Transferring Knowledge of Daily Life Routines with Wearable Accessory
Benefits from characterization of the patient’s routines goes into medical diagnosis and care of elderly people to assess individual’s physical and mental conditions. Annotation techniques to collect reference data are expensive, time-consuming and error-prone.
This project aims to implement an energy-efficient daily routine recognition algorithm able to learn activities and to transfer available knowledge to new activities. Daily life activity dataset from wrist-worn inertial sensors is provided.
|Project type||Bachelor/Master Thesis|
|Work distribution||100% Data Analysis|
|Useful knowledge||Machine learning|