Machine Learning

Description

This module provides a comprehensive grounding in Machine Learning (ML) algorithms and their application in a multi-disciplinary range of domains.

The course will cover the 4 areas of ML:

  • Supervised Learning (Classification)
  • Unsupervised Learning (Clustering)
  • Regression (Predictive modelling)
  • Dimensionality reduction

The module will also cover practical aspects including dataset pre-processing and will describe techniques for algorithms selection and parameterisation. Additionally, the process for training the model and reporting model performance will be detailed.

Learning Outcomes

  1. Navigate and utilise machine learning algorithms from state-of-the-art libraries

  2. Identify problems that can be modelled using machine learning techniques.

  3. Determine the class of problem and the model required, based upon the available data and desired outcome.

  4. Pre-process the data for use by machine learning libraries

  5. Select an appropriate algorithm and identify a training strategy and parameter set to develop a working model.

  6. Analyse and present results on the performance of the model using best-practice techniques.

Credits
10
% Coursework 100%