Speaker: Subhadeep Paul, Assistant Professor, Department of Statistics
Title: Modeling Temporal Networks of Relational Events Data
Abstract: Continuous-time temporal networks or relational events data are commonly encountered in several application problems, including online social media communications, human mobility, financial transactions, and international relations. Such datasets consist of directed instantaneous interaction events among entities at specific time points. For example, in online social media, users interact with each other through events that occur at specific time instances such as liking, mentioning, commenting, or sharing another user's content. In international relations and conflicts, nations commit acts of hostility or disputes through discrete time-stamped events. The relational events data often exhibit community structure and strong dependence patterns through mutual excitations among node pairs. We will introduce statistical models and methods for analyzing such datasets combining network models and multivariate point processes. We will also describe scalable estimation methods and study the asymptotic properties of the estimators. Finally, we will demonstrate that the models can fit several real datasets well and predict temporal structures in those datasets.
Zoom Link: https://osu.zoom.us/j/98099909210?pwd=V2d4ZTk5bFRUWThSNmxJTTN5WEFJQT09
Note: Seminars are free and open to the public
