Unsupervised learning of word sense disambiguation rules by estimating an optimum iteration number in the EM algorithm

Published 1 January 2003Open access

Hiroyuki Shinnou, Minoru Sasaki

DOI10.3115/1119176.1119182

Citations16

TL;DR

An unsupervised learning method using the Expectation-Maximization (EM) algorithm proposed by Nigam et al. for text classification problems is improved and confirmed to be effective also for verb WSD problems.

Abstract

In this paper, we improve an unsupervised learning method using the Expectation-Maximization (EM) algorithm proposed by Nigam et al. for text classification problems in order to apply it to word sense disambiguation (WSD) problems. The improved method stops the EM algorithm at the optimum iteration number. To estimate that number, we propose two methods. In experiments, we solved 50 noun WSD problems in the Japanese Dictionary Task in SENSEVAL2. The score of our method is a match for the best public score of this task. Furthermore, our methods were confirmed to be effective also for verb WSD problems.

Keywords

Computer Science

Combining labeled and unlabeled data with co-training

5,604 Citations1998Avrim Blum, Tom M. Mitchell

View PDF

Machine LearningText Classification from Labeled and Unlabeled Documents using EM

2,749 Citations2000Kamal Nigam, Andrew Kachites McCallum +2 more

This paper shows that the accuracy of learned text classifiers can be improved by augmenting a small number of labeled training documents with a large pool of unlabeled documents, and presents two extensions to the algorithm that improve classification accuracy under these conditions.

View PDF

Unsupervised word sense disambiguation rivaling supervised methods

2,408 Citations1995David Yarowsky

An unsupervised learning algorithm for sense disambiguation that, when trained on unannotated English text, rivals the performance of supervised techniques that require time-consuming hand annotations.

View PDF

Analyzing the effectiveness and applicability of co-training

1,028 Citations2000Kamal Nigam, Rayid Ghani

It is demonstrated that when learning from labeled and unlabeled data, algorithms explicitly leveraging a natural independent split of the features outperform algorithms that do not and may out-perform algorithms not using a split.

View PDF

Word translation disambiguation using Bilingual Bootstrapping

34 Citations2001Cong Li, Hang Li

Experimental results indicate that word translation disambiguation based on bilingual bootstrapping consistently and significantly outperforms existing methods that are based on monolingualbootstrapping.

View PDF

Word sense disambiguation by learning from unlabeled data

11 Citations2000Seong-Bae Park, Byoung‐Tak Zhang +1 more

A learning method is described that exploits unlabeled data to tackle data sparseness problem and uses committee learning to predict the labels of unlabeling data that augment the existing training data.

View PDF

Learning of word sense disambiguation rules by Co-training, checking co-occurrence of features.

2 Citations2002Hiroyuki Shinnou

This paper checks co-occurrence between two feature sets to avoid such undesirable situation when the authors add unlabeled instances to training data and applies this method to word sense disambiguation problems for the three Japanese words ‘koe’, ‘toppu’ and ’kabe’.