🗓️ Week 09 - Unsupervised Learning: Clustering Part 2, Anomaly detection and Dimensionality reduction

Theme: Unsupervised Learning

Welcome to the nineth week of this course!

We introduce another very useful clustering algorithm: DBSCAN.

Then, we leave clustering behind and turn our sights to another type of unsupervised learning: anomaly detection. We will explain in which cases anomaly detection applies before introducing a few techniques of anomaly detection (anomaly detection through clustering, and a density-based anomaly detection technique called Local Outlier Factor (LOF)).

We finish our exploration of unsupervised learning with yet another type of unsupervised learning: dimensionality reduction. We explain what dimensionality is about and what it is used for before introducing the most common and arguably the most well-known dimensionality reduction algorithm by far: PCA (Principal Component Analysis).

👩🏻‍🏫 Lecture Material

🎥 Looking for lecture recordings? You can only find those on Moodle, typically a day after the lecture. If you can’t find the recordings, please contact 📧 .

Download the archive containing the Quarto markdown file for this lecture (you might notice that the Markdown contains a few blanks- this is normal; the full Markdown will be released right after the lecture):