Home Banner

Informatics research centre

Cost-Sensitive Data Mining

This research area, led by Professor Sunil Vadera is developing new cost-sensitive data mining methods. It aims to develop new approaches that take account of the cost of mis-classifying examples and the cost of acquiring the information.

Such cost-sensitive algorithms have potential for application in medical diagnosis, credit scoring, and energy minimization.

Examples of research include projects with British Gas to apply data mining to SMART meters data to understand energy consumption behaviour and an FP7 project called SEEDS: Self-Learning Energy Efficient Buildings and Open Spaces.