Facial recognition has recently become very important in the fields of artificial intelligence, robotics, security, trade, etc. It can be used in stress analysis in industries by capturing the faces of the employees and by analysing their emotions in order to put forward the suitable therapies for cure in required. Face recognition can also be used in shopping malls to analyse the facial expressions of the users while purchasing various products, thus capturing their interests for purchasing the same. This can then be used together with the other data for the process of data analysis and further promotions for the users in terms of their liking products. In order to analyze facial expressions, a variety of techniques are employed, including support vector machines, the k-nearest neighbor algorithm, and multilayer perceptron. Although there are numerous ways to recognize expressions using machine learning and artificial intelligence techniques, this project makes an attempt to use Convolutional neural networks and MediaPipe to identify various expressions from the faces and classify them appropriately. In order to learn face representation from a smaller data set, this project suggests modifying a Convolutional neural network. Convolutional neural networks are well known to have a higher accuracy when compared to traditional machine learning methods. Our system continuously captures the images of the user using Webcam and analyses the facial expressions accordingly using a convolutional neural network and to identify the mood of the user which can be used for further analyses.