Handbook

This topic aims to enable you to use Python software to develop risk prediction models using common machine algorithms. In particular, it aims to  introduce you to a range of machine learning algorithms that have been written for use with Python software, teach you how to apply these to a data set and how to assess their accuracy in risk prediction. The focus will be on risk prediction of mortality in an elderly hospital population, but examples of other applications will also be discussed. You will be able to prepare a dataset for analysis, run Python code, extract and interpret various measures of accuracy and draw valid conclusions from the dataset. Concepts such as machine learning classification systems, assessing prediction accuracy, cross-validation, data reduction techniques and the methods behind specific machine learning algorithms are covered.