Data Mining is the computational process of extracting meaningful patterns and trends in large data sets, and the process of statistical learning from data. This course concentrates on the statistical and computational aspects of data mining. Students will learn concepts such as model assessment, model selection, model complexity, overfitting, train and test error, and loss functions. Students will learn how to use and implement supervised learning methods such as multiple (non)-linear regression, multiple logistic regression, linear (and quadratic) discriminant analysis, decision trees, and random forest. Unsupervised learning methods such as clustering and dimensionality reduction are also presented with real data applications. Students will also learn how to apply these methods to real-world problems and quantify and manage the risk. The intention is to concentrate more on the applications of the methods to gain business insight.Students who have taken STA 3920 cannot take STA 3950.S