Research Article
Classification of Microarray Data Using Kernel Fuzzy Inference System
Algorithm 1
Kernel subtractive clustering.
Input: The dataset , radius . | Output: Optimal number of clusters, their centroid and sigma (). | Compute the potential for each data point using (6). | Choose the data point whose potential value is highest as a cluster centroid. | Discard and recompute the potential value for each using (7). | If then | Accept as a cluster center and continue. | else if then | Reject and end the clustering process. | else | = shortest of the distance between and all previously found cluster centers. | if ( then | Accept as a cluster center and continue. | else | Reject and set the potential at to 0. Select the data point with the next highest potential as the new and reset. | end if | end if | |
|