Should We Feed the Trolls? Using Marketer-Generated Content to Explain Average Toxicity and Product Usage

Marketers and researchers recognize the importance impact on consumer behavior of marketer-generated content (MGC) in social media channels. In this study, authors present a method to classify MGC using combination unsupervised supervised machine learning. They gather large data set posts from Facebook, Instagram, Twitter use time-series model (panel-data vector autoregression) demonstrate how can be used explain average toxicity part users. contribute field by examining what types lead toxic comments these product usage. The find that demonstrates quality products is aimed at creating sense belonging group are more likely increase toxicity. Furthermore, higher communities leads an usage focal product. Finally, results literature providing insights