Abstract
Pattern analysis of mass spectra obtained from blood samples, has attracted the attention for early detection of cancer. In this paper, we present an unsupervised kernel based fuzzy c-means algorithm (KFCM), which is realized by modifying original Euclidean distance in classical fuzzy clustering algorithm (FCM) by kernel-induced distance metric. Our analysis on mass spectrometry dataset, shows that KFCM has better clustering performance and is more robust to noise than FCM. We evaluated the performance of our classification methods with some popular classification techniques like SVM, PCA, LDA/QDA and randomforests.