This is a core course of the Graduate Diploma 5372 and Master of Science in Health Data Science 9372.

Machine learning is the procedure of applying automated algorithms to data, in order to create knowledge. Investigating data in such a manner is often referred to as data mining. In general, the objective of data mining can be classified into one of four categories; pattern analysis, visualising data, pre-processing (e.g. dimensionality reduction), or outlier detection.

Dependent of the specific objective and data, machine learning can be supervised or unsupervised, involving techniques of classification, regression, clustering, and/or association rules. Further, various methods seek to optimise the training error rate.

This course will introduce machine learning through a series of health applications. Each application will pose a set of questions to be addressed. Machine learning techniques will then be applied, generating the knowledge required to address the contextual question. Once the student has mastered the contextual application, the underlying supporting theory will be presented.

1. Distinguish a range of task specific machine learning techniques appropriate for Health Data Science.
2. Design machine learning tasks for Health Data Science scenarios.
3. Construct appropriate training and test sets for health research data.
4. Generate knowledge via the application of machine learning techniques to health data.
5. Appraise methods of training error rate optimisation.

Tutorials: 3 hours
Web-based online learning activities: 7 hours