Machine Learning (ML) MCQ Quiz Hub

1. The effectiveness of an SVM depends upon:
selection of kernel
kernel parameters
soft margin parameter c
All of the above

2. The process of forming general concept definitions from examples of concepts to belearned.
deduction
abduction
induction
conjunction

3. Computers are best at learning
facts
concepts
procedures
principles

4. Data used to build a data mining model.
validation data
training data
test data
hidden data

5. Supervised learning and unsupervised clustering both require at least one
hidden attribute.
output attribute.
input attribute
categorical attribute.

6. Supervised learning differs from unsupervised clustering in that supervised learning requires
at least one input attribute.
input attributes to be categorical
at least one output attribute
output attributes to be categorical.

7. A regression model in which more than one independent variable is used to predict the dependent variable is called
a simple linear regression model
a multiple regression models
an independent model
none of the above

8. A term used to describe the case when the independent variables in a multiple regression modelare correlated is
regression
correlation
multicollinearity
None of these

9. A multiple regression model has the form: y = 2 + 3x1 + 4x2. As x1 increases by 1 unit (holding x2constant), y will
increase by 3 units
decrease by 3 units
increase by 4 units
decrease by 4 units

10. A multiple regression model has
only one independent variable
more than one dependent variable
more than one independent variable
None of the above

11. A measure of goodness of fit for the estimated regression equation is the
multiple coefficient of determination
mean square due to error
mean square due to regression
None of these

12. The adjusted multiple coefficient of determination accounts for
the number of dependent variables in the model
the number of independent variables in the model
unusually large predictors
None of the above

13. The multiple coefficient of determination is computed by
dividing ssr by sst
dividing sst by ssr
dividing sst by sse
None of the above

14. For a multiple regression model, SST = 200 and SSE = 50. The multiple coefficient ofdetermination is
0.25
4.00
0.75
None of the above

15. A nearest neighbor approach is best used
with large-sized datasets.
when irrelevant attributes have been removed from the data.
when a generalized model of the data is desirable.
when an explanation of what has been found is of primary importance.

16. Another name for an output attribute.
predictive variable
independent variable
estimated variable
dependent variable

17. Classification problems are distinguished from estimation problems in that
classification problems require the output attribute to be numeric.
classification problems require the output attribute to be categorical.
classification problems do not allow an output attribute.
classification problems are designed to predict future outcome.

18. Which statement is true about prediction problems?
the output attribute must be categorical.
the output attribute must be numeric.
the resultant model is designed to determine future outcomes.
the resultant model is designed to classify current behavior.

19. Which of the following is a common use of unsupervised clustering?
detect outliers
determine a best set of input attributes for supervised learning
evaluate the likely performance of a supervised learner model
determine if meaningful relationships can be found in a dataset

20. The average positive difference between computed and desired outcome values.
root mean squared error
mean squared error
mean absolute error
mean positive error

21. Selecting data so as to assure that each class is properly represented in both the training andtest set.
cross validation
stratification
verification
bootstrapping

22. The standard error is defined as the square root of this computation.
the sample variance divided by the total number of sample instances.
the population variance divided by the total number of sample instances.
the sample variance divided by the sample mean.
the population variance divided by the sample mean.

23. Data used to optimize the parameter settings of a supervised learner model.
training
test
verification
validation

24. Bootstrapping allows us to
choose the same training instance several times
choose the same test set instance several times.
build models with alternative subsets of the training data several times.
test a model with alternative subsets of the test data several times.

25. The average squared difference between classifier predicted output and actual output.
mean squared error
root mean squared error
mean absolute error
mean relative error

26. Simple regression assumes a __________ relationship between the input attribute and outputattribute.
linear
quadratic
reciprocal
inverse

27. Regression trees are often used to model _______ data.
linear
nonlinear
categorical
symmetrical

28. The leaf nodes of a model tree are
averages of numeric output attribute values.
nonlinear regression equations.
linear regression equations.
sums of numeric output attribute values.

29. Logistic regression is a ________ regression technique that is used to model data having a_____outcome.
linear, numeric
linear, binary
nonlinear, numeric
nonlinear, binary

30. This technique associates a conditional probability value with each data instance.
linear regression
logistic regression
simple regression
multiple linear regression

31. This supervised learning technique can process both numeric and categorical input attributes.
linear regression
bayes classifier
logistic regression
backpropagation learning

32. With Bayes classifier, missing data items are
treated as equal compares.
treated as unequal compares.
replaced with a default value
ignored.

