MULTISOLUTIONAL CLUSTERING AND QUANTIZATION ALGORITHM (MCQ)

We have developed a novel clustering and quantization algorithm that a llows the user to create multiple one-to-one correspondences between t he actual data and its transformed (clustered and quantized) values, b ased on the user's hypothesis regarding the nature of the classificati on task. The types of problems for which the algorithm can be benefici al are discussed. We report experiments employing simulated and real d ata that suggest the proposed algorithm may be useful in neural networ k analysis of various phenomena in medicine and biology.