The evaluation metrics for success is the accuracy level of correctly predicting the facial expression in the test data. In order to make the machine intelligent enough to recognize and differentiate between different facial expressions accurately, based on the knowledge of our team, we first thought of using basic machine learning techniques such as Artificial neural network, Linear Discriminant Analysis and Support Vector machine. The team has done extensive research on evaluating both the pros and cons of different machine learning techniques by analysing parameters such as complexity, results, feasibility, flexibility, runtime, model consistency, etc. After analysing different training models the team has settled down on designing a Convolutional Neural Network model for facial recognition.