Innovative brain tumor detection technique using K-nearest neighbors and compared with support vector machine

The objective of the work is to develop an innovative brain cerebrum tumor detection technique using K-Nearest Neighbors (KNN) and Support Vector Machine (SVM) algorithms to find the accuracy rate. Materials and methods: The sample size of brain tumor detection in the Cerebrum system with improved accuracy rate was sample 2000 (Group 1 = 1000 and Group 2 = 1000). For performance evaluation of SVM, accuracy values are calculated. G power is found to be 0.8 Results: The accuracy rate of Support Vector Machine (SVM) is 97.4325% compared to the accuracy rate of K-Nearest Neighbors (KNN) 94.945%. The SVM method is significant with P = 0.000 (2 tailed) (P < 0.05) in comparison with KNN Conclusion: Support Vector Machine (SVM) Classifier provides significantly better performance compared to K-Nearest Neighbors (KNN) Classifier in finding the accuracy for analysis of brain tumor detection in Cerebrum system.