Worst-case complexity and empirical evaluation of artificial intelligence methods for unsupervised word sense disambiguation

Schwab, Didier; Goulian, Jérôme; Tchechmedjiev, Andon

doi:10.1504/ijwet.2013.055713

Cited by 10 publications

(45 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The most accurate techniques are supervised (Iacobacci et al, 2016), but they require annotated training data which is not always available. In order to overcome this limitation, some researchers have proposed various unsupervised or knowledgde-based WSD methods (Banerjee and Pedersen, 2002;Schwab et al, 2012;Nguyen and Ock, 2013;Schwab et al, 2013a;Schwab et al, 2013b;Chen et al, 2014;Agirre et al, 2014;Bhingardive et al, 2015). Since our approach is unsupervised and based on WordNet (Miller, 1995;Fellbaum, 1998), our focus is to present related work in the same direction.…”

Section: Related Workmentioning

confidence: 99%

“…As we select a single context window at every possible location in a text, our algorithm becomes deterministic, obtaining the same global configuration for a given set of parameters and input document. Thus, our algorithm is not affected by random chance, unlike stochastic algorithms such as Ant Colony Optimization (Lafourcade and Guinand, 2010;Schwab et al, 2012;Schwab et al, 2013a).…”

Section: Introductionmentioning

confidence: 99%

“…We have conducted experiments on SemEval 2007 (Navigli et al, 2007), Senseval-2 (Edmonds and Cotton, 2001) and Senseval-3 (Mihalcea et al, 2004) data sets in order to compare ShotgunWSD with three state-of-the-art approaches (Schwab et al, 2013a;Chen et al, 2014;Bhingardive et al, 2015) along with the Most Common Sense (MCS) baseline 2 , which is considered as the strongest baseline in WSD (Agirre and Edmonds, 2006). The empirical results show that our algorithm compares favorably to these approaches.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

ShotgunWSD: An unsupervised algorithm for global word sense disambiguation inspired by DNA sequencing

Butnaru

Ionescu

Hristea

2017

Proceedings of the 15th Conference of the European Chapter of The Association for Computational Linguistics: Volume 1

View full text Add to dashboard Cite

In this paper, we present a novel unsupervised algorithm for word sense disambiguation (WSD) at the document level. Our algorithm is inspired by a widely-used approach in the field of genetics for whole genome sequencing, known as the Shotgun sequencing technique. The proposed WSD algorithm is based on three main steps. First, a brute-force WSD algorithm is applied to short context windows (up to 10 words) selected from the document in order to generate a short list of likely sense configurations for each window. In the second step, these local sense configurations are assembled into longer composite configurations based on suffix and prefix matching. The resulted configurations are ranked by their length, and the sense of each word is chosen based on a voting scheme that considers only the top k configurations in which the word appears. We compare our algorithm with other state-of-the-art unsupervised WSD algorithms and demonstrate better performance, sometimes by a very large margin. We also show that our algorithm can yield better performance than the Most Common Sense (MCS) baseline on one data set. Moreover, our algorithm has a very small number of parameters, is robust to parameter tuning, and, unlike other bioinspired methods, it gives a deterministic solution (it does not involve random choices).

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

ShotgunWSD: An unsupervised algorithm for global word sense disambiguation inspired by DNA sequencing

Butnaru

Ionescu

Hristea

2017

Proceedings of the 15th Conference of the European Chapter of The Association for Computational Linguistics: Volume 1

View full text Add to dashboard Cite

show abstract

“…Schwab studies the influence of 3 unsupervised algorithms on word sense disambiguation, including genetic algorithm, simulated annealing algorithm and ant colony algorithm. Experiment results show that the performance of ant colony algorithm is better and a shorter time is needed [10]. Imamura combines active learning method with semisupervised learning method to train WSD classifier, in which pseudo negative examples and unlabeled examples are used [11].…”

Section: Introductionmentioning

confidence: 99%

Determine Word Sense Based on Semantic and Syntax Information

Zhang¹,

Sun²,

Gao³

2016

IJDTA

View full text Add to dashboard Cite

show abstract

“…Schwab applies three unsupervised stochastic algorithms to word sense disambiguation including genetic algorithms, simulated annealing algorithms and ant colony algorithms. At the same time, comparative experiments are conducted to evaluate these 3 algorithms' performances [11]. Huang gives a position-based algorithm in order to measure the context similarity, in which contextual words are assigned with positional weights.…”

Section: Introductionmentioning

confidence: 99%