Machine Learning 1: UNIT 2: Classification MCQs

 1.     By default, SGD classifier follows this strategy for multi-class classification:

Top of Form

A)   OvO strategy

B)    OvA strategy

C)    Both

D)   None

Ans: B

2.     SGD Classifiers and Linear Classifiers are strictly

A)   Binary Classifier

B)    Multiclass classifier

C)    Both

D)   None

     Ans: A

3.     The greater the value for ROC AUC, better the model:Top of Form

A)   True

B)    FalseBottom of Form

Ans: A

4.     The maximum value of the ROC AUC isTop of Form

A)    0.8

B)     0.9 

C)     1

D)    0.7Bottom of Form

Ans: C

5.     Recall can be increased by increasing the decision threshold. True or FalseTop of Form?

A)    False

B)     True

Ans: A

6.     Precision can be increased by increasing the decision threshold. True or False?

A)   True

B)    False

Ans: A

 

7.     Which of these is a good measure to decide which threshold to use?

A)   Top of Form

A)   Confusion matrix

B)    F1 score

C)   ROC curve

D)   Precision & Recall versus Threshold Curve

Ans: D

8.     SVM Classifier scales poorly with the size of training dataset. For SVM, which strategy for multi-class classifier should be applied?

Top of Form

A)   OvA

B)    OvO

Ans: B

9.     Is RandomForestClassifier a multinomial classifier?

A)   Top of Form

A)   Yes

B)    NoBottom of Form

Ans: A

10.  For RandomForestClassifier, do we need to run either OvA or OvO classifier at all?

A)   Yes

B)    No

Ans: BBottom of Form

11.  Which of these classifiers support multilabel classification?

A)   Top of Form

A)   SGD Classifier

B)    SVM Classifier

C)    KNeighbours Classifier

D)   None

Ans: C

12.  For MNIST dataset multiclass classification, SGD Classifiers trains how many binary classifiers with OvA strategy?

A)    8 

B)    10 

C)    12

D)   45

Ans: BBottom of Form

13.  For MNIST dataset multiclass classification, SGD Classifiers train how many classifiers using OvO strategy?Top of Form

A)     8

B)     10

C)     12

D)    45

Ans: DBottom of Form

14.  Classifying MNIST dataset image into [large or small, odd or even] classification lables is an example of:

A)   Binary Classification

B)    Multiclass Classification

C)    Multilabel Classification

D)   Multioutput Classification

Ans: C

15.  In multilabel classification, which of the following 'average' method for calculating F1 score calculates the unweighted mean of the f1 score of individual labels?Top of Form

A)   macro 

B)    binary 

C)    micro 

D)   weighted

Ans: A

16.  In multilabel classification, which of the following average method for calculating F1 score calculates the metrics for each label, and find their average weighted by support (the number of true instances for each label)?Top of Form

A)    Macro

B)     Binary

C)     Micro

D)    Weighted

Ans: D

17.  Which of the following methods would not be a good measure for skewed datasets. For example, 5 and Not 5 classiifer in MNIST has a skewed dataset in which there are more 'Not 5's as compared to '5's?

A)   cross_val_score using accuracy

B)    confusion matrix

C)    Cross_val_score using precision

D)   None

Ans: A

18.  Multiclass classifiers are also known as:Top of Form

A)     Mutlilabel classifiers

B)      Multinomial classifiers

C)      Multioutput classifiers

D)    None

Ans: BBottom of Form

19.  MNIST - 5 and Not 5 problem is what kind of a problem?

A)   Classification

B)    Regression

C)    Clustering

D)   None

Ans: A

20.  MNIST - 5 and Not 5 Classification is what kind of a classification problem?Top of Form

A)   Binary Classification

B)    Multi-class

C)    Multi-label

D)   Multi-output

E)Bottom of Form

Ans: A

21.  Why do we use random_state in Stochastic Gradient Descent classifier?Top of Form

A)   For generating reproducible results 

B)    To specify the training size of the batch for each iteration

C)    Both

D)   None

Ans: A

 

