Description

The classification of different types of tumors is of great importance in cancer diagnosis and its drug discovery. Cancer classification via gene expression data is known to contain the keys for solving the fundamental problems relating to the diagnosis of cancer. The recent advent of DNA microarray technology has made rapid monitoring of thousands of gene expressions possible. Detection and confirmation of cancer have been one of the most transpiring clinical applications in micro array gene expression data. However, it remains an extremely difficult job. There are many reasons for this problem to arise. We do not have many samples that can be used for training. On the other side, we have plenty number of genes. The proposed approach uses the Principle Component Analysis to reduce the high dimensionality of the microarray dataset. The signal to noise ratio defines the difference in level between the signal and the noise for the all-aml data. The k-means clustering algorithm is implemented and find the correct label for the input data. The classifiers adopted to evaluate the proposed method are support vector machine and Naive Bayes. The experiments showed promising results in gene subset selection and classification.