Machine Learning From Basic to Advanced

Use kernel SVM to handle non-linearly separable data with nonlinear kernels like polynomial and RBF. Compare linear versus kernel-based models and observe accuracy gains.

Learn to implement a decision tree classifier by importing libraries, loading and splitting the dataset, training the model, and evaluating predictions with confusion metrics.

Learn random forest classification using a social network dataset, then split, scale, train a random forest classifier with ten estimators and entropy criterion, and evaluate with a confusion matrix.

Apply simple linear regression to predict salary from years of experience using a single predictor, fitting the best line by minimizing the squared errors and visualizing with matplotlib.

Master multiple linear regression by modeling profit with multiple independent variables, encoding state with one-hot or label encoding, and applying backward and forward feature selection to improve accuracy.

Explains polynomial regression for non-linear relationships, contrasts it with linear regression, and shows transforming features with degree four polynomial features to improve fit on a small level and salary dataset.

Explore how decision tree regression splits non-linear data into regions, uses the region average as predictions, and implements a regression tree with sklearn, highlighting higher resolution splits for interval predictions.

Learn k-means clustering, an unsupervised algorithm that partitions data into k clusters by updating centroids and reassigning points, and use the elbow method with WCSS on income and spending data.

Explore hierarchical clustering, a bottom-up method, comparing distance metrics such as closest, farthest, average, and center-based linkage, and using a threshold to decide the optimal number of clusters.