Measuring Influence and Topic Dependent Interactions in Social Media Networks Based on a Counting Process Modeling Framework.

Data extracted from social media platforms, such as Twitter, are both large in scale and complex in nature, since they contain both unstructured text, as well as structured data, such as time stamps and interactions between users. Some key questions for such platforms are (i) to determine influential users, in the sense that they generate interactions between members of the platform and (ii) identifying important interactions between nodes in the corresponding user network. Regarding the first question, common measures used both in the academic literature and by companies that provide analytics services are primarily variants of the popular web-search PageRank algorithm applied to networks that capture connections between users. In this work, we develop a modeling framework using multivariate interacting counting processes to capture the detailed actions that users undertake on such platforms, namely posting original content, reposting and/or mentioning other users’ postings. Based on the proposed model, we also derive a novel influence mea- sure. We discuss estimation of the model parameters through maximum likelihood and establish their asymptotic properties. The proposed model and the accompanying influence measure are illustrated on a data set covering a five year period of the Twitter actions of the members of the US Senate, as well as mainstream news organizations and media personalities. We then turn our attention to the problem of identifying important interactions both globally and also based on the particular topics under discussion. We modify the previously introduced modeling framework, so that topic dependent interactions can also be identified. We extend our previous algorithm to accommodate the new framework and also establish asymptotic properties of the key model parameters. We illustrate the results on the same Twitter data set.PhDStatisticsUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/113379/1/donggeng_1.pd