Using support vector machines and state-of-the-art algorithms for phonetic alignment to identify cognates in multi-lingual wordlists

Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers, pages 1205–1216, Valencia, Spain, April 3-7, 2017. c⃝2017 Association for Computational Linguistics Using support vector machines and state-of-the-art algorithms for phonetic alignment to identify cognates in multi-lingual wordlists Gerhard J¨ager T¨ubingen University Institute of Linguistics T¨ubingen, Germany gerhard.jaeger@ uni-tuebingen.de Johann-Mattis List CRLAO / AIRE EHESS / UPMC PARIS info@lingpy.org Pavel Sofroniev T¨ubingen University Institute of Linguistics T¨ubingen, Germany pavel.sofroniev@ student.uni-tuebingen.de Abstract Most current approaches in phylogenetic linguistics require as input multilingual word lists partitioned into sets of etymo- logically related words (cognates). Cog- nate identification is so far done manu- ally by experts, which is time consuming and as of yet only available for a small number of well-studied language families. Automatizing this step will greatly ex- pand the empirical scope of phylogenetic methods in linguistics, as raw wordlists (in phonetic transcription) are much easier to obtain than wordlists in which cognate words have been fully identified and an- notated, even for under-studied languages. A couple of different methods have been proposed in the past, but they are ei- ther disappointing regarding their perfor- mance or not applicable to larger datasets. Here we present a new approach that uses support vector machines to unify diffe- rent state-of-the-art methods for phonetic alignment and cognate detection within a single framework. Training and evaluat- ing these method on a typologically broad collection of gold-standard data shows it to be superior to the existing state of the art. 1 Introduction Computational historical linguistics is a relatively young sub-discipline of computational linguistics which uses computational methods to uncover how the world’s 7 000 human languages have de- veloped into their current shape. The discipline has made great strides in recent years. Excit- ing progress has been made with regard to auto- mated language classification (Bowern and Atkin- son, 2012; J¨ager, 2015), inference regarding the time depth and geographic location of ancestral language stages (Bouckaert et al., 2012), or the identification of sound shifts and the reconstruc- tion of ancestral word forms (Bouchard-Cˆot´e et al., 2013), to mention just a few. Most of the mentioned and related work relies on multilingual word lists manually annotated for cognacy. Unlike the classical NLP conception, cognate words are here understood as words in different languages which are etymologically related, that means, they have regularly developed from a common ances- tral form, such as both English tooth and Ger- man Zahn ‘tooth’ that go back to an earlier Proto- Germanic word tanT- with the same meaning. Manual cognate classification is a slow and la- bor intensive task requiring expertise in histori- cal linguistics and intimate knowledge of the lan- guage family under investigation. From a method- ological perspective, it can further be problematic to build phylogenetic inference on expert judg- ments, as the expert annotators necessarily base their judgments on certain hypotheses regarding the internal structure of the language family in question. In this way, the human-annotated cog- nate sets bear …

Content truncated to 500 words