Machine Learning (ML) MCQ Quiz Hub

1. Identify the various approaches for machine learning.
concept vs classification learning
symbolic vs statistical learning
inductive vs analytical learning
all above

2. what is the function of Unsupervised Learning?
find clusters of the data and find low-dimensional representations of the data
find interesting directions in data and find novel observations/
interesting coordinates and correlations
all

3. What are the two methods used for the calibration in Supervised Learning?
platt calibration and isotonic regression
statistics and informal retrieval
both (a) and (b)
None of These

4. Which of the following are several models for feature extraction
regression
classification
both (a) and (b)
None of the above

5. In a linear regression problem, we are using R-squared to measure goodness-of-fit. We add a feature in linear regression model and retrain the same model.Which of the following option is true?
if r squared increases, this variable is significant.
if r squared decreases, this variable is not significant.
individually r squared cannot tell about variable importance. we cant say anything about it right now.
None of These

6. Which of the following assumptions do we make while deriving linear regression parameters?1. The true relationship between dependent y and predictor x is linear2. The model errors are statistically independent3. The errors are normally distributed with a 0 mean and constant standard deviation4. The predictor x is non-stochastic and is measured error-free
1,2 and 3.
1,3 and 4.
1 and 3.
all of above.

7. Generally, which of the following method(s) is used for predicting continuous dependent variable?1. Linear Regression2. Logistic Regression
1 and 2
only 1
only 2
None of These

8. Suppose you are training a linear regression model. Now consider these points.1. Overfitting is more likely if we have less data2. Overfitting is more likely when the hypothesis space is small.Which of the above statement(s) are correct?
both are false
1 is false and 2 is true
1 is true and 2 is false
both are true

9. Suppose we fit Lasso Regression to a data set, which has 100 features (X1,X2X100). Now, we rescale one of these feature by multiplying with 10 (say that feature is X1), and then refit Lasso regression with the same regularization parameter.Now, which of the following option will be correct?
it is more likely for x1 to be excluded from the model
it is more likely for x1 to be included in the model
cant say
none of these

10. Which of the following is true aboutRidge or Lasso regression methods in case of feature selection?
ridge regression uses subset selection of features .
lasso regression uses subset selection of features
both use subset selection of features
none of above

11. Which of the following statement(s) can
1 and 2
1 and 3
2 and 4
none of above

12. We can also compute the coefficient of linear regression with the help of an analytical method called Normal Equation. Which of the following is/are true about Normal Equation?1. We dont have to choose the learning rate2. It becomes slow when number of features is very large3. No need to iterate
1 and 2
1 and 3.
2 and 3
1,2 and 3.

13. Which of the following option is true regarding Regression and Correlation ?Note: y is dependent variable and x is independent variable.
the relationship is symmetric between x and y in both.
the relationship is not symmetric between x and y in both.
the relationship is not symmetric between x and y in case of correlation but in case of regression it is symmetric.
the relationship is symmetric between x and y in case of correlation but in case of regression it is not symmetric.

14. When the C parameter is set to infinite, which of the following holds true?
the optimal hyperplane if exists, will be the one that completely separates the data
the soft-margin classifier will separate the data C.
both (a) and (b)
none of the above

15. Suppose you are building a SVM model on data X. The data X can be error prone which means that you should not trust any specific data point too much. Now think that you want to build a SVM model which has quadratic kernel function of polynomial degree 2 that uses Slack variable C as one of its hyper parameter.What would happen when you use very large value of C(C->infinity)?
we can still classify data correctly for given setting of hyper parameter c
we can not classify data correctly for given setting of hyper parameter c
cant say
None of These

16. Hyperplanes are boundaries that help classify the data points.
usual
decision
parallel
None of these

17. The of the hyperplane depends upon the number of features.
dimension
classification
reduction
None of These

18. What is the purpose of performing cross- validation?
to assess the predictive performance of the models
to judge how the trained model performs outside the sample on test data
both a and b
None of These

