Prépublication
Unsupervised clustering techniques are a valuable source of information for determining how to divide a dataset into subgroups. We present a comprehensive analysis of the quality of these algorithms by defining a clustering fuzziness metric. A statistical test and cluster probabilities corrections a...
