Swarm Intelligence and Data Mining
The aim of this research is to investigate the use of new swarm intelligence based techniques for data mining. This is an emerging new area of research where the data mining task is constructed as a set of biologically inspired agents. Each agent represents a simple task and the success of the method depends on the cooperative work of the agents. To our knowledge, Bonabeau et. al. developed ant-based clustering techniques successfully. Our aim is to extend this work to classification trees based on several biological models.
This research project is an important attempt to ameliorate the current applications of swarm intelligence in data mining. This should lead to novel techniques, which will enrich the integration between the newly emerging area of swarm intelligence and the somewhat mature area of data mining.
ACO, Memetic, and Immune
The aim of this research is develop new problem-solving techniques based on the emerging fields of Swarm intelligence and Immune systems. This requires scrutinising the behaviour of simple species such as Ants and Bees, and build underlying models which can be used to develop problem solving techniques. We are also interested in studying the Immune systems and explore how the defensive and adaptation mechanisms can be integrated within a swarm intelligence system. We believe that the behaviour of army ants such as Eciton hamatum, E. rapax, and E. burchelli can be used for problem solving if we integrate it with ideas from Immune systems. The project is expected to take an evolutionary approach where memetic algorithms will be the umbrella.