Speaker : Paula Lago, Department of Electrical and Computer Engineering, Concordia University. Member of PERFORM Center.
Whether on (wearable or implantable), in (ingestible) or around our bodies (ambient), we are now surrounded by sensors that continuously measure our behaviors and physiology with unprecedented extent and frequency.
How to make sense of the heterogenous, uncertain, and continuous data to improve our understanding of health and wellbeing?
In this talk, I will show how we can use data mining and machine learning techniques to understand short and long-term patterns from sensor data with two case studies. In the first one, I will show how we can analyze smart home sensor data to discover routines and routine changes that could be indicators of health deterioration. In the second, how to analyze wearable sensor data, which is more uncertain and noisier, to understand the current activities of an individual.