Cross-Domain Recommender Systems with Limited Human Annotated Data

Recommender systems are artificial intelligences that, as their name would suggest, make recommendations based on provided inputs. For example, recommending jobs a person can apply to based on their resumes. Existing research on recommender systems in the Job and Education domains have focused on a single domain. Our research focuses on bridging the gap between the Job and Education domains in an environment where we have little to no data created by humans. We demonstrate our methodologies by creating a recommender system for Prior Learning Assessment and Recognition (PLAR). PLAR is a system through which people can obtain college credits based on past life experience, such as job history. Our partner organisation hopes that such a recommendation tool can be used to streamline the PLAR process and promote education equity in Canada.

Faculty Supervisor:

Robert Mercer

Student:

Partner:

KnowMeQ Inc.

Discipline:

Computer science

Sector:

Manufacturing; Professional, scientific and technical services

University:

The University of Western Ontario

Program: