*Prerequisite: C520  Typically offered in the Winter

This course prepares students to effectively analyze data and improve prediction performance using machine learning approaches, and distillate useful information from a large amount of data to support managerial decision-making processes. It frames machine learning domain knowledge in four layers: applications, tasks, models, and algorithms. We will apply machine learning to various business contexts, e.g., customer behavior prediction, price forecasting, fraud detection, quality control, object and facial recognition. For each application, we will identify one or several machine learning tasks, such as classification, numeric prediction, and clustering. We will then accomplish each task by using various machine learning models, including logistic regression, distance-based models, support vector machines, decision-trees, ensemble learning models (e.g., random forests and boosting), and neural networks for deep learning. Students will be introduced to various machine learning algorithm packages (e.g., scikit-learn in Python) to implement the above models for business applications, making the advanced machine learning tools and platforms (e.g., TensorFlow) accessible to business students. Students without prior experience with data programming (e.g., using Python) can still take this course but are expected to use instructor-provided resources to prepare themselves for data programming basics.