The goal of this project was to solve the problem of a telecom operator losing customers to its competitor. A predictive model was built with classification algorithms to predict the likelihood of churn (leaving a service) of a mobile telephone user. Random Forest classifier, XGBoost & Decision Tree algorithms were used for training & testing the model while feature engineering was extensively carried out to fine-tune and improve the accuracy of the model. More work was done in cleaning the data – handling missing values and inconsistency in the data. This project was submitted in a Kaggle Competition. - View it on GitHub