Clustering is the process of splitting a given dataset into homogenous groups so that elements in one group are much similar to each other than the elements in different groups.
Many clustering techniques and algorithms including the most commonly used k-means have been developed and used in a variety of applications. Nevertheless, clustering is still considered as a challenge process because each individual clustering technique has its limits in some areas and none of them can adequately handle all types of clustering problems and produce reliable and meaningful results. Thus, there is still a need for exploring new approaches such as clustering ensemble that can combine the existing methods to improve clustering performance. This research aims to develop a methodology for building more effective ensembles for clustering problems.
Dr. Wenjia Wang, Geoffrey R. Guile, Jamil Al Shaqsi, Richard Harrison