Data Vault Ensemble Modeling

Beyond Data Modeling: Artificial Intelligence and Machine Learning

Dr. Raja Sooriamurthi, Carnegie Mellon University

The combined advances in hardware, software, and communication over the past few decades form the basis of our current disruptive age of Data.  Massive amounts of data (terabytes and beyond) are available in a range of domains: science, commerce, finance, healthcare, social media, real-time sensors etc. At historically unprecedented levels we are able to collect, transmit, curate, and process huge amounts of data at enormous speeds resulting in our ability to do ongoing tasks better and to do tasks we couldn’t do before.  Data, like fossils, tells us something about the past.  The premise is that past patterns are predictive of future behavior.  For example, a casino may want to identify whether there is a certain group of customers from which more business occurs.  A cell phone company may want to know if there is a risk of customers leaving for another carrier. In this talk, we will discuss various analytic tasks that facilitate such actionable insights such as prediction, optimization, recommendation, classification, clustering etc.

Dr. Raja Sooriamurthi is a Teaching Professor with the Information Systems Program at Carnegie Mellon University, Pittsburgh. His research and teaching interests span the fields of artificial intelligence and software development with a current focus on data-driven decision making. Along with his co-authors, he has investigated a novel approach to teaching critical thinking and problem solving termed puzzle-based learning resulting in the book Guide to Teaching Puzzle-based Learning (Springer, 2014). In addition to his university courses, Raja has taught several conference and industry workshops in the US, Australia, the Middle-East (Qatar, The United Arab Emirates), and India.  Over the years, since a graduate student, his pedagogical efforts have been recognized with several awards for teaching excellence.