GUSTAFSON-KESSEL AND FUZZY C-MEANS ALGORITHMS BY COLON CANCER DATA IN FUZZY CLUSTERING

Özer ÖZDEMİR¹ , Aslı KAYA²

Microarray technology has made it possible to simultaneously measure the expression levels of large numbers of genes in a short time. For the analysis of microarray data, clustering techniques are frequently used. So in this study, in cases where classical clustering analysis is insufficient to analyze data, fuzzy c-means algorithm and Gustafson-Kessel algorithm, which are improved to supply with advancing alternative statistical methods, are used. Firstly, the number of the optimum cluster was decided since the number of the cluster was not known at the beginning. Then, validity indexes and elbow criterion are applied to find the optimal number of clusters for both algorithms. It is seen that for both algorithms, the elbow was situated in the c=3 position as a result of the experimental result. At the end of the study, it is graphically stated that the fuzzy c-means algorithm is getting better clusters for the colon cancer dataset.

Keywords
Fuzzy Clustering, Fuzzy C-Means Algorithm, Cancer, Gustafson-Kessel Algorithm, Colon Cancer Data,