Leukemia Classification using Cloud based Map-Reduce with K-Nearest Neighbor Classifier Framework

Microarray-based gene expression profiling has emerged as an efficient technique for classification, prognosis, diagnosis, and treatment of cancer. Cancer diagnosis is one of the most emerging clinical applications of microarray data. Frequent changes in the behavior of disease generates an enormous volume of data. Microarray data satisfies both the accuracy and velocity of big data in recent development, as it keeps changing with time. Therefore, the analysis of microarray datasets needs a large amount of expression, but only a fraction of it comprises genes that are significantly expressed. The exact identification of genes are responsible for causing cancer are essential in microarray data analysis. Most existing schemes are two-phase process such as feature selection or extraction and classification. The various statistical methods based on MapReduce are proposed for selecting relevant features. After feature selection, a MapReduce based on a K-Nearest Neighbor (MRKNN) classifier is also employed to classify microarray data and the algorithms are successfully implemented in a Hadoop framework.