Introduction to machine learning with R and R studio

Overview

Key facts

• 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. 

• Radboud University's Introduction to machine learning with R and R studio 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.