REM-tutorial

The Reinforced EM algorithm https://epubs.siam.org/doi/10.1137/1.9781611977653.ch14 is applied to cluster the Glass Identification dataset from UCI: https://www.kaggle.com/datasets/uciml/glass?datasetId=738&sortBy=voteCount. The glass dataset comprises 9 features, where the first feature represents the refractive index, followed by weight percentages of the chemical elements Na, Mg, Al, Si, K, Ca, Ba, and Fe. Different compositions of these elements in glass result in different types of glass. The ground truth labels of the dataset consist of 7 labels. Plotted in the clustered glass dataset using the REM algorithm with GMM.