An Efficient Approach to Detect Liver Disorder Using Customised SVM in Comparison with Random Forest Algorithm to Measure Accuracy

Abstract

Aim: This work provides an effective approach to detect liver disorder using customised SVM algorithm in comparison with random forest algorithm to measure accuracy. Materials & Methods: For identification of effective approaches to detect liver disorder, the customised SVM algorithm is used comparatively with the random forest algorithm which is an existing algorithm. The number of samples used in each group is 20. Pretest analysis was done from clinical.cal.com by keeping G-power at 80%, threshold at 0.05% & confidential interval 95%. Result: This is an innovative liver disorder detection method and it is observed that the customised SVM algorithm is much better than random forest algorithm. Each sample gives different accuracies where customised SVM have better mean accuracy 93.80% which is greater than random forest algorithm 91.25%. The statistical results are also provided where SVM and random forest algorithms have statistical insignificance different values p=0.176 i.e., p>0.05 (Independent sample T-test). Conclusion: This study concludes that the random forest algorithm is having less accuracy value when compared to customised SVM algorithm 93.80%. By this study the effective approach for detection of liver disorder was determined.