MFE-HC: The maximizing feature elimination technique based hybrid classifier for cancer molecular pattern discovery

The most important application of Microarray for gene expression analysis is used to discover or classify the unknown tissue samples with the help of known tissue samples. Several Data Mining Classifiers have been proposed recently to predict/identify the cancer patterns. In this research work, we have focused and studied a few Classification Techniques such as Support Vector Machine (SVM), Nearest Neighbor Classifier (k-NN), ICS4, Non-Parallel Plane Proximal Classifier (NPPC), NPPC-SVM, and Margin-based Feature Elimination-SVM (MFE-SVM). The performances of these classifiers have been analyzed in terms of Threshold Level, Execution Time, Memory Usage and Memory Utilization. From our experimental results, we revealed that the Threshold level and Execution Time to predict the Cancer Patterns are different for different Classifiers. Our experimental results established that among the above identified classifiers, the k-NN classifier achieves less Threshold to predict the cancer pattern, but however it consumes more execution time, whereas the MFE-SVM achieves less execution time to predict the cancer pattern, but it still needs more threshold to predict the Pattern. That is to find the best single classifier in terms of Threshold and Execution Time is still complicated. To address this major issue, we have proposed an efficient Classifier called Maximizing Feature Elimination Technique based Hybrid Classifier (MFE-HC), which is the combination of both k-NN and SVM classifiers. From the results, it is established that our proposed work performs better than both the k-NN and MFE-SVM Classifiers interms of Threshold and Execution Time.