Supervised machine learning algorithms are created from basic principles to study and classify data from Pattern Recognition and Neural Networks by b. D. Ripley (https://www.stats.ox.ac.uk/pub/PRNN/).
A normalized, synthetic dataset is obtained from the aformentioned database to study, analyse and classify the data. The data has two features, X and Y. A label (0 or 1) is associated with every data point classifying it to either of the two classes. The objective of this project is to statistically model the data and derive a decision rule based on which each unknown, testing data will be assigned to the most fitting class. Maximum Likelyhood Estimator is being used to claculate the Gaussian mean and covariance matrix. The performance of each cases of Bayesian classifiers will be studied and estimated. A detailed report is provided for further information.