Active Learning Strategies for Time-series

Acquiring labelled data for training supervised learning models is a barrier to a wider adoption of machine learning strategies in several industries. The data annotation task is generally complex and time-consuming. In this project, we will use raw unlabeled data generated by fiber sensors to reduce the burden of labelling data with problems within the fiber industry.

Faculty Supervisor:

Amilcar Soares;Vinicius P. da Fonseca

Student:

Partner:

Instrumar

Discipline:

Computer science

Sector:

Manufacturing

University:

Memorial University of Newfoundland

Program:

Accelerate

Current openings

Find the perfect opportunity to put your academic skills and knowledge into practice!

Find Projects