With your data cleaned and explored, you're ready to start learning from it. This module introduces machine learning: the idea that instead of writing rules by hand, you let an algorithm find patterns in examples and use them to make predictions on data it has never seen.
You'll learn the core vocabulary (features, labels, training and test sets) and the difference between supervised learning (classification and regression) and unsupervised learning (clustering and dimensionality reduction). Then you'll build real models in Python with scikit-learn, including linear and logistic regression, K-nearest neighbors, decision trees, and K-Means.
Finally, you'll meet the central tension in all of machine learning, the bias-variance tradeoff, which explains underfitting and overfitting. That sets you up for Feature Engineering & Evaluation, the next module in this course.
A clear, visual overview of what machine learning is, the main ideas behind it, and the goals of the algorithms you'll meet in the rest of this module.
A beginner-friendly course on features, classification and regression, and preparing data, with hands-on implementations of K-nearest neighbors, Naive Bayes, logistic regression, SVMs, linear regression, and clustering.
Explains bias and variance with a simple example, and why balancing the two is the key to avoiding underfitting and overfitting.
A practical, in-depth course on building models with scikit-learn: regression, classification, decision trees, unsupervised learning, and approaching a full machine learning project from scratch.