We used two datasets in this work:

We used two datasets in this work: one for simple activitiesand one for complex activities. The first one is from ourprevious work [15] and its collection protocol can be foundin [15]. In that dataset, ten users performed seven physicalactivities while carrying smartphones (Samsung Galaxy S2)in their right jeans pocket and on their right wrist posi-tion, thereby emulating a smartwatch. These activities werewalking, jogging, biking, sitting, standing, walking upstairs,and walking downstairs. In sitting and standing activities, theuser sat and stood still alone without talking and doing anyother activity. The smartphones were used in same orientationon both positions. Because the new wrist-worn device areequipped with sensors like an accelerometer and a gyroscope,we simulate a smartwatch using Samsung Galaxy S2 on thewrist position. We collected data for multiple smartphone sen-sors, such as an accelerometer, linear acceleration, gyroscope,and magnetometer, but here we only consider an accelerometerand a gyroscope. The data was collected at 50 samples persecond for these sensors. For this dataset, each activity wasdone for 3 minutes by all participants (30 minutes of data foreach activity), thereby creating a balanced class distributionfor training and testing