Large-Scale Data Analytics Using Ensemble Clustering

Data clustering is a highly used analysis technique in many application domains. From the end user’s perspective, the wide variety of available algorithms and their technical parameterization bring major difficulties in the determination of a user-satisfying clustering result. To overcome this issue in the context of large-scale analysis, we developed a novel feedback-driven clustering process. Aside from presenting the theoretical concepts, we also describe our developed infrastructure to efficiently handle the still increasing data volumes, within our process.