Machine Learning Online Test | Quiz

1. Mean normalization can be used to simplify gradient descent for multivariate linear regression.

True

False

Question 1 of 50

2. Which of the following is expressed by the given equation Y = β₀ + β^TX + Ɛ which shows a real-valued dependent variable Y is modeled as function of multiple independent variables X₁, X₂, …, X_p ≡ X plus noise?

Multiple Regression

Linear Regression

Multi classification

Binary classification

Question 2 of 50

3. Which of the following statements is not true about reducing a real valued feature problem into binary feature?

Once the binary features are constructed the ID3 algorithm can be easily applied

After ID3 applied it is easy to verify that there exists a decision tree with different training error

It utilizes the threshold based splitting rules

After ID3 applied it is easy to verify that there exists a decision tree with same number of nodes

Question 3 of 50

4. It is possible that in the output, set S contains only phi.

False

True

Question 4 of 50

5. The target variable is represented along ____________

Depends on the dataset

Y axis

Either Y-axis or X-axis, it doesn’t matter

X axis

Question 5 of 50

6. Which of the following statements is not true about the picture shown below?

The data are not completely separable

Slack variables can be introduced to the objective function to allow error in the misclassification

Soft SVM can be used here to classify the data correctly

Hard SVM can be used here to classify the data flawlessly

Question 6 of 50

7. Which of the following objective functions is not solved by Subgradient?

L1 norm

Perceptron loss

Hinge loss

TanH function

Question 7 of 50

8. Multivariate linear regression belongs to which category?

Unsupervised learning

Neither supervised nor unsupervised learning

Supervised learning

Both supervised and unsupervised learning

Question 8 of 50

9. Spectrum Kernel count the number of substrings in common.

True

False

Question 9 of 50

10. (x⁽¹⁾, y⁽¹⁾) = 1, 1.5, (x⁽²⁾, y⁽²⁾) = 2, 3, (x⁽³⁾, y⁽³⁾) = 3, 4.5. Hypothesis: h(x) = t₁x, where t₁ = 1.5. How much error is obtained?

22.5

4.5

1.5

Question 10 of 50

11. The goal of a support vector machine is to find the optimal separating hyperplane which minimizes the margin of the training data.

False

True

Question 11 of 50

12. Consider the dataset given below where T and F represent True and False respectively. What is the entropy H (Rain)?

Temperature	Cloud	Rain
Low	T	T
Low	T	T
Medium	T	F
Medium	T	T
High	T	F
High	F	F

0.2

0.5

0.6

Question 12 of 50

13. What is the entropy at P = 0.5 from the given figure?

0.5

-0.5

-1

Question 13 of 50

14. The optimum separation hyperplane (OSH) is the linear classifier with the minimum margin.

False

True

Question 14 of 50

15. Who invented BFGS?

Cauchy

Bayes

Broyden, Fletcher, Goldfarb and Shannon

Quinlan

Question 15 of 50

16. Which of the following is represented by the below figure?

Random forest

Classification tree

Regression tree

Support vector machine

Question 16 of 50

17. Support vector machine is a generative classifier.

True

False

Question 17 of 50

18. Ax = b => [4 2, 2 3][x₁, x₂] = [2, 2]. Let x₀, the initial guess be [1, 1]. What is the residual vector?

[-4, -3]

[4, 3]

[4, -3]

[-4, 3]

Question 18 of 50

19. Identify the parametric machine learning algorithm.

Naïve Bayes

CNN (Convolutional neural network)

SVM (Support vector machines)

KNN (K-Nearest Neighbours)

