|Overview||Members||Projects||Publications||PhD Students||Meetings & Workshops|
Data Mining is also known as Data Analytics, Data Science and Machine Learning. Huge amount of data are being collected in every sector of life today, making data mining an extremely important research area. It is useful for data collection, data pre-processing and cleansing, knowledge discovery, event detection, future prediction and policy/strategy development.
We develop novel techniques and algorithms for all steps of data mining starting from data pre-processing and cleansing to future prediction and policy/strategy development. We work closely with industry partners to advance current practice and capability and thus solve real life problems by applying our in house and state of the art data mining techniques. We build systems for our industry partners to enhance their data analysis capability and thereby increase their business output.
Since we work on core data mining algorithms we have the ability to build custom made algorithms to cater for your specific needs and datasets. For example, in the past we developed algorithms to handle the challenges of imbalanced and cost sensitive datasets. We also developed algorithms to address the challenges of datasets with missing and incorrect data values. Our in house algorithms were used in real life to analyse and predict patient behaviour and re-admission pattern, water usage trend by farms and predict water requirement by the farms in future, analyse the behaviours of seniors in the community to improve drivers of healthy and active ageing and so on.
In addition to conventional data data mining algorithms such as classification by decision trees and forests we are also interested in privacy preserving data mining where we aim to ensure individual privacy of the data subjects while discovering general knowledge from the underlying data. For example, if general practitioners and hospitals have sensitive patient data we can develop a system that will discover knowledge from the combined datasets without disclosing individual privacy of the data subjects.
We are also interested in the visualisation of the datasets and discovered knowledge to make them more useful for end users such as doctors and nurses. We have strength on statistical analysis, image processing and machine learning for useful data analysis.
We are interested in collaboration with academics, researchers and industry partners from all over the world. We are also interested in supervising and training research higher degree students including PhD, DIT and Honours students.
The aims of the group are as follows:
The research Focus of the group is in Data Mining for Health, Agriculture and Security.