Given its ascendancy as a way to make connections worldwide, social media is affecting all areas of people’s lives. This paper focuses on analyzing how Twitter bots interact with scholarly articles. The growing number and increasing complexity of Twitter bots make it hard to identify who is actually tweeting about scholarly articles. The purpose of this paper is to provide metrics to determine—based on an analysis of the relationship between Twitter bots and several research factors—whether or not a given scholarly paper has been disseminated via a bot. We developed and tested several supervised machine learning classification models that address this problem in relation to both numerical and categorical features, based on which the best results achieved was F1-score of 63%.