Football game

Machine Learning

Research Group

Learn about the work of the Group

The Group has 25 years' experience in theoretical methodology and application of flexible models to data. 

Within these areas, members of the Machine Learning Research Group conduct two types of research – applied research and theoretical research.

Applied research

As part of the Group’s applied research, members conduct work into three main strands – computer-based decision support in healthcare, sports analytics and computational marketing.

In the latest research assessment (Research Excellence Framework 2014), outputs relating to machine learning were judged to be particularly strong, with most of the papers in this topic considered to be internationally excellent. 80% of the impact case studies for this Unit of Assessment were also ranked as internationally excellent and were judged to demonstrate very considerable reach and significance.

Members from the Machine Learning Research Group collaborate with a wide range of organisations and institutions, including: Universitat Autònoma de Barcelona, ESAT, University of Pisa, Universitat Politecnica de Calatunya, University of Valencia, Whiston NHS Trust, Public Health England, Prozone Performance Lab and the Football Exchange. 

Computer-based decision support: The computer-based decision work builds on longstanding research into brain tumours that used methods of signal separation applied to spectral data, mainly from magnetic resonance measurements. This work complements research on stratification of breast cancer patients, including molecular histology with bioinformatics. Robust probabilistic models have also been developed to model hazard rates with longitudinal data.

Sports analytics: The sports analytics research is a joint venture between members from the Group and the Football Exchange in the Research Institute for Sport and Exercise Science. As part of this work, researchers examine performance analysis and performance intelligence in relation to the tactical analysis of football games and player rating models. Researchers from the Group also work with the Prozone Performance Lab when conducting this research.

Computational marketing: In computational marketing, the Machine Learning Research Group were among the first in the world to publish the outcomes of live interventional studies with personalised recommender systems in retail. This led to the development of new methodologies for targeting recommendations that were tested online in a collaborative project with Unilever Research and Development and LeShop.ch.

Theoretical research

Researchers from the Group have developed a highly scalable implementation of an efficient algorithm to define graphical models. This involves finding the correlation structure in discrete and continuous data sets.

In addition to this work, the Machine Learning Research Group also use rigorous methods to interpret complex analytical models. This work validates the analytical models, which in turn builds trust with end-users.


  • ALL
  • A
  • B
  • C
  • D
  • E
  • F
  • G
  • H
  • I
  • J
  • K
  • L
  • M
  • N
  • O
  • P
  • Q
  • R
  • S
  • T
  • U
  • V
  • W
  • X
  • Y
  • Z

Loading staff profiles…

Contact details

If you’d like to ask a question or find out more about information about this Group, please contact the team using the details below.

Contact: Paulo Lisboa

Email: P.J.Lisboa@ljmu.ac.uk 

Call: 0151 231 2225

Address:

School of Computing and Mathematical Sciences
Liverpool John Moores University
Byrom Street
Liverpool
L3 3AF