Classification of Brain Tumor Images with Improved Accuracy Using KNN and Comparing with SVM

The main object of the study is to evaluate the correctness parameter of brain MRI images using KNN compared to SVM. The KNN algorithm has good accuracy in MRI images. Materials and Methods: A total of 40 samples of different MRI images has been collected from a dataset available in Github. These samples are tested for better accuracy in MRI images using the KNN algorithm and SVM algorithm with a pretest power of 80%. The values of accuracy are calculated to improve the accuracy in MRI Images. Results: The performance of the KNN algorithm is analyzed for brain MRI and it is observed that the accuracy is better for KNN (93.4%) compared to the SVM algorithm (87.1%). From this study, the results showed that the KNN algorithm gives desirable results with more accurate values for MRI images(p<0.05). Conclusion: It is found that the KNN algorithm shows a better accuracy compared to the SVM algorithm.