3.2 Machine Learning Basics

Section 3.2 introduces fundamental concepts that form the foundation of machine learning. It consists of four sub-sections:

• 3.2.1 Types of Machine Learning: This part explains the four main categories of machine learning: supervised learning, unsupervised learning, semi-supervised learning, and reinforcement learning. It describes how each type learns from data and the typical tasks they are used for.

• 3.2.2 Model Training and Evaluation: This section covers the typical workflow of training a machine learning model. It includes data splitting (training, validation, and test sets), algorithm selection, loss functions, optimization techniques, and evaluation metrics like accuracy, precision, recall, and F1-score.

• 3.2.3 Overfitting and Underfitting: This part discusses two critical issues that affect model performance. Overfitting occurs when a model learns noise from the training data, while underfitting happens when a model cannot capture the underlying pattern. The section also introduces ways to detect and mitigate both issues.

• 3.2.4 Feature Engineering and Selection: This sub-section explains the importance of selecting the right input features and transforming raw data into meaningful representations. Techniques such as normalization, encoding categorical variables, and feature selection methods (e.g., filter, wrapper, and embedded methods) are highlighted.