Data Science Project: Dimension Reduction and Naive Bayes Analysis

This assignment explores two key data science techniques: Principal Component Analysis (PCA) for dimension reduction and the Naive Bayes classifier. The first part focuses on the Utilities dataset, where PCA is applied to reduce the dimensionality of corporate data from 22 US public utilities. The analysis includes evaluating the results, discussing the normalization of data, and comparing PCA with other methods. The second part involves the UniversalBank dataset and the application of a Naive Bayes classifier to predict personal loan acceptance based on customer data. This includes creating pivot tables, calculating probabilities, and computing Naive Bayes probabilities. The assignment demonstrates the application of these techniques to real-world business problems, like predicting the cost impact of deregulation and determining personal loan acceptance.