K-means seeding via MUS algorithm

K-means algorithm is one of the most popular procedures in data clustering. Despite its large use, one major criticism is the impact of the initial seeding on the final solution. A modified version of K-means is proposed, based on a suitable choice of the initial centers. Similarly to clustering ensemble methods, our approach takes advantage of the information contained in a co-association matrix. Such matrix is given as input for the MUS algorithm that allows to define a pivot-based initialization step. Preliminary results concerning the comparison with the classical approach are discussed.