I thrive on leveraging data to bring business value through a smart mix of advanced statistics, analytics and ML, design thinking, and business knowledge.
My experience has taught me the importance of skillfully translating complex business problems into lean, innovative, implementable, and sustainable data science solutions.
I have experience leading successful data science projects focused on end-user impact and increasing ROI. My approach includes strong collaboration, integration od data science best practices, and matching right technology with the right level of rigor, creativity and flexibility to the right and well-defined business problem for tailored results.
Technical advisor to Data Science SVP & ML Team Lead
Vision: Build a powerful and lean ML solution stack adapted to a rapidly changing data ecosystem
Developed the team's ML products & growth strategy.
Lead prototype projects for sales impact measurement systems
Developed workshops and training for the team's data scientists
Organized and lead implementation of DS best practices throughout the data science teams
Led and built end-to-end marketing ROI solutions, optimization, and business forecasting
Focus: Empowering stakeholders with actionable, high-impact media measurement solutions.
Translated stakeholder business needs into statistical problems.
Conceptualized and developed prototypes for standardized ROI modeling systems
Developed data-driven solutions for strategy and spend optimization
Built end-to-end MMM solutions ready to production.
Coached and mentored data scientists and DS managers on MMM & ROI project delivery.
Increased model velocity through automation and agile management
Led data science efficiency strategies and best practices
Strategic development of DS/ML technical strategy
Developed custom measurement approaches
Leveraged advanced statistical methods applied to large unstructured datasets
Translated technical findings and complex ideas into easy-to-relate terms and concepts for stakeholders and non-technical team members.
Lead Data Scientist
Technical lead of a team of data scientists/engineers
Contributed to the creation of a data-driven system using machine learning.
Built a predictive modeling system prototype.
Senior Data Scientist
Developed and prototyped ML analytics methods for user-experience optimization
Proposed robust nonparametric approaches for large datasets.
Machine learning
Scientific Publications
· Roy, M.H., Larocque, D. (2019). Prediction Intervals for Random Forests. Journal of Statistical Methods for Medical Research, 0962280219829885.
· Owen, V.E., Roy, M.H., Thai, K.P., Burnett, V., Jacobs, D., Keylor, E. (2019). Detecting Wheel Spinning and Productive Persistence in Educational Games. Educational Data Mining 2019.
· Roy, M.H., Larocque, D. (2012). Robustness of Random Forests for Regression. Journal of Nonparametric Statistics, Volume 24, Issue 4.
· Lahaise, C., Pozzebon, M., Roy, M.H., L’intelligence d’affaires au service d’un programme de developpement durable (Business Intelligence for Sustainable Development). Actes du Congres 2011 de l’ASAC, Montreal, Quebec, Canada.
Conference Presentations
· Adapting Predictive Modelling to e-Learning Data. IDEAS Conference 2018. Los Angeles Convention Center, October 2018.
· Robust Variable Selection with a Multiple Step Bootstrap Procedure. Joint Statistical Meeting. Seattle, USA, August 2015.
· A Study of Random Forests Using Robust Aggregation Methods and Splitting Critetion. Joint Statistical Meeting. Montreal, Canada, August 2013.
· Robustness of Random Forests for Regression. International Conference on Robust Statistics (ICORS). Burlington, USA, August 2012.