Supervised Learning - Classification Week 7 Challenge

Answer 1:

Pros of SVM

1. It perform well in high-dimensional space and have excellent accuracy.
2. Requires less memory for computation.
3. Best suited for higher dimensions.
4. Perfoms well when there is large gaps between classes.

Cons of SVM :

1. It requires long training period and hence not feasible for large datasets.
2. Cannot handle overlapping classes.
3. Can not handle large noise dataset.
4. Performs bad when number of features is more than training samples.

Answer 2:

Code is provided in Week_7.ipynb file.