19. Which of the following is true about Naive Bayes ?
assumes that all the features in a dataset are equally important
assumes that all the features in a dataset are independent
both a and b
none of the above option

20. Which of the following isnotsupervised learning?
pca
decision tree
naive bayesian
linerar regression

21. can be adopted when it's necessary to categorize a large amount of data with a few complete examples or when there's the need to impose some constraints to a clustering algorithm.
supervised
semi-supervised
reinforcement
clusters

22. In reinforcement learning, this feedback is
overfitting
overlearning
reward
None of the above

23. In the last decade, many researchers started training bigger and bigger models, built with several different layers that's why this approach is called
deep learning
machine learning
reinforcement learning
unsupervised learning

24. there's a growing interest in pattern recognition and associative memories whose structure and functioning are similar to what happens in the neocortex. Such an approach also allows simpler algorithms called
regression
accuracy
modelfree
scalable

25. showed better performance than other approaches, even without a context- based model
machine learning
deep learning
reinforcement learning
supervised learning

26. If Linear regression model perfectly first i.e., train error is zero, then
test error is also always zero
test error is non zero
couldnt comment on test error
test error is equal to train error

27. In syntax of linear model lm(formula,data,..), data refers to
matrix
vector
array
list

28. Suppose you are training a linear regression model. Now consider these points.1. Overfitting is more likely if we have less data2. Overfitting is more likely when the hypothesis space is small. Which of the above statement(s) are correct?
both are false
1 is false and 2 is true
1 is true and 2 is false
both are true

29. Which of the following option is true regarding Regression andCorrelation ?Note: y is dependent variable and x is independent variable.
the relationship is symmetric between x and y in both.
the relationship is not symmetric between x and y in both.
the relationship is not symmetric between x and y in case of correlation but in case of regression it is symmetric.
the relationship is symmetric between x and y in case of correlation but in case of regression it is not symmetric.

30. How many coefficients do you need to estimate in a simple linear regression model (One independent variable)?
1
2
3
4

31. For the given weather data, Calculate probability of playing
0.4
0.64
0.29
0.75

32. In reinforcement learning, this feedback is usually called as .
overfitting
overlearning
reward
None of the above

33. Reinforcement learning is particularly
the environment is not
its often very dynamic
its impossible to have a
All of the above

34. Lets say, you are working with categorical feature(s) and you have not looked at the distribution of the categorical variable in the test data.You want to apply one hot encoding (OHE) on the categorical feature(s). What challenges you may face if you have applied OHE on a categorical variable of train dataset?
all categories of categorical variable are not present in the test dataset.
frequency distribution of categories is different in train as compared to the test dataset.
train and test always have same distribution.
. both a and b

35. Which of the following sentence is FALSE regarding regression?
it relates inputs to outputs.
it is used for prediction.
it may be used for interpretation.
it discovers causal relationships.

36. Which of the following method is used to find the optimal features for cluster analysis
k-means
density-based spatial clustering
spectral clustering find clusters
All of the above

37. scikit-learn also provides functions for creatingdummy datasets from scratch:
make_classification()
make_regression()
make_blobs()
All of the above

38. which can accept a NumPy RandomState generator or an integer seed.
make_blobs
random_state
test_size
training_size

39. In many classification problems, the target dataset is made up of categorical labels which cannot immediately be processed by any algorithm. An encoding is needed and scikit-learn offers atleast valid options
1
2
3
4

40. In which of the following each categorical label is first turned into a positive integer and then transformed into a vector where only one feature is 1 while all the others are 0
labelencoder class
dictvectorizer
labelbinarizer class
featurehasher

41. is the most drastic one and should be considered only when the dataset is quite large, the number of missing features is high, and any prediction could be risky.
emoving the whole line
creating sub-model to predict those features C.
using an automatic strategy to input them according to the other known values
All of the above

