End-to-End Data Pipeline and Predictive Modelling for Insurance Analytics

Abstract

Insurance telematics is an emerging and exciting field. It combines the advancements in GPS tracking, 
computational analytics, data processing, and machine learning into a useful tool to help insurance 
companies make the best product for their consumers. This is why National Indemnity looked to 
implement a telematics portion to their business processes of underwriting insurance policies and 
sponsored a School of Computing Senior Design project. In this report, we will first review existing 
solutions that been used to solve problems and subproblems like that we are given in this project. We 
then propose designs for the data pipeline and machine learning model that will be optimal in 
providing predictions on the risk level of drivers. National Indemnity will be able to use this project to 
leverage predictions to optimize insurance rates to more accurately account for risk among the 
insured.