Can Cognate Prediction Be Modelled as a Low-Resource Machine Translation Task?

Can Cognate Prediction Be Modelled as a Low-Resource Machine Translation Task? Clémentine Fourrier Rachel Bawden Inria, France firstname.lastname@inria.fr Benoît Sagot Abstract Cognate prediction is the task of generating, in a given language, the likely cognates of words in a related language, where cognates are words in related languages that have evolved from a common ancestor word. It is a task for which little data exists and which can aid linguists in the discovery of previously undis- covered relations. Previous work has applied machine translation (MT) techniques to this task, based on the tasks’ similarities, with- out, however, studying their numerous differ- ences or optimising architectural choices and hyper-parameters. In this paper, we inves- tigate whether cognate prediction can bene- fit from insights from low-resource MT. We first compare statistical MT (SMT) and neural MT (NMT) architectures in a bilingual setup. We then study the impact of employing data augmentation techniques commonly seen to give gains in low-resource MT: monolingual pretraining, backtranslation and multilingual- ity. Our experiments on several Romance lan- guages show that cognate prediction behaves only to a certain extent like a standard low- resource MT task. In particular, MT architec- tures, both statistical and neural, can be suc- cessfully used for the task, but using supple- mentary monolingual data is not always as ben- eficial as using additional language data, con- trarily to what is observed for MT. 1 Introduction The Neogrammarians (Osthoff and Brugmann, 1878) formalised one of the main hypotheses of the then recent field of comparative linguistics, the regularity of sound changes: if a phone in a word, at a given moment in the history of a given language, evolves into another phone, then all occurrences of the same phone in the same phonetic context in the same language evolve in the same way. Sound changes are usually identified by look- ing at the attested (or hypothesised) phonetic form of specific sets of words, called cognates, whose definition varies in the literature depending on the field.1 We use an extension of the customary defini- tion used in historical linguistics, as described for instance in (Hauer and Kondrak, 2011; List et al., 2017), which is the following: given two languages with a common ancestor, two words are said to be cognates if they are an evolution of the same word from said ancestor, having undergone the sound changes characteristic of their respective languages’ evolution. We extend it by also allowing the an- cestor word (from the parent language) to also be considered a cognate. For example, Latin bonus ‘good’ gave Italian buono ‘id.’, Spanish bueno ‘id.’ and Spanish bono ‘id.’ by inheritance, and they are all cognates, whereas Spanish abonar ‘to fertilise’, obtained by derivation, is related but not a cognate (Figure 1). Figure 1: Related words in Italian and Spanish, both the outcome of Latin bonus ‘good’. Plain arrows rep- resent inheritance, dotted arrows derivation. “?” indi- cates that the word is not present in our database. Cognate identification (finding cognate pairs in a multilingual …

Content truncated to 500 words