An Etymological Approach to Cross-Language Orthographic Similarity. Application on Romanian

Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 1047–1058, October 25-29, 2014, Doha, Qatar. c⃝2014 Association for Computational Linguistics An Etymological Approach to Cross-Language Orthographic Similarity. Application on Romanian Alina Maria Ciobanu, Liviu P. Dinu Faculty of Mathematics and Computer Science, University of Bucharest Center for Computational Linguistics, University of Bucharest alina.ciobanu@my.fmi.unibuc.ro,ldinu@fmi.unibuc.ro Abstract In this paper we propose a computational method for determining the orthographic similarity between Romanian and related languages. We account for etymons and cognates and we investigate not only the number of related words, but also their forms, quantifying orthographic similari- ties. The method we propose is adaptable to any language, as far as resources are available. 1 Introduction Language relatedness and language change across space and time are two of the main questions of the historical and comparative linguistics (Rama and Borin, 2014). Many comparative methods have been used to establish relationships between lan- guages, to determine language families and to re- construct their proto-languages (Durie and Ross, 1996). If grouping of languages in linguistic fam- ilies is generally accepted, the relationships be- tween languages belonging to the same family are periodically investigated. In spite of the fact that linguistic literature abounds in claims of classifi- cation of natural languages, the degrees of similar- ity between languages are far from being certain. In many situations, the similarity of natural lan- guages is a fairly vague notion, both linguists and non-linguists having intuitions about which lan- guages are more similar to which others. McMa- hon and McMahon (2003) and Rama and Borin (2014) note that the computational historical lin- guistics did not receive much attention until the beginning of the 1990s, and argue for the necessity of development of quantitative and computational methods in this field. 1.1 Related Work According to Campbell (2003), the methods based on comparisons of cognate lists and sound corre- spondences are the most popular approaches em- ployed for establishing relationships between lan- guages. Barbanc¸on et al. (2013) emphasize the va- riety of computational methods used in this field, and state that the differences in datasets and ap- proaches cause difficulties in the evaluation of the results regarding the reconstruction of the phylo- genetic tree of languages. Linguistic phylogeny reconstruction proves especially useful in histor- ical and comparative linguistics, as it enables the analysis of language evolution. Ringe et al. (2002) propose a computational method for evolutionary tree reconstruction based on a “perfect phylogeny” algorithm; using a Bayesian phylogeographic ap- proach, Alekseyenko et al. (2012), continuing the work of Atkinson et al. (2005), model the expan- sion of the Indo-European language family and find support for the hypothesis which places its homeland in Anatolia; Atkinson and Gray (2006) analyze language divergence dates and argue for the usage of computational phylogenetic meth- ods in the question of Indo-European age and ori- gins. Using modified versions of Swadesh’s lists1, Dyen et al. (1992) investigate the classification of Indo-European languages by applying a lexicosta- tistical method. The similarity of languages …

Content truncated to 500 words