Comparing SVM and KNN Algorithm for High Accuracy in Brain Tumor Prediction

The main aim of this research is to compare the efficacy of Support Vector Machine (SVM) and k-Nearest Neighbors (kNN) algorithms in high-precision brain tumor prediction, solving for the issue of low precision of current methods. Two approaches were tested in this study Group 1 that uses the KNN algorithm with its hyperparameters optimized and kernel functions, and Group 2 using the SVM algorithm with distance measures and feature scaling. 95% confidence level was utilised to establish statistical significance in the analysis of 1,000 brain MRI scans. Group 1, which utilised the KNN algorithm, performed much better than Group 2, which utilized the SVM algorithm, with an accuracy of 94.56%. In comparison between the two approaches, the performance difference was statistically significant (p < 0.05), indicating that the SVM algorithm proved to be more effective in predicting brain tumors with high accuracy. The prediction of brain tumor was made more accurate and reliable by the SVM algorithm as opposed to the KNN approach, especially in situations that demand accurate and efficient classification.