Topics covered:
k-Nearest Neighbors (kNN):
1. Classifying a point based on its k-nearest neighbors, for different values of k
2. Drawing decision boundaries for 1NN
3. Choosing k (overfitting, underfitting, cross-validation)
4. Transforming data, and different metrics for measuring distance
5. Working backwards to find training data
Identification Trees (ID Trees, aka decision trees or classification trees)
1. Calculating disorder
2. Constructing a greedy disorder-minimizing ID tree
3. Classifying unknown test points
4. Drawing decision boundaries for numeric ID trees
Example problems:
- 2008 final (rocks)
- 2014 Quiz 2 (trees)