Question

Q54) [1 Point] Which of the following learning curves represent a good linear regression model? Validation Error Training Error Error Error Error Training set size Training set size Model-1 Training set size Model-3 Model-2 A) Model-1 B) Model-2 C) Model-3 Q55) [1 Point] Maximal margin classifiers are sensitive to outliers in training data. A) True B) False Q56) [1 Point] Soft margin classifiers allows for misclassification in training data. A) True B) False Q57) [1 Point] Which of the following methods do we use to find the best fit line for data in Linear Regression? A) Least Square Error B) Maximum Likelihood C) Logarithmic Loss D) Both A and B Midterm Exam | Page 12

Answer 1

Question 55:

Answer : Option A -True, because, maximal margin classifier are very much sensitive to outliers in training data which makes them weak.

It is false because maximal classifier uses the svm (support vector machines) to find a decision boundary with maximum width.

Question 56:

Answer : Option A -True, because the idea behind it is, it allows SVM to make a some mistakes and keep margin as wide as possible so that different points can be still classified correctly.

Its false because soft margins classifiers are used for the svm methods

Question 57:

Answer : Option A - Least square Error , it is used to find the best fit line fot data linear regression which is also called as line of best fit. It is used because it minimizes the sum of the squares of reiduals in the result of every single equation and also it can be determined easily by using eyeball method.