Implementation and Performance Analysis of Novel Support Vector Machine Classifier for Detecting Eye Cancer Image in comparison with Decision Tree

Abstract

The focus of the research is to identify and detect eye cancer using novel Support Vector Machine (SVM) in contrast with Decision tree (DT). Materials and Methods: Samples are analyzed using two groups with 50 eye images. The SVM algorithm was considered as g1 and g2 as a decision tree algorithm for detection of cancerous cells in the eye image. Results: SVM has achieved a notable value of 95.0% when compared with a decision tree algorithm of 87.45% with significance (p<0.05). Conclusion: The SVM algorithm has better implication accuracy of 95% to the decision tree for the analysis and detection of eye cancer.