Classical Machine Learning

Here you find an index for content and posts on classical machine learning models used in machine learning and data science. This page is constantly being updated as I produce more content.

Supervised Machine Learning

Understanding basic supervised machine learning models gives you a great foundation for more advanced machine learning and deep learning methods.
We distinguish between models for regression and classification.

Regression models are used to predict a continuous numerical value such as the price of a house, the blood pressure of a person, etc. They include models such as linear regression, polynomial regression, and Bayesian linear regression.

Classification models classify content into one of several classes based on the input data. For example, you can use them to predict the category of products a customer is likely interested in based on her past purchasing behavior and demographic. Examples of linear classification models are logistic regression and linear discriminant analysis.

Linear Regression

• Residuals and the Least Squares Regression Line (Simple Introduction to Linear Regression)
• Ordinary Least Squares Regression (Technical Introduction to Linear Regression)
• The Coefficient of Determination and Linear Regression Assumptions
• Linear Regression in Python (Programming Tutorial)

Logistic Regression

• The Sigmoid Function and Binary Logistic Regression
• The Softmax Function and Multinomial Logistic Regression
• Logistic Regression in Python

Support Vector Machines & Kernel Methods

• What is a Kernel?
• What is a Support Vector?
• Understanding Hinge Loss and the SVM Cost Function

Unsupervised Machine Learning

• Principal Components Analysis Explained for Dummies

Further Resources

For writing these posts I’ve relied on several textbooks, online courses, and blogs.

Here you find a comprehensive list of educational resources on machine learning and data science that I’ve used to write these posts and that I recommend for further study.