42. When it is necessary to allow the model to develop a generalization ability and avoid a common problemcalled
overfitting
overlearning
classification
regression

43. What does learning exactly mean?
robots are programed sothat they can
a set of data is used todiscover the
learning is the ability tochange
it is a set of data is used todiscover the

44. In the last decade, many researchers started trainingbigger and bigger models, built with several different layers that's why this approach is called .
deep learning
machine learning
reinforcement learning
unsupervised learning

45. In reinforcement learning, this feedback is usually called as .
overfitting
overlearning
reward
None of the above

46. can be adopted when it's necessary to categorize a large amount of data with a fewcomplete examples or when there's the need to
supervised
semi- supervised
reinforcement
clusters

47. 100 people are at party. Given data gives information about how many wear pink or not, and if a man or not. Imagine a pink wearing guest leaves, what is the probability of being a man
0.4
0.2
0.6
0.45

48. For the given weather data, what is the probability that players will play if weather is sunny
0.5
0.26
0.73
0.6

49. If I am using all features of my dataset and I achieve 100% accuracy on my training set, but ~70% on validation set, what should I look out for?
underfitting B. C.
nothing, the model is perfect
overfitting
None of these

50. What is the purpose of performing cross- validation?
to assess the predictive performance of the models
to judge how the trained model performs outside the
both a and b
None of these

51. We usually use feature normalization before using the Gaussian kernel in SVM. What is true about feature normalization? 1.We do feature normalization so that new feature will dominate other 2. Some times, feature normalization is not feasible in case of categorical variables3. Feature normalization always helps when we use Gaussian kernel in SVM
1
1 and 2
1 and 3
2 and 3

52. What do you mean by generalization error in terms of the SVM?
how far the hyperplane is from the support vectors
how accurately the svm can predict outcomes for unseen data
the threshold amount of error in an svm
None of the above

53. The minimum time complexity for training an SVM is O(n2). According to this fact, what sizes of datasets are not best suited for SVMs?
large datasets
small datasets
medium sized datasets
size does not matter

54. Suppose you have trained an SVM with linear decision boundary after training SVM, you correctly infer that your SVM model is under fitting.Which of the following option would you more likely to consider iterating SVM next time?
you want to increase your data points
you want to decrease your data points
you will try to calculate more variables
you will try to reduce the features

55. For the given weather data, Calculate probability of not playing
0.4
0.64
0.36
0.5

56. Suppose, you got a situation where you find that your linear regression model is under fitting the data. In such situation which of the following options would you consider?1. I will add more variables2. I will start introducing polynomial degree variables3. I will remove some variables
1 and 2
2 and 3
1 and 3
1, 2 and 3

57. We have been given a dataset with n records in which we have input attribute as x and output attribute as y. Suppose we use a linear regression method to model this data. To test our linear regressor, we split the data in training set and test set randomly. What do you expect will happen with bias and variance as you increase the size of training data?
bias increases and variance increases
bias decreases and variance increases
bias decreases and variance decreases
bias increases and variance decreases

58. We have been given a dataset with n records in which we have input attribute as x and output attribute as y. Suppose we use a linear regression method to model this data. To test our linear regressor, we split the data in training set and test set randomly. Now we increase the training set size gradually. As the training set size increases, what do you expect will happen with the mean training error?
increase
decrease
remain constant
cant say

59. Suppose that we have N independent variables (X1,X2 Xn) and dependent variable is Y. Now Imagine that you are applying linear regression by fitting the best fit line using least square error on this data. You found that correlation coefficient for one of its variable(Say X1) with Y is -0.95.Which of the following is true for X1?
relation between the x1 and y is weak
relation between the x1 and y is strong
relation between the x1 and y is neutral
correlation cant judge the relationship

60. there's a growing interest in pattern recognition and associative memories whose structure and functioningare similar to what happens in the neocortex. Such an
regression
accuracy
modelfree
scalable