Dataset for Temporal Analysis of English-French Cognates

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 855–859 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 855 Dataset for Temporal Analysis of English-French Cognates Esteban Frossard♠ Micka¨el Coustaty♠ Antoine Doucet♠ Adam Jatowt♦ Simon Hengchen♣ ♠University of La Rochelle, L3i Laboratory, {firstname.lastname}@univ-lr.fr ♦Kyoto University, adam@dl.kuis.kyoto-u.ac.jp ♣University of Helsinki, simon.hengchen@helsinki.fi Abstract Languages change over time and, thanks to the abundance of digital corpora, their evolutionary analysis using computational techniques has recently gained much research attention. In this paper, we focus on creating a dataset to support investigating the similarity in evolution between different languages. We look in particular into the similarities and differences between the use of corresponding words across time in English and French, two languages from different linguistic families yet with shared syntax and close contact. For this we select a set of cognates in both languages and study their frequency changes and correlations over time. We propose a new dataset for computational approaches of synchronized diachronic investigation of language pairs, and subsequently show novel findings stemming from the cognate-focused diachronic comparison of the two chosen languages. To the best of our knowledge, the present study is the first in the literature to use computational approaches and large data to make a cross-language diachronic analysis. Keywords: Crosslingual semantic change, cognates, temporal analysis, semantic analysis 1. Introduction Languages, our main tools of communication, evolve con- stantly: words obtain new and lose old meanings over time, they become popular or fade into obscurity. Because of its importance, language is studied by academics and pub- lic alike, as shown by the large number of publications and websites devoted to language evolution, etymology and semantic changes (Cresswell, 2010; Ayto, 2011; Lewis, 2013). Most of these focus on individual words only or are done on a small scale, mainly because the analysis re- quires manual work to locate occurrences of features in old texts, and then to compare manually their contexts or other characteristics. In the recent years, large amounts of digitized old books and texts were made available, such as Google’s Books ini- tiative (Michel et al., 2010) with 5% of books ever pub- lished. Computational approaches have also been con- ducted to analyze them (Gulordava and Baroni, 2011), proposing novel approaches for understanding lexical se- mantic change – for an overview, we refer to the survey by (Tahmasebi et al., 2018). However, to the best of our knowledge, no cross-language temporal analysis has been proposed in the literature using computational approaches and large data. In addition, most prior studies focused only on English, whereas comparing two or more languages can shed light on how they actually co-evolved over time. To study multiple languages over time, we assume the most intuitive approach: we focus on their similar connecting as- pects. We use in particular words in both languages that have the same origins and similar meaning, also known as cognate words. We propose to study the temporal charac- teristics of cognate words as …

Content truncated to 500 words