Ridesharing systems are transforming urban transportation in an economically feasible, environmentally friendly and socially beneficial way. The Ridesharing problem is an intriguing problem in transportation research and has been studied for several years. The objective is to find efficient assignments for transportation of items through a complex network while respecting the capacity constraints of the available vehicles in the fleet. In recent times, more work has been focused on using data driven and intelligent approaches to solve problems arising in transportation research. In this thesis we aim to solve the ridesharing problem using neural network architectures. Our results indicate that the neural network architecture models generate solutions with reasonable accuracy and consume less computational time.