33. This clustering algorithm merges and splits nodes to help modify nonoptimal partitions.
agglomerative clustering B.
expectation maximization
conceptual clustering
k-means clustering

34. This clustering algorithm initially assumes that each data instance represents a single cluster.
agglomerative clustering
conceptual clustering
k-means clustering
expectation maximization

35. This unsupervised clustering algorithm terminates when mean values computed for the currentiteration of the algorithm are identical to the computed mean values for the previous iteration.
agglomerative clustering
conceptual clustering
k-means clustering
expectation maximization

36. In reinforcement learning if feedback is negative one it is defined as____.
Penalty
Overlearning
Reward
None of the above

37. According to____ , it’s a key success factor for the survival and evolution of all species.
Claude Shannons theory
Gini Index
Darwin’s theory
None of above

38. What is ‘Training set’?
Training set is used to test the accuracy of the hypotheses generated by the learner.
A set of data is used to discover the potentially predictive relationship.
Both A & B
None of the above

39. Common deep learning applications include____
Image classification,Real-time visual tracking
Autonomous car driving,Logistic optimization
Bioinformatics,Speech recognition
All of the above

40. Reinforcement learning is particularly efficient when______________.
the environment is not completely deterministic
its often very dynamic
its impossible to have a precise error measure
All of the above

41. if there is only a discrete number of possible outcomes (called categories),the process becomes a______.
Regression
Classification
Modelfree
Categories

42. Which of the following are supervised learning applications
Spam detection,Pattern detection,Natural Language Processing
Image classification,Real-time visual tracking
Autonomous car driving,Logistic optimization
Bioinformatics,Speech recognition

43. During the last few years, many ______ algorithms have been applied to deepneural networks to learn the best policy for playing Atari video games and to teach an agent how to associate the right action with an input representing the state.
Logical
Classical
Classification
None of the above

44. _____is much more difficult because it's necessary to determine a supervised strategy to train a model for each feature and, finally, to predict their value
Removing the whole line
Creating sub-model to predict those features
Using an automatic strategy to input them according to the other known values
All of the above

45. How it's possible to use a different placeholder through the parameter_______.
regression
classification
random_state
missing_values

46. If you need a more powerful scaling feature, with a superior control on outliers and the possibility to select a quantile range, there's also the class________
RobustScaler .
DictVectorizer
LabelBinarizer
FeatureHasher

47. scikit-learn also provides a class for per-sample normalization, Normalizer. It can apply________to each element of a dataset
max, l0 and l1 norms
max, l1 and l2 norms
max, l2 and l3 norms
max, l3 and l4 norms

48. There are also many univariate methods that can be used in order to select the best features according to specific criteria based on________.
F-tests and p-values
chi-square
ANOVA
All of the above

49. ____performs a PCA with non-linearly separable data sets.
SparsePCA
KernelPCA
SVD
None of the Mentioned

50. The parameter______ allows specifying the percentage of elements to put into the test/training set
. test_size
training_size
All above
None of these

51. In many classification problems, the target ______ is made up of categorical labels which cannot immediately be processed by any algorithm.
random_state
dataset
test_size
All of the above

52. If Linear regression model perfectly first i.e., train error is zero, then _____________________
Test error is also always zero
Test error is non zero
Couldn’t comment on Test error
Test error is equal to Train error

53. In syntax of linear model lm(formula,data,..), data refers to ______
Matrix
Vector
Array
List

54. Which of the following methods do we use to find the best fit line for data in Linear Regression?
Least Square Error
Maximum Likelihood
Logarithmic Loss
Both A and B

55. Which of the following is true about Residuals ?
Lower is better
Higher is better
A or B depend on the situation
None of These

56. Naive Bayes classifiers are a collection ------------------of algorithms
Classification
Clustering
Regression
All of the above

57. Naive Bayes classifiers is _______________ Learning
Supervised
Unsupervised
Both
None

58. Features being classified is __________ of each other in Naïve Bayes Classifier
Independent
Dependent
Partial Dependent
None

59. Bernoulli Naïve Bayes Classifier is ___________distribution
Continuous
Discrete
Binary
None of these

60. Multinomial Naïve Bayes Classifier is ___________distribution
Continuous
Discrete
Binary
None of These

61. Gaussian distribution when plotted, gives a bell shaped curve which is symmetric about the _______ of the feature values.
Mean B. C. Discrete D.
Variance
Discrete
Random

62. SVM is a ------------------ algorithm
Classification
Clustering
Regression
All

63. SVM is a ------------------ learning
Supervised
Unsupervised
Both
None

64. Even if there are no actual supervisors ________ learning is also based on feedback provided by the environment
Supervised
Reinforcement
Unsupervised
None of the above

65. When it is necessary to allow the model to develop a generalization ability and avoid a common problem called______.
Overfitting B. C. D. Regression
Overlearning
Classification
Regression

66. Techniques involve the usage of both labeled and unlabeled data is called___.
Supervised
Semi-supervised
unsupervised
None of the above

67. A supervised scenario is characterized by the concept of a _____.
Programmer
Teacher
Author
Farmer