Statistical Pattern Recognition (classic machine learning)

Implemented using Linear SVM(SVC) and Non-Linear SVM(RBF). ML ASSIGNMENT 2 => Q2

Build a linear SVM classifier to classify emails into spam and ham. The dataset, taken from the UCI ML repository, contains about 4600 emails labelled as spam or ham. Build a non-linear SVM classifier to classify emails and compare the performance with the linear SVM model.

The popular IRIS dataset is used for the training of linear and non-linear SVM models. The hyper-parameters are fine-tuned of the models are fine-tuned using K-Fold Cross-Validation and GridSearch to improve model performance.