Introduction to Machine Learning

About

Machine learning methods involve methods that deal with multivariate data, learning hidden structures and prediction. In particular within prediction methods, classification is a special case where explanatory variables are used to predict which one of a number of classes an object belongs to, e.g. is an email spam or not spam, is a person likely to vote for a particular party over the others, etc. For multivariate data you may wish to reduce the number of variables for either simpler modelling or to try to discover hidden concepts within the data.

This course will look at a subset of these types of methods including: principal component and a brief discussion of factor analysis, classification using k-nearest neighbours, classification and regression trees and discriminant analysis classification. In addition to lectures giving background on the methods and the intuition behind them to aid understanding, there will be computing sessions in R showing how to implement these methods on real data examples.

Course leader

Dr Nema Dean

Target group

Students (all groups) and researchers who have familiarity with R and basics of statistics (linear regression, normal distributions and probability).

Course aim

To provide a general and applied introduction to concepts and techniques in machine learning.

Fee info