What will you in the Practical Machine Learning Course
Learn the full workflow of predictive modeling, from data preprocessing to final model evaluation.
Understand critical concepts like overfitting, cross-validation, and out-of-sample error.
Apply machine learning algorithms including decision trees, random forests, and regularized regression.
Use the caret package in R for building, training, and validating models.
Learn how to combine multiple models and use unsupervised methods for prediction.
Program Overview
1. Introduction to Prediction and Study Design
Duration: ~2 hours
Overview of predictive modeling concepts.
Introduction to training/test sets, error types, and cross-validation.
Basics of designing a machine learning study.
2. Machine Learning with caret in R
Duration: ~2 hours
Working with the caret package to train and evaluate models.
Data splitting, preprocessing (scaling, PCA), and model tuning.
Plotting predictors and using caret’s modeling workflow.
3. Decision Trees, Random Forests, and Boosting
Duration: ~1.5 hours
Understanding and implementing tree-based models.
Random forests and boosting explained with practical examples.
Introduction to model-based prediction approaches.
4. Regularization and Model Combination
Duration: ~2 hours
Concepts of regularized regression (e.g., ridge and lasso).
Combining multiple predictive models to improve accuracy.
Brief introduction to forecasting and unsupervised prediction.
Final assignment: build a working prediction model and submit for peer review.
Get certificate
Job Outlook
Data Scientists: Gain practical experience in model building and validation.
Machine Learning Engineers: Learn foundational methods for scalable ML applications.
Business Analysts: Use data-driven techniques to support strategic decision-making.
Academic Researchers: Apply machine learning methods to experimental or observational data.
R Programmers: Advance your skills in applying machine learning using the caret package.
Specification: Practical Machine Learning
|