Just “OneSeC” for Producing Multilingual Sense-Annotated Data

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 699–709 Florence, Italy, July 28 - August 2, 2019. c⃝2019 Association for Computational Linguistics 699 Just “OneSeC” for Producing Multilingual Sense-Annotated Data Bianca Scarlini, Tommaso Pasini and Roberto Navigli Department of Computer Science Sapienza University of Rome {scarlini,pasini,navigli}@di.uniroma1.it Abstract The well-known problem of knowledge acqui- sition is one of the biggest issues in Word Sense Disambiguation (WSD), where anno- tated data are still scarce in English and al- most absent in other languages. In this paper we formulate the assumption of One Sense per Wikipedia Category and present OneSeC, a language-independent method for the automatic extraction of hundreds of thou- sands of sentences in which a target word is tagged with its meaning. Our automatically- generated data consistently lead a supervised WSD model to state-of-the-art performance when compared with other automatic and semi-automatic methods. Moreover, our ap- proach outperforms its competitors on mul- tilingual and domain-specific settings, where it beats the existing state of the art on all languages and most domains. All the train- ing data are available for research purposes at http://trainomatic.org/onesec. 1 Introduction The problem of acquiring knowledge (i.e., the knowledge acquisition bottleneck) is an open is- sue in Natural Language Processing (NLP). This problem has become even more critical with the advent of deep learning, as a bigger amount of data is needed to meet the requirements of more and more difficult tasks and increasingly complex models. Word Sense Disambiguation (WSD), i.e., the task of associating a word with its meaning in a context (Navigli, 2009), is one of the most affected research areas (Navigli, 2018). The in- terest in this field has grown remarkably due to the variety of applications that can benefit from it, such as Machine Translation (Neale et al., 2016) or Information Extraction (Delli Bovi et al., 2015). Most approaches to WSD are either supervised or knowledge-based. The former frames the problem as a classification (Zhong and Ng, 2010) or se- quence learning (Raganato et al., 2017b) task, in which either a target word or all the content words in a sequence have to be tagged with one of their possible meanings. The latter, instead, exploits graph algorithms on knowledge bases, such as the Personalized PageRank method (Haveliwala, 2002; Agirre et al., 2014), or the densest subgraph heuristic (Moro et al., 2014). Hence, knowledge- based approaches rely on semantic networks such as WordNet1 (Miller et al., 1990), a manually- curated resource where synonyms are grouped into so-called synsets, or BabelNet2 (Navigli and Ponzetto, 2010), a large multilingual encyclopedic dictionary that merges together different resources like WordNet, Wikipedia, Wikidata etc. There- fore, in one form or another both approaches to WSD need lexical-semantic data. This is espe- cially crucial in the case of supervised systems, which have proved capable of attaining higher re- sults on English, for which annotated data are available, whereas they fall behind knowledge- based approaches when tested on other languages. Unfortunately, carrying out semantic annotations for …

Content truncated to 500 words