Automatically Building a Multilingual Lexicon of False Friends With No Supervision

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 3001–3007 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 3001 Automatically Building a Multilingual Lexicon of False Friends With No Supervision Ana-Sabina Uban, Liviu P. Dinu Human Language Technologies Research Center, Faculty of Mathematics and Computer Science, University of Bucharest ana.uban@gmail.com, ldinu@fmi.unibuc.ro Abstract Cognate words, defined as words in different languages which derive from a common etymon, can be useful for language learners, who can leverage the orthographical similarity of cognates to more easily understand a text in a foreign language. Deceptive cognates, or false friends, do not share the same meaning anymore; these can be instead deceiving and detrimental for language acquisition or text understanding in a foreign language. We use an automatic method of detecting false friends from a set of cognates, in a fully unsupervised fashion, based on cross-lingual word embeddings. We implement our method for English and five Romance languages, including a low-resource language (Romanian), and evaluate it against two different gold standards. The method can be extended easily to any language pair, requiring only large monolingual corpora for the involved languages and a small bilingual dictionary for the pair. We additionally propose a measure of ”falseness” of a false friends pair. We publish freely the database of false friends in the six languages, along with the falseness scores for each cognate pair. The resource is the largest of the kind that we are aware of, both in terms of languages covered and number of word pairs. Keywords: cognates, false friends, database, language acquisition, word embeddings, low-resource languages, unsupervised 1. Introduction Cognates are words in genetically related languages (lan- guages descending from a common ancestor) with a com- mon proto-word. For example, the Romanian word victo- rie and the Italian word vittoria are cognates, as they both descend from the Latin word victoria (meaning victory) – see Figure 1. Cognates can be useful for language learners, who can have a prior understanding of what a word in a foreign language means only through knowing its cognate: a Romanian speaker can easily learn the Italian word vitto- ria, and, for related languages where much of the vocabu- lary consists of cognates, non-native speakers of the related language can even achieve basic understanding of texts in the foreign language without ever having studied it. In most cases, cognates have preserved similar meanings across languages, but there are also exceptions. In some cases, the meanings of cognates has diverged from the common etymon through their use in each of the two lan- guages, and their meanings became different from each other. These are called deceptive cognates or, more com- monly, false friends. Here we use the definition of cognates that refers to words with similar appearance and some com- mon etymology, while true cognates is used to refer to cog- nates which also have a common meaning, and deceptive cognates or false friends refers to cognate pairs which do not have …

Content truncated to 500 words