Translating Unknown Words by Analogical Learning

Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, pp. 877–886, Prague, June 2007. c⃝2007 Association for Computational Linguistics Translating Unknown Words by Analogical Learning Philippe Langlais and Alexandre Patry Dept. I.R.O. Universit´e de Montr´eal {felipe,patryale}@iro.umontreal.ca Abstract Unknown words are a well-known hindrance to natural language applications. In particu- lar, they drastically impact machine transla- tion quality. An easy way out commercial translation systems usually offer their users is the possibility to add unknown words and their translations into a dedicated lex- icon. Recently, Stroppa and Yvon (2005) have shown how analogical learning alone deals nicely with morphology in different languages. In this study we show that ana- logical learning offers as well an elegant and effective solution to the problem of identify- ing potential translations of unknown words. 1 Introduction Analogical reasoning has received some attention in cognitive science and artificial intelligence (Gentner et al., 2001). It has been for a long time a faculty as- sessed in the so-called SAT Reasoning tests used in the application process to colleges and universities in the United States. Turney (2006) has shown that it is possible to compute relational similarities in a corpus in order to solve 56% of typical analogical tests quizzed in SAT exams. The interested reader can find in (Lepage, 2003) a particularly dense treat- ment of analogy, including a fascinating chapter on the history of the notion of analogy. The concept of proportional analogy, denoted [A : B = C : D], is a relation between four entities which reads: “A is to B as C is to D”. Among proportional analogies, we distinguish for- mal analogies, that is, ones that arise at the graph- ical level, such as [fournit : fleurit = fournie : fleurie] in French or [believer : unbelievable = doer : undoable] in English. Formal analogies are often good indices for deeper analogies (Stroppa and Yvon, 2005). Lepage and Denoual (2005) presented the sys- tem ALEPH, an intriguing example-based system entirely built on top of an automatic formal anal- ogy solver. This system has achieved state-of-the- art performance on the IWSLT task (Eck and Hori, 2005), despite its striking purity. As a matter of fact, ALEPH requires no distances between exam- ples, nor any threshold.1 It does not even rely on a tokenization device. One reason for its success probably lies in the specificity of the BTEC corpus: short and simple sentences of a narrow domain. It is doubtful that ALEPH would still behave adequately on broader tasks, such as translating news articles. Stroppa and Yvon (2005) propose a very help- ful algebraic description of a formal analogy and describe the theoretical foundations of analogical learning which we will recap shortly. They show both its elegance and efficiency on two morphologi- cal analysis tasks for three different languages. Recently, Moreau et al. (2007) showed that for- mal analogies of a simple kind (those involving suf- fixation and/or prefixation) offer an effective way to extend …

Content truncated to 500 words