Small Round Blue Cell Tumor Classification using Pipeline Genetic Algorithm
Computational classification of cancerous tumors is an important research problem in machine learning. A number of approaches have been proposed by researchers to achieve accurate differentiation of samples as cancerous or non- cancerous or to differentiate different stages of a cancer. This process of computational classification has also been successfully carried out by using gene expression data as input. In this paper, we have proposed an evolutionary technique based on genetic algorithms for classification of small round blue cell tumors. This tumor occurs in four subtypes, our method has been able to differentiate these four types with 100% accuracy. The method has been compared with existing methods and has been shown to perform very well with respect to classification accuracy, recall, precision and support.
Keywords: Gene Expression Data, Small Round Blur Cell Tumor, Cancer Classification, Genetic Algorithm, Evolutionary Feature Selection