Divisive Intelligent K-Means algorithm (DiviK) for joint feature selection and clustering of heavily multidimensional data.
Method to find contrastive dimensions between experimental conditions
