Using Deep Analysis of Driver Behavior for Vehicle Theft Detection and Recovery

Advancement in technology has resulted to a vast amount of spatio-temporal data. When this data is properly mined it translates to knowledge, which is used for auto-theft detection and recovery. In this study, we analyze vehicle driving dataset using deep learning algorithms such as the Convolutional Neural Network, the Recurrent Neural Network with Long Short-Term Memory, Deep Boltzmann Machines and Deep Autoencoders, as well as bio-inspired algorithms such as the Particle Swarm Optimization, Artificial Bee Colony, Ant Colony Optimization and Bat Algorithm. The performance of these algorithms is analyzed using benchmark functions such as the Ackley function, Rastrigin function, Rosenbrock function, Sphere function, Schaffer function and Himmelblau's function. Finally, efficiency of the algorithms was measured using the Mean Squared Error.