Semi-Supervised Learning of Partial Cognates Using Bilingual Bootstrapping

Proceedings of the 21st International Conference on Computational Linguistics and 44th Annual Meeting of the ACL, pages 441–448, Sydney, July 2006. c⃝2006 Association for Computational Linguistics Semi-Supervised Learning of Partial Cognates using Bilingual Bootstrapping Oana Frunza and Diana Inkpen School of Information Technology and Engineering University of Ottawa Ottawa, ON, Canada, K1N 6N5 {ofrunza,diana}@site.uottawa.ca Abstract Partial cognates are pairs of words in two languages that have the same meaning in some, but not all contexts. Detecting the actual meaning of a partial cognate in context can be useful for Machine Trans- lation tools and for Computer-Assisted Language Learning tools. In this paper we propose a supervised and a semi- supervised method to disambiguate par- tial cognates between two languages: French and English. The methods use only automatically-labeled data; therefore they can be applied for other pairs of lan- guages as well. We also show that our methods perform well when using cor- pora from different domains. 1 Introduction When learning a second language, a student can benefit from knowledge in his / her first lan- guage (Gass, 1987), (Ringbom, 1987), (LeBlanc et al. 1989). Cognates – words that have similar spelling and meaning – can accelerate vocabu- lary acquisition and facilitate the reading com- prehension task. On the other hand, a student has to pay attention to the pairs of words that look and sound similar but have different meanings – false friends pairs, and especially to pairs of words that share meaning in some but not all contexts – the partial cognates. Carroll (1992) claims that false friends can be a hindrance in second language learning. She suggests that a cognate pairing process between two words that look alike happens faster in the learner’s mind than a false-friend pairing. Ex- periments with second language learners of dif- ferent stages conducted by Van et al. (1998) suggest that missing false-friend recognition can be corrected when cross-language activation is used – sounds, pictures, additional explanation, feedback. Machine Translation (MT) systems can benefit from extra information when translating a certain word in context. Knowing if a word in the source language is a cognate or a false friend with a word in the target language can improve the translation results. Cross-Language Information Retrieval systems can use the knowledge of the sense of certain words in a query in order to re- trieve desired documents in the target language. Our task, disambiguating partial cognates, is in a way equivalent to coarse grain cross-language Word-Sense Discrimination. Our focus is disam- biguating French partial cognates in context: de- ciding if they are used as cognates with an English word, or if they are used as false friends. There is a lot of work done on monolingual Word Sense Disambiguation (WSD) systems that use supervised and unsupervised methods and report good results on Senseval data, but there is less work done to disambiguate cross-language words. The results of this process can be useful in many NLP tasks. Although French and English belong to differ- ent branches …

Content truncated to 500 words