CRCS Lunch Seminar

Date: Monday, November 9, 2015
Time: 11:30am – 1:00pm
Place: 33 Oxford St., Maxwell Dworkin 119

Speaker: Manuel Gomez Rodriguez (MPI for Software Systems)

Title: Learning Opinion Dynamics in Social Networks

Abstract: Social media and social networking sites have become a global pinboard for exposition and discussion of news, topics, and ideas, where social media users increasingly form their opinion about a particular topic by learning information about it from her peers. In this context, whenever a user posts a message about a topic, we observe a noisy estimate of her current opinion about it but the influence the user may have on other users’ opinions is hidden. In this talks, we introduce SLANT, a probabilistic modeling framework of opinion dynamics, which allows the underlying opinion of a user to be modulated by those expressed by her neighbors over time. We then identify a set of conditions under which users’ opinions converge to a steady state, find a linear relation between the initial and steady state opinions, and develop an efficient estimation method to fit the model parameters from historical fine-grained opinion and information diffusion event data. Experiments on data gathered from Twitter, Reddit and Amazon show that our model provides a good fit to the data and more accurate predictions than alternatives.

Biography: Manuel Gomez Rodriguez is a tenure-track faculty at Max Planck Institute for Software Systems. Manuel develops machine learning and large-scale data mining methods for the analysis, modeling and control of large real-world networks and processes that take place over them. He is particularly interested in problems arising in the Web and social media and has received several recognitions for his research, including an Outstanding Paper Award at NIPS'13 and a Best Research Paper Honorable Mention at KDD'10. Manuel holds a PhD in Electrical Engineering from Stanford University and a BS in Electrical Engineering from Carlos III University in Madrid (Spain). You can find more about him at http://www.mpi-sws.org/~manuelgr/.