Principles of Machine Learning

About This Course

Machine learning uses computers to run predictive models that learn from existing data in order to forecast future behaviors, outcomes, and trends.

In this data science course, you will be given clear explanations of machine learning theory combined with practical scenarios and hands-on experience building, validating, and deploying machine learning models. You will learn how to build and derive insights from these models using R, Python, and Azure Machine Learning.

What you'll learn

Explore classification

Regression in machine learning

How to improve supervised models

Details on non-linear modeling

Clustering

Recommender systems

The hands-on elements of this course leverage a combination of R, Python, and Microsoft Azure Machine Learning

Course Syllabus

Explore classification

Understand the operation of classifiers
Use logistic regression as a classifier
Understand the metrics used to evaluate classifiers
Lab: Classification with logistic regression taught using Azure Machine Learning

Regression in machine learning

Understand the operation of regression models
Use linear regression for prediction and forecasting
Understand the metrics used to evaluate regression models
Lab: Predicting bike demand with linear regression taught using Azure Machine Learning

How to improve supervised models

Process for feature selection
Understand the problems of over-parameterization and the curse of dimensionality
Use regularization on over-parameterized models
Methods of dimensionality reduction Apply cross validation to estimating model performance
Lab: Improving diabetes patient classification using Azure Machine Learning
Lab: Improving bike demand forecasting using Azure Machine Learning

Details on non-linear modeling

Understand how and when to use common supervised machine learning models Applying ML models to diabetes patient classification
Applying ML models to bike demand forecasting

Clustering

Understand the principles of unsupervised learning models
Correctly apply and evaluate k-means clustering models
Correctly apply and evaluate hieratical clustering model
Lab: Cluster models with AML, R and Python

Recommender systems

Understand the operation of recommenders
Understand how to evaluate recommenders
Know how to use alternative to collaborative filtering for recommendations
Lab: Creating and evaluating recommendations

Meet the instructors

Dr. Steve Elston

Steve is a big data geek and data scientist, with over two decades of experience using R and S/SPLUS for predictive analytics and machine learning. He holds a PhD degree in Geophysics from Princeton University, and has led multi-national data science teams across various companies

Cynthia Rudin

Cynthia leads the Prediction Analysis Lab at MIT, and is associated with the Computer Science and Artificial Intelligence Laboratory and the Sloan School of Management. She holds a PhD in applied and computational mathematics from Princeton University, and was previously, an associate research scientist at the Center for Computational Learning Systems at Columbia U.

Graeme Malcolm

Graeme has been a trainer, consultant, and author for longer than he cares to remember, specializing in SQL Server and the Microsoft data platform. He is a Microsoft Certified Solutions Expert for the SQL Server Data Platform and Business Intelligence. After years of working with Microsoft as a partner and vendor, he now works in the Microsoft Learning Experiences team as a senior content developer, where he plans and creates content for developers and data professionals who want to get the best out of Microsoft technologies.