About the Institute

The aim of the Institute is to become a leading centre for research in applicable methods for extracting exploitable knowledge from huge amount of interconnected, multiscale and multivariable data and relating this knowledge to the underlying system that generates the data. The Institute brings together researchers in both mathematics and computer science. The research ranges from micro-level simulation through to the presentation of information to end users and support for decision making. The fundamental goal is to empower non-statistically trained users to understand and control their domain. It has distinctive strengths in analysis and control of networks, data visualisation, time series characterisation, inference combining simulation and observation, and social system models.

Taking such large-scale problems requires multi-disciplinarity approach: system modelling using a combination of data analysis, simulation, and physical models; semantic models and probabilistic inference to integrate multiple sources of information; and innovative methods of presenting uncertain information to end users and supporting rational decision making.

Research Methods

  • Machine learning and pattern analysis
  • Data visualisation/visual analytics
  • Statistical physics of complex systems
  • Natural-language processing
  • Visual information processing
  • Nonlinear and stochastic differential systems
  • Computational intelligence
  • Semantic web and cyber-physical social systems
  • Software engineering
  • Self-adaptive and autonomous systems
  • Cognitive science

Key Applications

  • Prediction, classification, and clustering of data.  Non-linear models and Bayesian methods.  Advanced inference, time series analysis/forecasting
  • Projection of high-dimensional data for visual interpretation
  • Routing, network analysis, emergent behaviour in nonlinear and evolving systems, optimisation and scheduling
  • Smart cities research - urban data analytics, smarter transportation, intelligent building and urban resource planning and optimisation
  • Text data (documents and social networks); topic and sentiment analysis
  • Analysis of image and video data; information retrieval
  • Physical and biological system modelling, fluid dynamics, econophysics
  • Agent-based system modelling based on micro-level modelling
  • Semantic modelling of software, hardware and physical systems and their interaction with human society.
  • Reliable software engineering, large-scale and non-SQL databases, parallel and cluster computing.
  • Software systems that react to data to improve their performance
  • Human factors in data analysis and knowledge interpretation; decision-support systems