Question 19 of 50

20. More sophisticated averaging schemes can improve the convergence speed in the case of strongly convex functions.

False

True

Question 20 of 50

21. Ensembles tend to yield better results when there is a significant diversity among the models.

False

True

Question 21 of 50

22. When does the hypothesis change in the Find-S algorithm, while iteration?

Amy negative example is encountered

Any example (positive or negative) is encountered

Any positive example is encountered

Positive Example inconsistent with the hypothesis is encountered

Question 22 of 50

23. Which of the following statements is not a step in Minimum error pruning?

Calculate the expected error rate for that node if subtree is not pruned

If pruning the node leads to greater expected error rate, then keep the subtree

At each non leaf node in the tree, calculate expected error rate if that subtree is pruned

If pruning the node leads to smaller expected error rate, then don’t prune it

Question 23 of 50

24. When was logistic regression invented?

1958

1960

1957

1959

Question 24 of 50

25. Consider there are 7 weak learners, out of which 4 learners are voted as FAKE for a social media account and 3 learners are voted as REAL. What will be the final prediction for the account if we are using a majority voting method?

REAL

Error

Undefined

FAKE

Question 25 of 50

26. Setting large values of K in kNN is computationally inexpensive.

True

False

Question 26 of 50

27. The output in a logistic regression problem is yes (equivalent to 1 or true). What is its possible value?

Depends on the algorithm’s threshold value

Greater than 0.5

Equal to 1

Greater than 0.6

Question 27 of 50

28. Which of the following statements is false about the base - learners?

The base - learners are chosen for their accuracy

The base - learners are chosen for their simplicity

Base - learners do not require them to be very accurate individually

The base - learners has to be diverse

Question 28 of 50

29. What is assumed while using empirical risk minimization with inductive bias?

The learner has some knowledge about labeling function

Reduction of overfitting may lead to underfitting

The learner has some prior knowledge about training data

No assumptions are made

Question 29 of 50

30. Which of the following ECOC designs uses n = (Nc−1).T dichotomizers, where T stands for the number of binary tree structures to be embedded?

DECOC

Forest-ECOC

One-versus-all

One-versus-one

Question 30 of 50

31. How is the version space represented?

Most general and least general members

Most general members

Arbitrary members chosen form hypothesis space

Least general members

Question 31 of 50

32. Which of the following statements is not true about the C parameter in SVM?

Small values of C give solutions with bigger margin

Small values of C give solutions with less classification errors

Large values of C give solutions with less misclassification errors

Large values of C give solutions with smaller margin

Question 32 of 50

33. What is the advantage of the list-then-eliminate algorithm?

Computation is less

Time-effective

Contains all hypotheses consistent with observed data

Overfitting never occurs

Question 33 of 50

34. A learner can be deemed consistent if it produces a hypothesis that perfectly fits the __________

training data

test data

overall dataset

cross-validation data

Question 34 of 50

35. Ax = b => [3 2, 2 3][x₁, x₂] = [8, 6]. Let x₀, the initial guess be [2, 1]. What is the residual vector?

[1, 0]

[0, 1]

[0, -1]

[-1, 0]

Question 35 of 50

36. Which of the following statements is not true about Subgradient method?

It has no good stopping criterion

It often leads to very simple algorithms

Its convergence can be very fast

It handles general non-differentiable convex problem

Question 36 of 50

37. Error is defined over the _____________

test Set

cross-validation set

training set

domain set

Question 37 of 50

38. The Subgradient is a descent method.

False

True

Question 38 of 50

39. In Classification trees the value obtained by terminal node in the training data is the mode of observations falling in that region.

True

False

Question 39 of 50

40. Error strongly depends on distribution D.

False

True

Question 40 of 50

41. Stochastic gradient descent has the possibility of escaping from local minima.

False

True

Question 41 of 50

42. In error-correcting output codes (ECOC), the main classification task is defined in terms of a number of subtasks that are implemented by the base-learners.

False

True

Question 42 of 50

43. Assume that we are training a boosting classifier using decision stumps on the given dataset. Then which of the given examples will have their weights increased at the end of the first iteration?

Both

Circle

No increment in weight

Square

Question 43 of 50

44. From the below table where the target is to predict play or not (Yes or No) based on weather condition, what is the Gini index for Climate = Sunny?

