Trend Emergence and Evolution Prospection using Machine Learning in Fully Developed Insurance Claims

This Mitacs Business Strategy Internship project unites Intact Insurance’s extensive insurance data and industry expertise with Western University’s Banking Analytics Lab’s advanced skills in deep learning, actuarial science, and data analysis. The project’s goal is to build an unsupervised deep learning model for detecting significant changes in claim patterns, surpassing the limitations of traditional change-point detection methods. This will enable Intact to proactively manage risks, optimize premium calculations, and respond more effectively to emerging threats. The intern will gain hands-on experience in developing and validating this model over real-life data lakes at the partner institution, significantly enhancing their skills in deep learning, data analysis, and the application of these techniques to real-world insurance problems. The collaboration brings together the complementary strengths of both partners; Intact provides real-world data, business context, and industry mentorship while Western University offers advanced research expertise and computational resources for model development and validation, creating a strong foundation for future collaborations.

Faculty Supervisor:

Cristian Bravo Román;Kristina Sendova

Student:

Partner:

Intact

Discipline:

Business

Sector:

Finance and Insurance

University:

The University of Western Ontario

Program: