A dataset for elderly action recognition using indoor location and activity tracking data
Description
This dataset describes the data collected from physical activity and indoor location systems. these data can be used to train two different machine learning model then the outcome from both systems to be combined to study the Actions (sub-activity). The data from the physical activity tracking system are accelerometer X-axis, Y-axis, Z-axis, and Gyroscope X-axis, Y-axis, Z-axis, and the name of the activity performed by the volunteer obtained from the MPU6050 at sampling rate of 20 Hz. The MPU6050 sensor was placed inside the waist wearable device worn by the subject. The data from indoor location positioning obtained is the Received Signal Strength Indicator (RSSI) from Bluetooth Low Energy (BLE) beacon and BLE scanners. The indoor locations dataset is recorded in the Digital Smart Home Lab (DSHL) located at Multimedia University. DSHL consists of 5 rooms namely living room(10m×5m), kitchen(4.5m×4.5m), bedroom (5m×4.5m), Office (6m×4m), Foyer(4m×3.8m), and Toilet(1.6m ×0.95m). The advertising interval for the BLE beacon is set to 100ms with a transmitting power of up to -30 dBm. If an RSSI is equal to -120, this means that the BLE beacon is very far from the BLE scanner or not found. The actions data obtained by integrating the results from indoor location and activities tracking systems to study the actions. Actions mean the outcome from doing one activity in one location because the same activity at the different location might illustrate something different for example lying in the bedroom means sleeping but lying in the living room means relaxing while lying in the toilet means an emergency.