COMPARATIVE PERFORMANCE OF THE FSCL NEURAL-NET AND K-MEANS ALGORITHM FOR MARKET-SEGMENTATION

Given the success of neural networks in a variety of applications in e ngineering, such as speech and image quantization, it is natural to co nsider its application to similar problems in other domains. A related problem that arises in business is market segmentation for which clus tering techniques are used. In this paper, we explore the ability of a specific neural network, namely the Frequency-Sensitive Competitive L earning Algorithm (FSCL), to cluster data for developing strategic mar keting decisions. To this end, we investigate the comparative performa nce of FSCL vis-a-vis the K-means clustering technique. A cluster anal ysis conducted on brand choice data for the coffee category revealed t hat the two methodologies resulted in widely differing cluster solutio ns. In an effort to address the dispute over the appropriate methodolo gy, a comparative performance investigation was undertaken using simul ated data with known cluster solutions in a fairly large experimental design to mimic varying data quality to reflect data collection and me asurement error. Based on the results of these studies, it is observed that a combination of the two methodologies, wherein the results of t he FSCL network are input as seeds to the K-means, seems to provide mo re managerially insightful segmentation schemes.