The mathematics of predictive analytics applied to viral growth in social media

We will present the underlying mathematical concepts behind predictive analytics and discuss the pros and cons of different approaches in the context of social media analytics. Concrete examples will be presented along with an evaluation of the performance costs of each approach. We will show how to extract meaningful data from content shared on social media in real time as the data comes streaming in. We will also present current research that we are doing using deep learning methodologies to train a system to automatically recognize the characteristics of viral content.

Alain Chesnais is Chief Scientist at TrendSpottr. He is a long time ACM volunteer who served as President of the organization from 2010 through 2012. He served as ACM SIGGRAPH president from 2002 through 2005.