Analysis and Comparison of SVM-RBF Algorithms for Colorectal Cancer Detection over Convolutional Neural Networks with Improved Accuracy.
DOI:
https://doi.org/10.47750/pnr.2022.13.S04.011Keywords:
SVM-RBF Algorithm, CNN Algorithm, Cancer Detection, Innovative colorectal cancer detection, Machine learningAbstract
Aim: This research is to study the comparison and analysis of (Support Vector Machine - Radial Basis Function) SVM-RBF algorithm for colorectal cancer detection over Convolutional Neural Network (CNN).
Materials and Methods: The study was conducted using SVM-RBF and Convolutional Neural Network algorithms to analyze and compare colorectal cancer detection. results were computed using matlab software. For each group, twenty samples were used to compare accuracy of SVM-RBF algorithm and CNN algorithm. Sample size was calculated by maintaining G-power 80%, α=0.05, confidence interval 95%.
Results: Study performance exhibits colorectal cancer detection accuracy over SVM-RBF algorithm and detection from CNN with improved accuracy. The mean value of SVM-RBF algorithm is 91% and CNN algorithm is 89%. It is observed that SVM-RBF algorithm performed better than CNN algorithm (p=0.013) by performing an independent sample t-test. Conclusion: SVM-RBF algorithm has significantly greater accuracy in predicting colorectal cancer than CNN algorithm.
