IMU-based Human Activity Recognition Dataset

This dataset was collected using an Inertial Measurement Unit (IMU) sensor connected to a Raspberry Pi, capturing the motion data for six different activities that include: walking upstairs, walking downstairs, walking, jogging, sitting, and standing. The target column, being activity was initially stored as text then later it was mapped into numerical values for easier analysis.

The dataset contains 15,980 observations with seven columns, six of them representing the sensor readings and one indicating the activity label. The six sensor data includes three-axis acceleration and three-axis gyroscope which provides information on both linear acceleration and angular velocity.

This dataset is useful for Human Activity Recognition (HAR) research, with potential applications in areas of machine learning-based classification and healthcare monitoring such as rehabilitation or fall detection.

Accelerometer Data:
- ax : acceleration along the X-axis.
- ay : acceleration along the Y-axis.
- az : acceleration along the Z-axis.

Gyroscope Data:
- gx : angular velocity along the X-axis.
- gy : angular velocity along the Y-axis.
- gz : angular velocity along the Z-axis.

Activity label:
1 = walking upstairs.
2 = walking downstairs.
3 = walking.
4 = sitting.
5 = standing.
6 = jogging.