Examining the Shape of Data through Topological Data Analysis

For our May Meetup, we're thrilled to welcome Jean-Ezra Yeung (http://www.linkedin.com/in/jeanezrayeung) back to DC to tell us about a new approach to learning from and exploring data that leverages ideas from the mathematical field of topology. In many machine learning and statistical approaches to data analysis, you are reasoning about points in a high-dimensional space. But in Topological Data Analysis, you replace the points with mathematical objects that identify and capture important properties of the shape of the data and relationships among points. Jean-Ezra, who works for Ayasdi, a company working to commercialize this technology, will walk us through some of the math, and all of the reasons you should care about TDA.

• 6:30pm -- Networking, Empenadas, and Refreshments

• 7:00pm -- Introduction

• 7:15pm -- Presentation and Discussion

• 8:30pm -- Data Drinks (Tonic, 2036 G St NW, 3rd Floor)

Abstract:

The current challenge in big data is the ability to process huge amounts of data and extract insights. Using traditional methods, we would write a query and then apply statistical methods or machine learning algorithms to analyze the data. This is problematic because it requires a data analyst to know a priori what question to ask and consequently limits the ability to find insights. Topological data analysis (TDA) does not require writing queries. Instead, it harnesses information from the shape of data by creating a compressed representation. What this means is that we can discover insights by exploiting the geometric properties of the data. We discover answers for questions we did not know to ask. This talk will discuss how TDA serves as a framework for machine learning algorithms. We will demonstrate how to conduct both supervised and unsupervised learning.

Bio:

Jean-Ezra Yeung is a Data Scientist at Ayasdi (http://www.ayasdi.com/). He previously worked in technical consulting for pharmaceutical and medical devices clients, and health policy evaluation for the Department of Health and Human Services. He received his BA in Economics from New York University, and an MPH from Columbia University.

Sponsors:

This event is sponsored by the GWU Dept. of Decision Sciences (http://business.gwu.edu/decisionsciences/), Cloudera (http://www.cloudera.com/), Statistics.com (http://bit.ly/12YljkP), MicroStrategy (http://www.microstrategy.com/free), IBM Analytics Solution Center (https://www.ibm.com/ascdc), Novetta Solutions (http://novetta.com), Elder Research (http://datamininglab.com/), Five 9 Group (http://www.five9group.com/), and InformIT (http://informit.com/). Would you like to sponsor too? Please get in touch!