Phone sensor data while driving a car and normal or aggressive driving behaviour classification

Published: 3 June 2021| Version 1 | DOI: 10.17632/5stn873wft.1
Shardul Nazirkar


(1_20210317_184512.csv, 2_20210317_171452.csv) - Data has been recorded on an android phone attached to the dashboard of the car. Data was collected while driving the car on city roads in mild traffic. The parameters recorded are : Longitude, Latitude, Speed, Distance, Time, Acc X, Acc Y, Acc Z, Heading, gyro_x, gyro_y, gyro_z (Acc - Accelerometer, Gyro - Gyroscope). (3_FinalDatasetCsv.csv) - Above two files merged but only Accelerometer and Gyroscope parameters are present. Also another column present called 'label'. This column basically has two classes - '0' and '1'. '0' represents normal driving behavior while '1' represents aggressive driving behavior. These datasets can be used for training machine learning models that classify a person's driving into normal or aggressive.



Driver Behavior