RSS04.19 Introduction to machine learning with R and R studio

Due to the further digitization of society in general, digital data have become available in large quantities (Big Data). Because Big Data (many observations) is also often high-dimensionally (many features, variables), it allows us to use machine learning techniques to make predictions and classify data into groups and uncover hidden patterns in data.

Machine learning consists of a collection of quantitative techniques to make predictions and to classify data into categories or uncover groups of observations. The course will focus on popular machine learning techniques. The first category of techniques are supervised techniques (identifying an explicit outcome variable), such as regression models, K-nearest neighbor, decision trees, support vector machines and deep learning. These models aim to make predictions based on a set of input variables. Another class of machine learning techniques are unsupervised techniques. These techniques (a) reduce many variables (columnwise) into a limited number of dimensions (PCA, autoencoders), or (b) reduce many observations (rowwise) into homogeneous groups (clustering techniques). Further, topic modeling of textual (unstructured) data will be discussed.

In this course we will use R and Rstudio to use machine learning techniques to analyze data. R is free and open source data analysis software, widely used in the academic and business world.