Predictive Analytics and Machine Learning Module 2: Predictive Modelling for Categorical Data (synchronous E-Learning)

At the end of the 2-day module, participants will be able to:

• Understand core concepts of predictive models for classification
• Internalize a "workflow" of supervised machine learning for numerical outcomes.
• Grasp the XG Boost Algorithm
• Execute binary class and multi-class predictive models using tidymodels framework

What You'll Learn

Building on a previous module on predictive models with numerical outcome data, the module will further improve participants' understanding of classification problems (predicting categorial outcomes) in supervised machine learning, executing and interpreting predictive models with Random Forest algorithm and XG Boost (a decision-tree-based ensemble Machine Learning algorithm that uses a gradient boosting framework).

Participants will learn the differences between solving regression and classification problems in the workflow of predictive models (e.g., performance metrics). They will be shown how to run and draw insights from both binary classification and multi-class (multinomial) classification tasks.

Minimum Entry Requirement

Participants with at least a diploma qualification and completed Certified Data Analytics (R) Specialist or equivalent