Supervised Learning - Classification Week 7 Challenge

Ques1.

Pros of SVM:

Effective in high-dimensional spaces (e.g., text classification or gene expression data).

Memory efficient – uses a subset of training points (support vectors).

Versatile – supports different kernel functions (linear, polynomial, RBF, etc.).

Works well when classes are separable – both linearly and non-linearly

Cons of SVM:

Not suitable for large datasets – training time increases significantly with sample size.

Struggles with overlapping classes – sensitive to noisy data.

Requires careful parameter tuning – kernel, regularization, gamma, etc.

Not ideal for multi-label classification directly – needs One-vs-Rest strategy.

Ques2.

Attached in the file