Model based clustering for high dimensional data

Abstract

The rapid increase in the size of data sets makes clustering all the more important to capture and summarize the information; at the same time clustering is more difficult to be accomplished. If model-based clustering is applied directly to a large data set, it is too slow for practical application. A simple and common approach is to first cluster a random sample of moderate size, and then use this clustering model in this way to classify the remainder of the objects.