Basics of Machine Learning

Naive Bayes, decision trees, zero-frequency, missing data, ID3 algorithm, information gain, overfitting, confidence intervals, nearest-neighbour method, Parzen windows, K-D trees, K-means, scree plot, gaussian mixtures, EM algorithm, dimensionality reduction, principal components, eigen-faces, agglomerative clustering, single-link vs. complete link, lance-williams algorithm

doubleclick for fullscreen, ← → keys forwards / rewinds 5 seconds, 1 2 ... 9 jumps to 10% 20% ... 90% of the video

Lecture 5: Naive Bayes (play all)

Lecture 6: Decision Tree (play all)

Lecture 7: Generalization and Overfitting (play all)

Lecture 9: Nearest Neighbour Method (play all)

Lecture 16: K-means Clustering (play all)

Lecture 17: Mixture Models and the EM algorithm (play all)

Lectures 18-19: Principal Component Analysis (play all)

Lecture 20: Hierarchical Agglomerative Clustering (play all)