Module aims
Traditional data mining has proved to be a successful approach to extracting new knowledge from collections of structured digital data usually stored in databases. Whereas data mining was done in the early days primarily on numerical data, the tools needed today are tools for discovering relationships between objects or segments within multimedia document components, such as classifying images based on their content, extracting patterns in sound, categorising speech and music, and recognising and tracking objects in video streams. This module will introduce the fundamental concepts of multimedia data analytics and will demonstrate how to apply proven analytics techniques to large multimedia datasets.
Module learning outcomes
Having successfully completed this module, the student will be able to:
- Describe techniques for feature extraction, selection and combination on multimedia data
- Compare and contrast models and algorithms for mining multimedia datasets
- Pre-process or clean multimedia data
- Reduce the dimensionality of multimedia data whilst conserving the relevant information
- Apply proven data science techniques for finding implicit patterns in large multimedia datasets
- Characterise the performance of various mining algorithms on multimedia data
Indicative module assessment
Continuous assessment will include weekly practical exercises, a major assignment and problem sheets based on lecture material.