Sachin Adlakha is a Sr. Data Scientist at Ayasdi Inc. and also a post doctoral scholar the California Institute of Technology's Center for the Mathematics of Information. He received his Ph.D. from Stanford University in 2010 where his thesis was focused on the design of complex interconnected systems.
One of the big challenge in big data is to figure out what features should one use to build successful predictive models. As the number of features in the data increases, there is an exponential increase in possible combinations of feature space to explore. In this talk we will look at a new and innovative technique called "topological data analysis" and understand how it can be used to extract relevant features that are essential to model building. This technique works by looking at the data in its entirety and find data points that are "similar". We will also present some interesting practical applications of this technique.