DAN 611Applied Machine Learning

Students in this course will learn about supervised and unsupervised training methods. The focus is on identifying relationships that cannot be found by basic statistics and used for example in customer satisfaction, branding, machine failure, resource allocation, fraud detection, and fraudulent activities. Techniques include Nearest Neighbors, Naive Bayes, deep learning, text mining, clustering, association rules, regularization and dimensionality reduction. The bias/variance trade-off and model selection is a focal point of the course and will be illustrated from multiple angles. Students will acquire hands-on experience on all techniques taught.