The Data Science theme investigates techniques and applications to intelligently process complex data streams. The group uses techniques from machine learning and information retrieval to better understand patterns in big data and use this information to predict likely future behaviour. In processing data, the group also considers security and privacy requirements, and data trustworthiness issues.
Much of the world’s useful data comes in semi-structured data formats, e.g. plain text written in a common style. Understanding this data and being able to search for similarities is key to making large data driven systems, and a fundamental underlying technology in the application of machine learning techniques to large and complex data sets.
Current and recent research projects
A sample of our current research projects include:
- Predictive Analysis for Tourism
- Crowdsourcing Dangerous Road Sections
- Large-scale Protein Topology Mapping for Drug Design
- Understanding the annotation process for Big Data
- Improving Predictions of Buyer Behaviour
- Trust Management in Data-Intensive Systems
- Structural Comparison of Labelled Graph Data
- Feasibility of Image Similarity Detection on the Monterey Cloud Architecture