Efficient Fuzzy Techniques for Medical Data Clustering

Researchers in data mining field aims to design efficient and scalable algorithms when clustering high dimensional data. Nowadays, data have become more massive and complex resulting in big data and high dimension datasets that require efficient methods to analyze them. For that reason, as the data become larger and more diverse, maintaining cluster quality and speed is necessary when adapting an existing algorithms. Various clustering methods can deal with low dimension datasets yet as the dimensionality increases these methods tend to fail. There are number of fuzzy clustering techniques that are available for clustering high dimension data efficiently. Fuzzy C-mean, FCM, and its extension Gustafson Kessel, GK, are two examples for clustering data of high dimensions. In this paper, FCM and GK were used for medical data clustering in particular, Wisconsin Breast Cancer, WBC, dataset obtained from UCI repository were used in the experiments. The two algorithms were compared to other work. The results of the experiments showed a high performance in accuracy for both FCM and GK.