Improve of Fuzzy C-Means Clustering in Feature Extraction Phase on the Breast Cancer Analysis

Cancer analysis is one of the broadly advised acreage in the healthcare domain. The objective of the breast cancer problem is to predict the property of a new tumor (malignant or benign). The existing method hybridizes K-means algorithm and SVM (K-SVM) for breast cancer diagnosis. To reduce the high dimensionality of feature space, it extracts abstract malignant and benign tumor patterns separately before the original data is trained to obtain the classifier. In order to improve the quality of prediction, Fuzzy c-means clustering is hybridizes with SVM (F-SVM). An improved fuzzy c-means algorithm is proposed to deal with the cancer data. The proposed algorithm improves the traditional Fuzzy c-means algorithm in terms of selecting the initial cluster centre. Thereby, it avoids the basic drawback of Fuzzy C-means and improves the quality of prediction over k-means algorithm. It helps to predict the benign and malignant tumors. Based on the derived membership, each tumor pattern is considered as a model. Further support vector machine (SVM) technique is used to obtain the new classifier to discriminate the data.