K-Boost: a Scalable Algorithm for High-Quality Clustering of Microarray Gene Expression Data

In this paper we propose K-Boost, a novel clustering algorithm basedon a combination of the Furthest-Point-First (FPF) heuristic for solving themetric k-center problem, a {\em stability-based} method for determining thenumber of clusters, and a k-means-like cluster refinement. Experiments showthat \textit{K-Boost} exhibits a good quality/running time tradeoff that makesit ideal for large data sets, with quality measured by several internal andexternal criteria.