This project aims to predict whether a person earns more than 50k/year, taking into account some socio-economic input attributes. Supervised machine learning methods are applied in order to solve the classification task and different models are performed. The analysis deals with the class imbalance problem, handled with the minority class undersampling. Trying to understand the most valuables attributes, feature selection is implemented using both multivariate filter and wrapper. J48, Random Forest and Logistic Regression appear to be the most appropriate models. The performance is evaluated in terms of Accuracy, Precision, Recall, F1 measure and ROC-Curve.