Day	Climate	Temperature	Wind	Decision
1	Sunny	Cool	Strong	Yes
2	Sunny	Hot	Weak	No
3	Rainy	Medium	Weak	Yes
4	Winter	Cool	Weak	Yes
5	Rainy	Cool	Strong	No
6	Winter	Cool	Strong	No
7	Sunny	Hot	Strong	No

0.47

0.45

0.49

0.43

Question 44 of 50

45. Who invented logistic regression?

Vapnik

Ross Quinlan

DR Cox

Chervonenkis

Question 45 of 50

46. Which of the following is an example of stacking?

AdaBoost

Voting Classifier

Bagged Decision Trees

Random Forest

Question 46 of 50

47. Which of the following statements is not true about Lagrange multipliers?

Once the stationary points have been identified from the first-order necessary conditions, it determines the maxima, minima or saddle points

The basic idea behind this is to convert a constrained problem into a form such that the derivative test of an unconstrained problem can still be applied

The Hessian matrix determines the maxima, minima, or saddle points before the stationary points have been identified

It is done by converting a constrained problem to an equivalent unconstrained problem with the help of certain unspecified parameters

Question 47 of 50

48. In a linear regression problem, h(x) is the predicted value of the target variable, y is the actual value of the target variable, m is the number of training examples. What do we try to minimize?

(h(x) – y)² / 2*m

(h(x) – y) / 2*m

(h(x) – y) / m

(y – h(x))

Question 48 of 50

49. Which of the following statements is not true about the Regression trees?

Algorithm starts from the leftmost leave nodes

The algorithm selects the split that minimizes the sum of the squared deviations from the mean in the two separate partitions

The algorithm allocates the data into the first two partitions or branches, using every possible binary split on every field

Initially, all records in the training set (pre-classified records that are used to determine the structure of the tree) are grouped into the same partition

Question 49 of 50

50. Given the entropy for a split, E_split = 0.39 and the entropy before the split, E_before = 1. What is the Information Gain for the split?

0.39

2.56

0.61

Question 50 of 50

Machine Learning Certification Test

Machine Learning Certification Test is a free certification exam. However, you need to score an A grade in each of the "Certification Level Tests 1 to 10" to be eligible to take part in this certification test. So, take all the "10 Tests" starting from Certification Level 1 upto Level 10, before taking the final Certification test.

Level 1 Test Certification Test

Level 1 to 10 Tests:
Total Questions: 25, Total Time: 30 min, Correct Answer: 2 points, Wrong Answer: -1 point

Certification Test:
Total Questions: 50, Total Time: 1 hour, Correct Answer: 2 points, Wrong Answer: -1 point

Machine Learning Internship Test

If you scored either Grade A* or Grade A in our Machine Learning Internship Test, then you can apply for Internship at Sanfoundry in Machine Learning.

Machine Learning Internship Test

Total Questions: 50, Total Time: 1 hour, Correct Answer: 2 points, Wrong Answer: -1 point

Machine Learning Job Test

It is designed to test and improve your skills for a successful career, as well as to apply for jobs.

Machine Learning Job Test

Total Questions: 50, Total Time: 1 hour, Correct Answer: 2 points, Wrong Answer: -1 point

Note: Before you get started on these series of online tests, you should practice our collection of 1000 MCQs on Machine Learning .

Sanfoundry Scoring & Grading System

Sanfoundry tests and quizzes are designed to provide a real-time online exam experience. Here is what you need to know about them.

Scoring System: You get 2 points for each correct answer but lose 1 point for every wrong answer.
Grading System: Your grade depends on your final score and can be one of the following:

Grade A* - Genius (100%)
Grade A - Excellent (80% to 99%)
Grade B - Good (60% to 80%)
Grade C - Average (40% to 60%)
Grade D - Poor (0% to 40%)