A Multi-Criteria Evaluation Approach for Ensemble Machine Learning Algorithms in Precision Brain Cancer Detection

A brain tumour is a lump or collection of abnormal brain cells that has the potential to spread to nearby tissues and a precise diagnosis is necessary for effective treatment planning. The development of an algorithm for multi-criteria ensemble learning to identify brain cancer with decrease the time needed to process the classifier, which will improve accuracy. After using SMOTE and data augmentation, the model pre-processes the images. After that, a convolution neural network is used to analyze the data to extract features with Squared Hinge Loss function. The images that have been optimized are used as input for ensemble learning algorithms that are based on the gradient-boost and Ada-boost decision tree methods. The algorithm was selected with the intention of reducing processing complexity and temporal complexity without sacrificing CNN's accuracy. The system's accuracy for cancer diagnosis is discovered to be 98.45%, which is a superior value than the current approach.