ALTA) 2016 Shared Task, we devised Pairwise FastText Classifier (PFC), an efficient embedding-based text classifier, and used it for entity disambiguation. Compared with a few baseline algorithms, PFC achieved a higher F1 score at 0.72 (under the team name BCJR). To generalise the model, we also created a method to bootstrap the training set deterministically without human labelling and at no financial cost. By releasing PFC and the dataset augmentation software to the public1, we hope to invite more collaboration.",

author = "Cheng Yu and Bing Chu and Rohit Ram and James Aichinger and Lizhen Qu and Hanna Suominen",

year = "2016",

language = "English",

isbn = "9781510833166",

pages = "175{\^a}ˆ{\textquoteright}179pp",

editor = "{Trevor Cohn}",

booktitle = "Proceedings of Australasian Language Technology Association Workshop 2016 Workshop",

publisher = "Association for Computational Linguistics",

edition = "Peer Reviewed",

note = "Australasian Language Technology Association Workshop (ALTA 2016) ; Conference date: 01-01-2016",

url = "http://alta2016.alta.asn.au/U16/U16-1.pdf",

}

Yu, C, Chu, B, Ram, R, Aichinger, J, Qu, L & Suominen, H 2016, Pairwise FastText Classifier for Entity Disambiguation. in TC (ed.), Proceedings of Australasian Language Technology Association Workshop 2016 Workshop. Peer Reviewed edn, Association for Computational Linguistics, Pennsylvania, USA, pp. 175âˆ’179pp, Australasian Language Technology Association Workshop (ALTA 2016), 1/01/16.

Pairwise FastText Classifier for Entity Disambiguation. / Yu, Cheng; Chu, Bing; Ram, Rohit et al.
Proceedings of Australasian Language Technology Association Workshop 2016 Workshop. ed. / Trevor Cohn. Peer Reviewed. ed. Pennsylvania, USA: Association for Computational Linguistics, 2016. p. 175âˆ’179pp.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Pairwise FastText Classifier for Entity Disambiguation

AU - Yu, Cheng

AU - Chu, Bing

AU - Ram, Rohit

AU - Aichinger, James

AU - Qu, Lizhen

AU - Suominen, Hanna

PY - 2016

Y1 - 2016

N2 - For the Australasian Language Technology Association (ALTA) 2016 Shared Task, we devised Pairwise FastText Classifier (PFC), an efficient embedding-based text classifier, and used it for entity disambiguation. Compared with a few baseline algorithms, PFC achieved a higher F1 score at 0.72 (under the team name BCJR). To generalise the model, we also created a method to bootstrap the training set deterministically without human labelling and at no financial cost. By releasing PFC and the dataset augmentation software to the public1, we hope to invite more collaboration.

AB - For the Australasian Language Technology Association (ALTA) 2016 Shared Task, we devised Pairwise FastText Classifier (PFC), an efficient embedding-based text classifier, and used it for entity disambiguation. Compared with a few baseline algorithms, PFC achieved a higher F1 score at 0.72 (under the team name BCJR). To generalise the model, we also created a method to bootstrap the training set deterministically without human labelling and at no financial cost. By releasing PFC and the dataset augmentation software to the public1, we hope to invite more collaboration.

M3 - Conference contribution

SN - 9781510833166

SP - 175âˆ’179pp

BT - Proceedings of Australasian Language Technology Association Workshop 2016 Workshop

A2 - null, Trevor Cohn

PB - Association for Computational Linguistics

CY - Pennsylvania, USA

T2 - Australasian Language Technology Association Workshop (ALTA 2016)

Y2 - 1 January 2016

ER -

Pairwise FastText Classifier for Entity Disambiguation

Abstract

Conference

Fingerprint

Cite this