Supervised Learning - Classification Week 7 Challenge

Q1. Answer

Pros- 1)it works really well with a clear margin of separation.

         2)it is effective in high dimensional spaces.

         3) it is effective in cases where no of dimensions is greater than the number of samples.

         4) it is uses a subset of training points in the dicision function, so its also memory efficient.

Cons- 1)it doesn't perform well when we have large data set beacuse required training time is higher.

          2)it also doen't perform very well , when the data set has more noise. Ex. target classes are overlapping

          3)SVM dosen't directly provide probability estimates, these are calculated using an expensive five-fold-cross

             valiadation.it is included in the related SVC method of python scikit-learn library.