Data Mining and Machine Learning Group

This is the home page of the Machine Learning Group (formerly the Data Mining and Machine Learning Group), part of the Artificial Intelligence section in the Department of Computer Science at the University of Liverpool.

The aim of the group is to investigate automated learning in intelligent systems by developing computational models and algorithms. Our expertise spans the core areas of data mining, pattern recognition, prediction, reinforcement and multi-agent learning, and learning to behave.

The group is led by Professor Danushka Bollegala

About the Machine Learning Group

    The theoretical group interests and capabilities include:
  • adaptive signal processing
  • artificial neural networks
  • big data analytics
  • combinatorial data analysis
  • data fusion
  • data visualisation
  • decision theory
  • deep learning
  • distributed systems
  • feature representation learning
  • image, text and graph mining
  • kernel-based learning
  • mathematical modelling and optimisation
  • markov decision processes and generalisations
  • multi-agent learning
  • multi-view learning
  • multi-way modelling and tensor decompositions
  • natural language processing
  • reinforcement learning
  • scalable learning and mining algorithms
  • sequential decision making
  • statistical pattern recognition
  • supervised and un/semi-supervised learning
  • tracking
We have strong expertise in the application of such methodologies to engineering, medicine and science. Examples include:
  • analysis of software on modern multicore systems
  • auctions
  • bioinformatics
  • biomechanics
  • biometric technologies
  • biomedical diagnostics
  • credit card fraud detection
  • e-commerce
  • health data analytics
  • industrial monitoring and control
  • interbank markets
  • market behaviour analysis
  • medical image analysis
  • multi-robot systems
  • nuclear forensics and security
  • psychology
  • security games
  • social media analysis
  • systems biology
  • wireless sensor networks