INF6028   Data Mining   (15 credits)

Description

As the volume of and types of information collected and stored in databases grows, there is a growing need to gain new insights into the data by identifying important patterns and trends. Data Mining is the process by which this is done. This module will examine the two main goals served by data mining: (i) insight (identifying patterns and trends on which to base actions), and (ii) prediction (modelling future activities or outcomes based on input data) and how algorithms are used to support these. An overview will be provided on the algorithms that underpin the most commonly used machine learning methods for building models and identifying patterns in data. Practical experience will be gained through the use of appropriate software to complete weekly tasks (i.e. KNIME).  Students will be introduced to key themes in data mining, including types of data mining problem  (e.g. classification, clustering, rule mining), common algorithms used in machine learning (e.g. SVM, decision trees, k-means, neural networks, etc.), feature selection and evaluation issues (e.g. measures and standardised benchmarks). Case studies will be used throughout the module to highlight the use of data mining methods for tackling real-world problems as well as the various ethical, social and legal issues associated with its use.

Reading List