The detailed analysis of the data, can be understood that it needs to be properly pre-processed to feed it to the predictive model. Therefore, the data is converted to a format the model best understands, and then exploratory data analysis is performed. Lot of facts has been discovered in this analysis. Later in the prediction part, there are 3 main models used. With the help of python’s well-known library package “sci-kit learn” it is easily possible to implement and execute different types of model. Initially, we use a model called Support vector classifier. This gives a decent amount of accuracy of predictions. Later the same model is optimized and cross validated. It still stays with decent accuracy rate. Secondly, used other models like linear regression, adaptive boosting, and sci-kit learn’s simple neural network called MLP – multi-layered perceptron package. All give fair amount of accuracy. Adaptive booting gives 100 percent accuracy. Since the out is binary, adaptive boosting algorithm is the most accurate one.