TITLE:
Machine Learning for Social Systems: Modeling Opinions, Activities, and 
Interactions

ABSTRACT:

The proliferation of user-generated content on the web provides a 
wealth of opportunity to study human behavior through users' online 
traces. My research aims to model such behavior through the lens of 
opinions. Opinions are ubiquitous online: people share opinions 
explicitly in the form of ratings, reviews, and "likes"; and 
implicitly, through the products they purchase, the connections they 
form, and the communities they join. Such opinions are much more than
a simple number or quantity: they are multi-dimensional, they develop 
over time, and they are influenced by our friends and communities. In 
this talk, I will present machine learning techniques in order to 
model and understand such rich, structured, and time-evolving data.