Wetin dey with these comments? Modeling Sociolinguistic Factors Affecting Code-switching Behavior in Nigerian Online Discussions

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 6204–6214 Florence, Italy, July 28 - August 2, 2019. c⃝2019 Association for Computational Linguistics 6204 W´etin dey with these comments? Modeling Sociolinguistic Factors Affecting Code-switching Behavior in Nigerian Online Discussions Innocent Ndubuisi-Obi∗ School of Information University of Michigan innoobi@umich.edu Sayan Ghosh∗ Department of EECS University of Michigan sayghosh@umich.edu David Jurgens School of Information University of Michigan jurgens@umich.edu Abstract Multilingual individuals code switch between languages as a part of a complex communi- cation process. However, most computational studies have examined only one or a handful of contextual factors predictive of switching. Here, we examine Naij´a-English code switch- ing in a rich contextual environment to un- derstand the social and topical factors elicit- ing a switch. We introduce a new corpus of 330K articles and accompanying 389K com- ments labeled for code switching behavior. In modeling whether a comment will switch, we show that topic-driven variation, tribal affilia- tion, emotional valence, and audience design all play complementary roles in behavior. 1 Introduction Multilingual individuals frequently switch be- tween different languages throughout a discourse, a process known as code switching (Heller, 2010; Gamb¨ack and Das, 2016). This switching pro- cess is thought to be driven from a variety of factors, including grammatical constraints (Pfaff, 1979; Poplack, 1980), audience design (Gumperz, 1977; Bell, 1984), or even to evoke a specific perception of the speaker’s identity (Niedzielski, 1999; Schmid, 2001). In common social situa- tions, many of these factors are in play, yet we of- ten do not have an idea of how they interact. Here, we present a large scale study of code switching in Nigeria between English and Naij´a, the widely- spoken Nigerian creole, to quantify which factors predict switching. Computational studies of code switching have largely focused on linguistic aspects of switching (Solorio and Liu, 2008; Adel et al., 2013; Vyas et al., 2014; Hartmann et al., 2018). However, sev- eral recent works have begun to examine the con- textual factors that influence switching behavior, ∗Authors contributed equally. finding that the topic driving a discussion spurs on language variation (Shoemark et al., 2017; Stew- art et al., 2018) and that individuals are sensitive to the scope of their audience when choosing a lan- guage (Papalexakis et al., 2014; Pavalanathan and Eisenstein, 2015). Given that the social context is known to be strongly influential on code switching (Gumperz, 1977; Thomason and Kaufman, 2001; Gardner-Chloros and Edwards, 2004), our work builds on these recent advancements to quantify the impact of social and contextual factors influ- encing code switching. Here, we examine the social and contextual fac- tors predictive of English-Naij´a code switching in online discussions across five major Nigerian newspapers. Our work makes three contributions towards computational sociolinguistics. First, we introduce a massive new corpus of Naij´a and En- glish text that presents code switching behavior in context, using 330K articles and 389K comments from nine years of longitudinal data. Second, we develop a new classifier for distinguishing Naij´a and English, …

Content truncated to 500 words