

We not only study what makes a meme viral, but we also analyze what incremental predictive power image related attributes have over textual attributes on memes popularity. We stand apart from other authors by investigating whether the success of a meme can be predicted based on its content alone, excluding social network factors.

Using advanced machine learning techniques (such as convolutional neural networks, gradient boosting, and random forest), we perform a content-based analysis of what makes a meme successful, considering several features from both text and image data. The main contributions of this work can be summarized as follows: Here we analyze the popularity of coronavirus memes based on 129,326 records scraped from Reddit, the largest social news and entertainment site. 2018 Tsur and Rappoport 2015 Wang and Wood 2011). Our paper joins a growing body of literature that employs network science and data science techniques to predict the popularity of Internet memes (Weng et al. Along the way, it exposes what topics were popular on the Internet during the global COVID-19 pandemic. This paper investigates the relationship between a meme’s content, excluding social network features, and its popularity. In other areas of human achievement viral success is closely linked with merit (Yucesoy and Barabási 2016), but it is unclear what characteristics lead a meme to have merit.

Many authors have explored the social network factors that lead a meme to go viral but bracketed the impact that meme content may have on popularity (Gleeson et al. 2011), they often show what the public is noticing most. Memes tend to reflect pressing global issues and while they are not always loyal to the facts (Simmons et al. Today, memes are not only a source of humor but also draw attention to poignant cultural and political themes (Brodie 2009). The increasingly participatory nature of the Internet has made memes into a social phenomenon, created, altered, and spread by Internet users themselves. As a result, Web culture is moving faster than ever and social media sites have exploded with coronavirus memes as people all over the world try to take this serious situation with a pinch of humor (Bischetti et al. In the time of the COVID-19 pandemic, memes have become an even more important part of social life since due to social distancing orders more people turned to the Internet for everyday interactions. Due to their popularity, memes have received considerable attention in areas such as pop culture, marketing, sociology, and computer science (Bauckhage et al. Over the past decade, Internet memes have become a pervasive phenomenon in contemporary Web culture (Laineste and Voolaid 2017).
