Graduate Courses

Part A: Theory. Introduction to databases and SQL. Data preparation, data cleaning and completion of missing values. Supervised learning: decision trees, nearest neighbours, Bayesian algorithms, comparison of classification methods and classifier ensembles. Regression: multiple linear regression, prediction trees and neural networks. Selection of independent variables through filter and wrapper methods and principal component analysis. Unsupervised learning: clustering, hierarchical clustering models and density-based models. Association rules and sparse matrices. Big data: Hadoop and MapReduce. Part B: Laboratory. The R language for Data Science, queries on multiple tables using JOIN, operators, subqueries, connection with R, and use of R packages such as sqldf, lattice, ggplot2, dplyr, party, C50, Rattle, mlr, randomForest, rpart, caret, factoextra, cluster, fpc, arules, arulesViz and RHadoop.