Intelligent Systems

Intelligent Systems logo

A central theme of Intelligent Systems research is the extraction of information and knowledge from data.

Within this general aim there are many sub-themes. Issues of interest may include:

  • data amounts can be very large and we consider the development and use of novel methods to make the data analysis tractable;
  • with many application domains there are disparate types of data present, which can be inter-linked in various ways - hence we consider models for data integration;
  • find those features in the data which are most salient: feature extraction;
  • find unanticipated structure present in data: unsupervised learning;
  • construct predictors for various applications.

Research includes machine learning, pattern recognition, web technologies, data mining, bioinformatics, semantic image analysis and natural intelligent systems.

These and many other themes, are common across the various application projects pursued by the group. Examples include:

Nello Cristianini has devised a Flu Detector based on Twitter stream data. In this application, feature extraction is crucial since we need to identify those single words and word-pairings which are most informative for predicting flu incidence in the population.

Colin Campbell and collaborators have devised a predictor for determining if single nucleotide variants in the human genome are functional in human disease. Many types of data are relevant to this problem so we consider data integration. There are 3.2 billion nucleotides in the genome, so the computational efficiency of the algorithm is another important issue.

In short, the group is interested in:

  • novel algorithms and methods for analysing, interpreting and using data;
  • applications to exploit data, from web technologies, bioinformatics, semantic image analysis, robotics through to natural intelligent systems;
  • the fundamental theory of learning.

Further information:

Contacts and further information