Data stream mining is a natural and necessary progression from traditional data mining. However, it presents additional challenges to batch analysis: along with strict time and memory constraints, change is a major consideration. In a dynamic data stream the underlying concepts may drift and change over time. The challenge of recognising and reacting to change in a stream is compounded by the scarcity of labels problem. This talk presents our recent work to evaluate unsupervised learning as the basis for online classiﬁcation in dynamic data streams with a scarcity of labels. A novel stream clustering algorithm based on the collective behaviour of ants, called Ant Colony Stream Clustering (ACSC), is present. Furthermore, a novel framework, Clustering and One class Classiﬁcation Ensemble Learning (COCEL), for classiﬁcation in dynamic streams with a scarcity of labels is described. The proposed framework can identify and react to change in a stream and hugely reduces the number of required labels (typically less than 0.05% of the entire stream). Finally, some conclusions will be made.
Dr. Shengxiang Yang is currently a Professor of Computational Intelligence and the Director of the Centre for Computational Intelligence, School of Computer Science and Informatics, De Montfort University, Leicester, U.K. He has over 280 publications. His current research interests include evolutionary computation, swarm intelligence, artiﬁcial neural networks, data mining and data stream analysis, and relevant real-world applications. He serves as an Associate Editor/Editorial Board Member of several international journals, such as the IEEE Transactions on Cybernetics, IEEE Transactions on Evolutionary Computation, Information Sciences, Enterprise Information Systems, and Soft Computing.