Recognizing problem behaviors of children with developmental disabilities using smartwatch
Yeongju Lee, Minseok Song · 2016
We develop a scheme to analyze problem behaviors of children with developmental disabilities by making use of 3-axis accelerometer, gyroscope, and microphone sensors in a smartwatch. It consists of four phases: classification of behaviors, feature collection, classifier construction and behavior recognition. Real measurements show that the recognition accuracy is above 92% for all test cases.