Chapter 1 is the introduction to the course. Will do Chapter 2 at the end of the course.
Chapter 3: Linear Learning Machines
Linear Regression Classification Logistic Regression Multi-class Logistic Regression
Chapter 4: Nearest Neighbor Methods
Nearest-Neighbor Method K-Nearest Neighbor Method
Chapter 5: Classical Artificial Neural Network
Chapter 6: Support Vector Machines
Optimal Hyperplane Generalized Optimal Hyperplane Kernel Trick Usage of SVMs Statistical Learning Theory
Chapter 7: Feature Selection and Extraction for Classification
Separability of Classes Feature Extraction
Chapter 8: Decision Trees, Random Forest, and Ensemble Learning
Decision Trees Ensemble Learning
Chapter 9: Bayesian Classifier
Basic Bayesian Statistics Bayesian Rule Forms Bayesian Decision with Gaussian Distribution Markov Chain Naive Bayes Classifier Neyman-Pearson Criterion
Chapter 10: Probability Density Estimation
We need this since we don’t know the probability density. So, to use Bayesian decision, we need to estimate the probability density from the training data. There are two types of estimation, parametric, and non-parametric.
Parametric estimation: Maximum Likelihood Estimation Bayesian Estimation To do this, the data must be normal. If there is no good model, then ditch it and use non-parametric estimation instead, which includes KNN method.