This module explores various machine learning techniques for predictive analytics, such as decision trees and k-nearest neighbors. Students will discover how predictive models leverage historical data and machine learning algorithms to anticipate future outcomes and trends, aiding businesses in making well-informed decisions.

Advanced Topics in Predictive Modeling cover sophisticated techniques such as ensemble methods, deep learning, and model interpretability, enabling practitioners to tackle complex data challenges and interpret the performance of their predictions. It also provides overview of methods for numeric prediction, such as regression analysis, and time series forecasting.

This module demonstrates how to organize a design sprint for ideating predictive modeling projects with team members. The process starts with brainstorming sessions focused on a customer churn problem for a company, breaking it down into clear, actionable data analytics questions using Situation-Complication-Question (SCQ) analysis. These questions are then prioritized and structured using an issue tree, ensuring a systematic approach to problem-solving and highlighting the most critical areas for data-driven insights.

This module covers the creation of outcome sketches and results mapping for predictive modeling projects. It includes industry expert interviews, dashboard mock-ups, and methods for mapping questions to analytics models. The module concludes with a final project plan review and an assignment on predictive quality control and maintenance.