Automatic Machine Learning for Recommender Systems

Crossings Minds provides machine learning-based recommendation systems that allow companies to integrate their data and easily obtain on-demand personalized recommendations for their users. Currently, onboarding a new customer requires significant work on the behalf of Crossing Minds engineers to create and polish a machine learning model for that specific customer. The research project will explore the field of auto-machine learning, which studies the way that machine learning models themselves learn. The goal would be to provide tools that make the machine learning decision-making process easier for Crossing Minds engineers and will explore potential opportunities in each step of the machine learning pipeline. Such tools will reduce the turnaround time for onboarding of new customers and will therefore improve the company’s scalability.

Faculty Supervisor:

Nisarg Shah;Scott Sanner;Eldan Cohen

Student:

Partner:

Crossing Minds Canada Inc.

Discipline:

Computer science

Sector:

Professional, scientific and technical services

University:

University of Toronto

Program: