Predictive Analytics

Catalog description

Introduces the end-to-end, data-driven statistical and predictive modeling approach with applications and case studies. Includes all the data and modeling steps in a full modeling cycle, including data ETL process, exploratory data analysis, and data cleansing for outlier imputation and data normalization. Commonly applied modeling techniques such as k-nearest neighbors, GLM, random forest, neural networks, and Naive Bayes are heavily utilized and explained using advanced visualization techniques and simplified mathematical derivations to enhance understanding. Predictive analytic modeling steps such as model training, validation, and testing are widely utilized, as are tools and languages for data processing, analysis, and modeling.