New Relic Churn Prediction & Prevention Micro-Segmentation and Predictive Analytics

New Relic, Inc. Computer Science, 2016-17

Liaison(s): Kira Chuchom, Tarun Arora
Advisor(s): Ran Libeskind-Hadas
Students(s): Felis Perez, Rose Choi, William Chen (PM-S), Yiqing Cai (PM-F)

The goal of the New Relic Clinic project is to develop a data-driven approach to predicting potential churn. New Relic has collected substantial data. The types of data that has been collected have varied over time, leading to a heterogeneous dataset that is difficult to analyze. To address this issue, the New Relic Clinic project aims to first restructure the data to be consistent and then apply machine learning techniques to identify features capable of predicting churn.