Blood Cancer Classification Based on Microarray Data

This study delves into the classification of microarray data from patient samples—a critical step in deciphering disease mechanisms and improving patient stratification. Motivated by the potential of microarray technology to measure thousands of gene expression levels simultaneously, this research addresses the analytical challenges posed by the high dimensionality of such data. Employing eight distinct machine learning classifiers (J48, Random Forest, Random Tree, JRip, OneR, k-NN, SMO, and BayesNet), the study compares their performance using accuracy, precision, recall, F-measure, and ROC metrics. The results shed light on each classifier’s efficacy in handling complex microarray data and inform their application in clinical genomics. Despite offering valuable insights, the findings are subject to the limitations of microarray technology, including noise and variability in gene expression measurements. Ultimately, this research provides practical recommendations for leveraging machine learning classifiers in clinical genomics, thereby contributing to the progress of personalized medicine.