Prediction Of Insufficient Accuracy For Kidney Cancer Renal Cell Carcinoma Using Naive Bayes In Comparison With K-Nearest Neighbour Algorithm With Improved Accuracy

Abstract

Aim: The main objective of this article is comparing K-Nearest Neighbour (KNN) with Naive Bayes to find out the best classification algorithm among these two based on accuracy. Materials and Methods: The study included 450 samples, two sets of algorithms, a G-power value of 80%, and kidney photos from a range of websites. The study's cutoff was 0.05 percent, and its mean, standard deviation, and confidence intervals were all 95%. Because the KNN algorithm has previously achieved 91.05 percent accuracy in predicting the renal failure rate for individuals, this research needs to develop a higher accuracy for kidney failure prediction using the Novel Naïve Bayes machine learning algorithm. Results: The accuracy for detecting renal failure using the Naïve Bayes algorithm was determined to be 94.23% in this study, with a significant value of 0.001 (p 0.05) for two-tailed testing. Conclusion: This study's findings demonstrate that the Naïve Bayes algorithm performs noticeably better on kidney images than the KNN algorithm.