Rand index value on the y-axis for different datasets on the x-axis: all data try to define two clusters (‘MC1’, first column), ‘MHD25’ vs. ‘HD25’ (second column), and ‘MHD50’ vs. ‘HD50’ (third column). The colours show the different data transformations: polynomial with blue and green, standard with yellow, and logarithm with red (see the legend for details). The higher the Rand index, the more effective the clustering. The best results were obtained with standard and polynomial data for the first column (all data, distance difference).