Train-O-Matic: Large-Scale Supervised Word Sense Disambiguation in Multiple Languages without Manual Training Data

Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 78–88 Copenhagen, Denmark, September 7–11, 2017. c⃝2017 Association for Computational Linguistics Train-O-Matic: Large-Scale Supervised Word Sense Disambiguation in Multiple Languages without Manual Training Data Tommaso Pasini and Roberto Navigli Department of Computer Science Sapienza University of Rome {pasini,navigli}@di.uniroma1.it Abstract Annotating large numbers of sentences with senses is the heaviest requirement of current Word Sense Disambiguation. We present Train-O-Matic, a language- independent method for generating mil- lions of sense-annotated training instances for virtually all meanings of words in a language’s vocabulary. The approach is fully automatic: no human interven- tion is required and the only type of hu- man knowledge used is a WordNet-like resource. Train-O-Matic achieves consis- tently state-of-the-art performance across gold standard datasets and languages, while at the same time removing the bur- den of manual annotation. All the training data is available for research purposes at http://trainomatic.org. 1 Introduction Word Sense Disambiguation (WSD) is a key task in computational lexical semantics, inasmuch as it addresses the lexical ambiguity of text by mak- ing explicit the meaning of words occurring in a given context (Navigli, 2009). Anyone who has struggled with frustratingly unintelligible transla- tions from an automatic system, or with the mean- ing bias of search engines, can understand the im- portance for an intelligent system to go beyond the surface appearance of text. There are two mainstream lines of research in WSD: supervised and knowledge-based WSD. Su- pervised WSD frames the problem as a classi- cal machine learning task in which, first a train- ing phase occurs aimed at learning a classification model from sentences annotated with word senses and, second the model is applied to previously- unseen sentences focused on a target word. A key difference from many other problems, however, is that the classes to choose from (i.e., the senses of a target word) vary for each word, therefore requir- ing a separate training process to be performed on a word by word basis. As a result, hundreds of training instances are needed for each ambiguous word in the vocabulary. This would necessitate a million-item training set to be manually created for each language of interest, an endeavour that is currently beyond reach even in resource-rich lan- guages like English. The second paradigm, i.e., knowledge-based WSD, takes a radically different approach: the idea is to exploit a general-purpose knowledge resource like WordNet (Fellbaum, 1998) to de- velop an algorithm which can take advantage of the structural and lexical-semantic information in the resource to choose among the possible senses of a target word occurring in context. For ex- ample, a PageRank-based algorithm can be devel- oped to determine the probability of a given sense being reached starting from the senses of its con- text words. Recent approaches of this kind have been shown to obtain competitive results (Agirre et al., 2014; Moro et al., 2014). However, due to its inherent nature, knowledge-based WSD tends to adopt bag-of-word approaches which do not ex- ploit …

Content truncated to 500 words