Geometry of compositionality

Research output: Contribution to conferencePaper

Abstract

This paper proposes a simple test for compositionality (i.e., literal usage) of a word or phrase in a context-specific way. The test is computationally simple, relying on no external resources and only uses a set of trained word vectors. Experiments show that the proposed method is competitive with state of the art and displays high accuracy in context-specific compositionality detection of a variety of natural language phenomena (idiomaticity, sarcasm, metaphor) for different datasets in multiple languages. The key insight is to connect compositionality to a curious geometric property of word embeddings, which is of independent interest.

Original languageEnglish (US)
Pages3202-3208
Number of pages7
StatePublished - Jan 1 2017
Event31st AAAI Conference on Artificial Intelligence, AAAI 2017 - San Francisco, United States
Duration: Feb 4 2017Feb 10 2017

Other

Other31st AAAI Conference on Artificial Intelligence, AAAI 2017
CountryUnited States
CitySan Francisco
Period2/4/172/10/17

Fingerprint

Geometry
Experiments

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Gong, H., Bhat, S. P., & Viswanath, P. (2017). Geometry of compositionality. 3202-3208. Paper presented at 31st AAAI Conference on Artificial Intelligence, AAAI 2017, San Francisco, United States.

Geometry of compositionality. / Gong, Hongyu; Bhat, Suma Pallathadka; Viswanath, Pramod.

2017. 3202-3208 Paper presented at 31st AAAI Conference on Artificial Intelligence, AAAI 2017, San Francisco, United States.

Research output: Contribution to conferencePaper

Gong, H, Bhat, SP & Viswanath, P 2017, 'Geometry of compositionality', Paper presented at 31st AAAI Conference on Artificial Intelligence, AAAI 2017, San Francisco, United States, 2/4/17 - 2/10/17 pp. 3202-3208.
Gong H, Bhat SP, Viswanath P. Geometry of compositionality. 2017. Paper presented at 31st AAAI Conference on Artificial Intelligence, AAAI 2017, San Francisco, United States.
Gong, Hongyu ; Bhat, Suma Pallathadka ; Viswanath, Pramod. / Geometry of compositionality. Paper presented at 31st AAAI Conference on Artificial Intelligence, AAAI 2017, San Francisco, United States.7 p.
@conference{582e5e791e81479abf6e2f56d02e30fc,
title = "Geometry of compositionality",
abstract = "This paper proposes a simple test for compositionality (i.e., literal usage) of a word or phrase in a context-specific way. The test is computationally simple, relying on no external resources and only uses a set of trained word vectors. Experiments show that the proposed method is competitive with state of the art and displays high accuracy in context-specific compositionality detection of a variety of natural language phenomena (idiomaticity, sarcasm, metaphor) for different datasets in multiple languages. The key insight is to connect compositionality to a curious geometric property of word embeddings, which is of independent interest.",
author = "Hongyu Gong and Bhat, {Suma Pallathadka} and Pramod Viswanath",
year = "2017",
month = "1",
day = "1",
language = "English (US)",
pages = "3202--3208",
note = "31st AAAI Conference on Artificial Intelligence, AAAI 2017 ; Conference date: 04-02-2017 Through 10-02-2017",

}

TY - CONF

T1 - Geometry of compositionality

AU - Gong, Hongyu

AU - Bhat, Suma Pallathadka

AU - Viswanath, Pramod

PY - 2017/1/1

Y1 - 2017/1/1

N2 - This paper proposes a simple test for compositionality (i.e., literal usage) of a word or phrase in a context-specific way. The test is computationally simple, relying on no external resources and only uses a set of trained word vectors. Experiments show that the proposed method is competitive with state of the art and displays high accuracy in context-specific compositionality detection of a variety of natural language phenomena (idiomaticity, sarcasm, metaphor) for different datasets in multiple languages. The key insight is to connect compositionality to a curious geometric property of word embeddings, which is of independent interest.

AB - This paper proposes a simple test for compositionality (i.e., literal usage) of a word or phrase in a context-specific way. The test is computationally simple, relying on no external resources and only uses a set of trained word vectors. Experiments show that the proposed method is competitive with state of the art and displays high accuracy in context-specific compositionality detection of a variety of natural language phenomena (idiomaticity, sarcasm, metaphor) for different datasets in multiple languages. The key insight is to connect compositionality to a curious geometric property of word embeddings, which is of independent interest.

UR - http://www.scopus.com/inward/record.url?scp=85030463530&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85030463530&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:85030463530

SP - 3202

EP - 3208

ER -