“…Prior work on online health forums focuses on analyzing participation (Schultz et al, 2019;Sadeque et al, 2015;van Campen and Iedema, 2007), social connection and engagement (Sharma et al, 2020a;Kushner and Sharma, 2020;Smith-Merry et al, 2019;Halder et al, 2017), and modelling user behavior (Hosseini and Caragea, 2021;Sharma et al, 2020b;Buechel et al, 2018;Elliott et al, 2018;Pérez-Rosas et al, 2017;Gibson et al, 2016;Choudhury and De, 2014). Further, existing research focuses on the linguistic analysis of conversations in social media platforms like Facebook (Dehouche, 2020;Tian et al, 2017;Vyas et al, 2014;Das and Gambäck, 2013), Reddit and Twitter (Zomick et al, 2019;Rudra et al, 2019;Kiesling et al, 2018;Rijhwani et al, 2017;Tran and Ostendorf, 2016;Celli and Rossi, 2012;Bak et al, 2012;Gouws et al, 2011), but has paid little attention to analyzing the conversations in small, close-knit WhatsApp communities. In this paper, we propose a hierarchical annotation framework to facilitate the analysis of health-focused WhatsApp groups.…”