You don’t often equate home loans with actuarial science. However, using advanced statistical and machine learning methods, Society of Actuaries (SOA) member Carlos Brioso, FSA, CERA, crafted a data solution that would help low-income consumers secure home loans. That innovative and inclusive solution earned him the coveted distinction of Kaggle Competitions Master.
Through Kaggle competitions, participants compete to produce the best models for predicting and describing data provided by third-party competition hosts. Brioso, who is the director of the Center for Data Science and Artificial Intelligence at New York Life, now ranks in the top 1% of more than 100,000 data scientists and machine learning engineers worldwide in Google’s Kaggle community. Reflecting on his participation, Brioso says, “The real prize comes from the learning experience and [being able to] contribute my code to the community.”
The SOA’s Kaggle Involvement Program recognizes actuaries for participating in these Kaggle competitions, and it is an opportunity for actuaries to develop and showcase their data modeling skills. Brioso is among the 15 individuals who earned prizes and bragging rights through the SOA’s 2018 Kaggle Involvement Program. Individuals or teams are recognized if they place in the top 10% of their competition or achieve Kaggle Competitions Master status.
“It’s important for actuaries to get more involved in data science,” says Brioso. “Actuaries have business acumen, but we have to combine it with the ability to handle data. Kaggle is a time commitment, but the skills you gain are worth it.
2018 Kaggle Involvement Program Winners
SOA members participated in data science competitions that challenged them to use cutting-edge technology to build models and find solutions with important societal implications, from preventing environmental accidents at sea to ensuring underserved populations have access to fair lending practices. Congratulations to these members who placed in the top 10% of competitors worldwide in these Kaggle challenges.
Airbus Ship Detection Challenge
Airbus Group Inc. challenged Kagglers to aid in building a model that detects ships in satellite images as quickly as possible to aid in preventing infractions at sea such as environmentally devastating ship accidents, piracy, illegal fishing, drug trafficking and prohibited cargo movement.
Home Credit Default Risk
Home Credit Group’s challenge asked competitors to help unlock the full potential of their data to ensure that clients capable of repayment are not rejected and that loans are given with a principal, maturity and repayment calendar that will empower their clients to be successful.
Santander Value Prediction Challenge
In this competition, Santander Group asked participants to identify the value of transactions in order to anticipate customer needs and provide personalized customer service solutions.
TGS Salt Identification Challenge
Several areas of Earth with large accumulations of oil and gas also have huge deposits of salt below the surface. To assist in avoiding potentially dangerous situations for oil and gas company drillers, TGS asked Kaggle’s machine learning community to build an algorithm that automatically and accurately identifies if a subsurface target on a seismic image also contains salt.
Copyright © 2019 by the Society of Actuaries, Schaumburg, Illinois.