MOA is an open source framework for Big Data stream mining. It includes a collection of machine learning algorithms (classification, regression, clustering, outlier detection, concept drift detection and recommender systems) and tools for evaluation.
since the kmeans is intended to be used on the micro clustering results from ClusTree or CluStream etc the data points to the kmeans are actually micro clusters which has an associated weight. A dominant micro cluster which has more points in it should be given more weightage when calculating new centers. Also ClusTree handles concept drift by reducing the weight of the micro-clusters. So it is important to use the weights also when finding centers.
sybernix
commented
4 years ago
since the kmeans is intended to be used on the micro clustering results from ClusTree or CluStream etc the data points to the kmeans are actually micro clusters which has an associated weight. A dominant micro cluster which has more points in it should be given more weightage when calculating new centers. Also ClusTree handles concept drift by reducing the weight of the micro-clusters. So it is important to use the weights also when finding centers.