The Trumpiest Trump? Identifying a Subject’s Most Characteristic Tweets

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, pages 1653–1663, Hong Kong, China, November 3–7, 2019. c⃝2019 Association for Computational Linguistics 1653 The Trumpiest Trump? Identifying a Subject’s Most Characteristic Tweets Charuta Pethe Department of Computer Science, Stony Brook University, NY, USA cpethe@cs.stonybrook.edu Steven Skiena Department of Computer Science, Stony Brook University, NY, USA skiena@cs.stonybrook.edu Abstract The sequence of documents produced by any given author varies in style and content, but some documents are more typical or represen- tative of the source than others. We quantify the extent to which a given short text is char- acteristic of a specific person, using a dataset of tweets from fifteen celebrities. Such analy- sis is useful for generating excerpts of high- volume Twitter profiles, and understanding how representativeness relates to tweet popu- larity. We first consider the related task of bi- nary author detection (is x the author of text T?), and report a test accuracy of 90.37% for the best of five approaches to this problem. We then use these models to compute characteri- zation scores among all of an author’s texts. A user study shows human evaluators agree with our characterization model for all 15 celebri- ties in our dataset, each with p-value < 0.05. We use these classifiers to show surprisingly strong correlations between characterization scores and the popularity of the associated texts. Indeed, we demonstrate a statistically significant correlation between this score and tweet popularity (likes/replies/retweets) for 13 of the 15 celebrities in our study. 1 Introduction Social media platforms, particularly microblog- ging services such as Twitter, have become in- creasingly popular (Statista, 2019) as a means to express thoughts and opinions. Twitter users emit tweets about a wide variety of topics, which vary in the extent to which they reflect a user’s person- ality, brand and interests. This observation mo- tivates the question we consider here, of how to quantify the degree to which tweets are character- istic of their author? People who are familiar with a given author ap- pear to be able to make such judgments confi- dently. For example, consider the following pair of tweets written by US President Donald Trump, at the extreme sides of our characterization scores (0.9996 vs. 0.0013) for him: Tweet 1: Thank you for joining us at the Lin- coln Memorial tonight- a very special evening! Together, we are going to MAKE AMERICA GREAT AGAIN! Tweet 2: “The bend in the road is not the end of the road unless you refuse to take the turn.” - Anonymous Although both these tweets are from the same account, we assert that Tweet 1 sounds more char- acteristic of Donald Trump than Tweet 2. We might also guess that the first is more popular than second. Indeed, Tweet 1 received 155,000 likes as opposed to only 234 for Tweet 2. Such an author characterization score has many possible applications. With the ability to identify the most/least characteristic …

Content truncated to 500 words