Data Mining

Stage 1: Dataset selection 1 Selection of a real-world cathegorical dataset suitable for binary classiaction 2 Stage 2: Pre-processing 1 Application of pre-processing techniques: normalization, scaling, missing data imputation, etc. 2 Visualization the pre-processed dataset by means of PCA dimensionality-reductions 3 Stage 3: Implementation of thresholded Naive Bayes 1 Implementation of a binary thresholded Naive Bayes classier with Python and Pandas 4 Stage 4: Comparison and report 1 Comparison of the Naive Bayes classier against a SVM classier with average CV ROC curves