Automatic Identification and Classification of Bragging in Social Media

Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics Volume 1: Long Papers, pages 3945 - 3959 May 22-27, 2022 c⃝2022 Association for Computational Linguistics Automatic Identification and Classification of Bragging in Social Media Mali Jin1, Daniel Preo¸tiuc-Pietro2, A. Seza Do˘gruöz3, Nikolaos Aletras1 1University of Sheffield, 2Bloomberg, 3Ghent University {mjin6, n.aletras}@sheffield.ac.uk dpreotiucpie@bloomberg.net as.dogruoz@ugent.be Abstract Bragging is a speech act employed with the goal of constructing a favorable self-image through positive statements about oneself. It is widespread in daily communication and es- pecially popular in social media, where users aim to build a positive image of their persona di- rectly or indirectly. In this paper, we present the first large scale study of bragging in computa- tional linguistics, building on previous research in linguistics and pragmatics. To facilitate this, we introduce a new publicly available data set of tweets annotated for bragging and their types. We empirically evaluate different transformer- based models injected with linguistic informa- tion in (a) binary bragging classification, i.e., if tweets contain bragging statements or not; and (b) multi-class bragging type prediction including not bragging. Our results show that our models can predict bragging with macro F1 up to 72.42 and 35.95 in the binary and multi-class classification tasks respectively. Fi- nally, we present an extensive linguistic and error analysis of bragging prediction to guide future research on this topic.1 1 Introduction The desire to be viewed positively is a key driver of human behavior (Baumeister, 1982; Leary and Kowalski, 1990; Sedikides, 1993; Tetlock, 2002) and creating a positive image often leads to per- sonal rewards (Gilmore and Ferris, 1989; Hogan, 1982; Schlenker, 1980). Self-presentation strategies are means for individuals to build and establish this positive social image to meet their goals (Goffman et al., 1978; Jones et al., 1982; Jones, 1990; Bak et al., 2014a). Bragging (or self-praise) is one of the most common strategies and involves disclos- ing a positively valued attribute about the speaker or their in-group (Dayter, 2014, 2018). 1Data is available here: https://archive.org/ details/bragging_data Social media platforms tend to promote self- presentation tendencies (Chen et al., 2016) and allow users to craft an idealized self-image of them- selves (Chou and Edge, 2012; Michikyan et al., 2015; Halpern et al., 2017). Self-presentation on- line is predominantly positive (Chou and Edge, 2012; Lee-Won et al., 2014; Matley, 2018). Fur- thermore, self-promotion is acceptable and even de- sired in certain online contexts (Dayter, 2018). This is also amplified by social media platforms through the presence of likes or positive reactions to users’ posts (Reinecke and Trepte, 2014) which often are used to quantify impact on the platform (Lampos et al., 2014). Bragging in particular was found to be more frequent on social media than face-to-face interactions (Ren and Guo, 2020). However, bragging is considered a high risk act (Brown and Levinson, 1987; Holtgraves, 1990; Van Damme et al., 2017) and can lead to the op- posite effect than intended, such as dislike or de- creased perceived competence (Jones et al., 1982; Sezer …

Content truncated to 500 words