COMPUTER SCIENCE AND ENGINEERING
MACHINE LEARNING
|
Question
[CLICK ON ANY CHOICE TO KNOW THE RIGHT ANSWER]
|
|
|
SVM has inbuilt L2 regularization capabilities
|
|
|
SVM solves both classification and regression problems
|
|
|
It does not require any feature scaling
|
|
|
Choosing an appropriate Kernel function is difficult
|
Detailed explanation-1: -Yes, SVMs are sensitive to feature scaling as it takes input data to find the margins around hyperplanes and gets biased for the variance in high values.
Detailed explanation-2: -c. They find “maximum margin” classifiers. It’s a true statement that Support Vector Machines (SVMs) find “maximum margin” classifiers. SVMs are used and helpful when one wants the classifier to have the largest margin possible because they maximize the margin.
Detailed explanation-3: -SVM is not suited for finding nonlinear decision boundaries.
Detailed explanation-4: -Feature extraction is of vital importance in the implementation of classification. Proper feature extraction can help simplify the design of the SVM. On the contrary, improper feature extraction will deteriorate the performance or even lead to failure of the designed SVM.