Mathematical Ranking of AI Classifiers Using Confusion Matrix and Matthews Correlation Coefficient

This research primarily deals with mathematically ranking the level of accuracy various Artificial Intelligence (AI) based machine learning classifiers, using Mathews Correlation Coefficient (MCC), leveraging Confusion Matrices. A detailed Literature survey was done to gather existing knowledge. knowledge used as foundational basis further build scope this project. The classifiers were Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Convolutional Neural Network (CNN). These for Face recognition individuals. total 33 adult test subjects 10 images per subject resulting in a 330 distinct part from Labeled Faces Wild publicly available database along teachers James B. Conant High School who voluntarily participated research. Python face program wrappers associated environment built around pre-existing image data passed these wrappers. recognized faces individuals over trials wherein each trial consisted images, varying degree presented that outputs. outputs determine Matrices which turn calculate MCC scores. scores plotted, results showed SVM AI classifier had highest relative recognition, followed by KNN CNN classifiers.