Mining the web for reciprocal relationships

Michael Paul, Roxana Girju, Chen Li

Research output: Contribution to conferencePaper

Abstract

In this paper we address the problem of identifying reciprocal relationships in English. In particular we introduce an algorithm that semi-automatically discovers patterns encoding reciprocity based on a set of simple but effective pronoun templates. Using a set of most frequently occurring patterns, we extract pairs of reciprocal pattern instances by searching the web. Then we apply two unsupervised clustering procedures to form meaningful clusters of such reciprocal instances. The pattern discovery procedure yields an accuracy of 97%, while the clustering procedures indicate accuracies of 91% and 82%. Moreover, the resulting set of 10,882 reciprocal instances represent a broad-coverage resource.

Original languageEnglish (US)
StatePublished - Jun 2009
Event13th Conference on Computational Natural Language Learning, CoNLL 2009 - Boulder, CO, United States
Duration: Jun 4 2009Jun 5 2009

Other

Other13th Conference on Computational Natural Language Learning, CoNLL 2009
CountryUnited States
CityBoulder, CO
Period6/4/096/5/09

Fingerprint

reciprocity
coverage
resources

ASJC Scopus subject areas

  • Artificial Intelligence
  • Human-Computer Interaction
  • Linguistics and Language

Cite this

Paul, M., Girju, R., & Li, C. (2009). Mining the web for reciprocal relationships. Paper presented at 13th Conference on Computational Natural Language Learning, CoNLL 2009, Boulder, CO, United States.

Mining the web for reciprocal relationships. / Paul, Michael; Girju, Roxana; Li, Chen.

2009. Paper presented at 13th Conference on Computational Natural Language Learning, CoNLL 2009, Boulder, CO, United States.

Research output: Contribution to conferencePaper

Paul, M, Girju, R & Li, C 2009, 'Mining the web for reciprocal relationships', Paper presented at 13th Conference on Computational Natural Language Learning, CoNLL 2009, Boulder, CO, United States, 6/4/09 - 6/5/09.
Paul M, Girju R, Li C. Mining the web for reciprocal relationships. 2009. Paper presented at 13th Conference on Computational Natural Language Learning, CoNLL 2009, Boulder, CO, United States.
Paul, Michael ; Girju, Roxana ; Li, Chen. / Mining the web for reciprocal relationships. Paper presented at 13th Conference on Computational Natural Language Learning, CoNLL 2009, Boulder, CO, United States.
@conference{c8fe499e395a46bc8479de0ceb34a483,
title = "Mining the web for reciprocal relationships",
abstract = "In this paper we address the problem of identifying reciprocal relationships in English. In particular we introduce an algorithm that semi-automatically discovers patterns encoding reciprocity based on a set of simple but effective pronoun templates. Using a set of most frequently occurring patterns, we extract pairs of reciprocal pattern instances by searching the web. Then we apply two unsupervised clustering procedures to form meaningful clusters of such reciprocal instances. The pattern discovery procedure yields an accuracy of 97{\%}, while the clustering procedures indicate accuracies of 91{\%} and 82{\%}. Moreover, the resulting set of 10,882 reciprocal instances represent a broad-coverage resource.",
author = "Michael Paul and Roxana Girju and Chen Li",
year = "2009",
month = "6",
language = "English (US)",
note = "13th Conference on Computational Natural Language Learning, CoNLL 2009 ; Conference date: 04-06-2009 Through 05-06-2009",

}

TY - CONF

T1 - Mining the web for reciprocal relationships

AU - Paul, Michael

AU - Girju, Roxana

AU - Li, Chen

PY - 2009/6

Y1 - 2009/6

N2 - In this paper we address the problem of identifying reciprocal relationships in English. In particular we introduce an algorithm that semi-automatically discovers patterns encoding reciprocity based on a set of simple but effective pronoun templates. Using a set of most frequently occurring patterns, we extract pairs of reciprocal pattern instances by searching the web. Then we apply two unsupervised clustering procedures to form meaningful clusters of such reciprocal instances. The pattern discovery procedure yields an accuracy of 97%, while the clustering procedures indicate accuracies of 91% and 82%. Moreover, the resulting set of 10,882 reciprocal instances represent a broad-coverage resource.

AB - In this paper we address the problem of identifying reciprocal relationships in English. In particular we introduce an algorithm that semi-automatically discovers patterns encoding reciprocity based on a set of simple but effective pronoun templates. Using a set of most frequently occurring patterns, we extract pairs of reciprocal pattern instances by searching the web. Then we apply two unsupervised clustering procedures to form meaningful clusters of such reciprocal instances. The pattern discovery procedure yields an accuracy of 97%, while the clustering procedures indicate accuracies of 91% and 82%. Moreover, the resulting set of 10,882 reciprocal instances represent a broad-coverage resource.

UR - http://www.scopus.com/inward/record.url?scp=79957458664&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79957458664&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:79957458664

ER -