Abstract:

In this paper we present a new heuristic that can be used to cluster ordinal data and highlight application potentials in the area of market segmentation. We prove the metric properties of the proposed heuristic which facilitates its usage as a distance measure while clustering. We highlight some peculiarities with respect to the proposed heuristic and show how the associated problems may be overcome. We also give an algorithm to cluster ordinal data using the proposed heuristic. We demonstrate the same with the help of an example. We also give a real-world application here. In the real-world application, we apply the conventional clustering method on a data set taken from preferences given with respect to attributes pertaining to the motor-vehicle industry. We also apply the proposed methodology on the same, and compare the results so obtained.