Skip to content Skip to main navigation Report an accessibility issue

Predicting Increases in Facebook Usage Frequency

BallingMichel Ballings, Ph.D.
Business Analytics and Statistics, UT
October 3rd, 2014, 2:30 – 3:30 PM

410 John D. Tickle Engineering Building

Dr. Michel Ballings joined the Department of Business Analytics and Statistics in 2014 after earning his Ph.D. in Applied Economics from Ghent University (Belgium). He holds a master’s degree in Business Economics from VLEKHO Business School in Brussels and a bachelor’s degree in Business Administration from University College Leuven. He teaches Marketing Analytics focusing on predictive analytics for analytical Customer Relationship Management (aCRM), including customer acquisition, customer retention, and customer development. His current research interests are concentrated in Social Media Analytics (SMA), a multi-disciplinary, data-intensive, and collaborative research area that lies at the intersection of computing, social media management, and machine learning. Dr. Ballings has worked with organizations such as Anheuser Bush InBev, Carglass, Vodafone, Friesland Campina, USG People, Concentra, Essent, and Club Brugge. He has research articles published in refereed journals and conference proceedings, such as Expert Systems With Applications, IEEE International Conference on Data Mining Workshops, Studies in Classification – Data Analysis and Knowledge Organization, and Management Intelligent Systems.

Talk Abstract: The purpose of this study is to (1) assess the feasibility of predicting increases in Facebook usage frequency, (2) evaluate which algorithms perform best, (3) and determine which predictors are most important and describe their relationship to the response. We benchmark the performance of Logistic Regression, Random Forest, Stochastic Adaptive Boosting, Kernel Factory, Neural Networks and Support Vector Machines using five times twofold cross-validation. The results indicate that it is feasible to create models with high predictive performance. The top performing algorithm was Stochastic Adaptive Boosting with a cross-validated AUC of 0.66 and accuracy of 0.74. The most important predictors include deviation from regular usage patterns, frequencies of likes of specific categories and group memberships, average photo album privacy settings, and recency of comments. Facebook and other social networks alike could use predictions of increases in usage frequency to customize its services such as pacing the rate of advertisements and friend recommendations, or adapting News Feed content altogether. The main contribution of this study is that it is the first to assess the prediction of increases in usage frequency in a social network.