22.  Which of these may have to be performed before analyzing and training the dataset?Top of Form

A)    Shuffling

B)     Cross-Validation

C)     F1 Score

D)    All of the above

Ans: A

23.  For the below confusion matrix, what is the total number of training datasets?

	Not 5	5
Not 5	53272	1307
5	1077	4344

Top of Form

A)     50000

B)      60000

C)      70000

D)     80000

Ans: B

24.  For the below confusion matrix, what is the count of True Positive?

	Not 5	5
Not 5	53272	1307
5	1077	4344

A)   53272

B)    1077

C)    1307

D)   4344

     Ans: D

25.  For the below confusion matrix, what is the count of True Negatives?

	Not 5	5
Not 5	53272	1307
5	1077	4344

Top of Form

A)    53272

B)     1077

C)     1307

D)    4344

Ans: A

26.  For the below confusion matrix, what is the count of False Negatives?

	Not 5	5
Not 5	53272	1307
5	1077	4344

Top of Form

A)     53272

B)     1077

C)     1307

D)    4344

Ans: B

27.  For the below confusion matrix, what is the count of False Positive?

	Not 5	5
Not 5	53272	1307
5	1077	4344

Top of Form

A)    53272

B)     1077

C)     1307

D)    4344

Ans: C

28.  For the below confusion matrix, what is the accuracy?

	Not 5	5
Not 5	53272	1307
5	1077	4344

A)   95%

B)    90%

C)    96%

D)   98%

Ans: C

29.  For the below confusion matrix, what is the recall?

	Not 5	5
Not 5	53272	1307
5	1077	4344

A)   0.7

B)    0.8

C)    0.9

D)   0.95

Ans: B

30.  For the below confusion matrix, what is the precision?

	Not 5	5
Not 5	53272	1307
5	1077	4344

A)   0.73

B)    0.76

C)    0.78

D)   0.82

Ans: B

31.  F1 score is:

A)   absolute mean of precision and recall

B)    harmonic mean of precision and recall

C)    squared mean of precision and recall

D)   All of the above

Ans: B

32.  For the below confusion matrix, what is the F1 score?

	Not 5	5
Not 5	53272	1307
5	1077	4344

A)   0.72

B)    0.784

C)    0.82

D)   0.84

Ans: B

33.  For a model to detect videos that are unsafe for kids, we need (safe video = postive class)

A)   High precision, low recall

B)    High recall, low precision

C)    High Precision, High Recall

D)   Low Precision, Low Recall

Ans: A

34.  For a model to detect shoplifters in surveillance images, we need (shoplifter is postive class)

A)   High precision, low recall

B)    High recall, low precision

C)    High Precision, High Recall

D)   Low Precision, Low Recall

Ans: B

35.  Which of the following can be treated as a multi-output classification problem?

Top of Form

A)    Removing noise from MNIST image

B)     Classifying MNIST dataset into 0 to 9

C)     Predicted the demand for rental bikes

D)    None

Ans: A

Machine Learning 1 : UNIT 2: Classification PPTs

                                                                              Unit II

Classification

1.     Training a Binary Classifier

2.     Performance Measures

3.     Measuring Accuracy Using Cross-Validation

4.     Confusion Matrix

5.     Precision and Recall

6.     Precision/Recall Trade-off

7.     The ROC Curve

8.     Multiclass Classification

9.     Error Analysis

10.  Multilabel Classification

11.  Multi Output Classification

12.  k-NN Classifier

Machine Learning 1: UNIT 2 NOTEs: Classification Notes

 Unit II

Classification

1.     Training a Binary Classifier

2.     Performance Measures

3.     Measuring Accuracy Using Cross-Validation

4.     Confusion Matrix

5.     Precision and Recall

6.     Precision/Recall Trade-off

7.     The ROC Curve

8.     Multiclass Classification

9.     Error Analysis

10.  Multilabel Classification

11.  Multi Output Classification

12.  k-NN Classifier

Machine Learning 1: UNIT 1 The Machine Learning Landscape Questions

 UNIT-1

The Machine Learning Landscape

Short Answer Questions

-------------------------------------------------------------------------------------------------------

 

1.     Define Machine Learning.

2.     Define Unsupervised Learning.

3.     Explain Supervised Learning Algorithm.

4.     List the different Machine Learning Challenges.

5.     What is the purpose of the Machine Learning.

6.     List the different Machine Learning Applications.

7.     Define Reinforcement Learning.

8.     Define batch learning

9.     Define online learning

10.  List the different machine learning systems

11.  Machine learning types.

12.  List the supervised learning algorithms

13.  List the unsupervised learning algorithms

14.  What is overfitting

15.  What is underfitting

16.  What is poor quality data

17.  What is irrelevant feature

18.  Insufficient quantity of training data

19.  Non representative training data

20.  Testing

21.  Validation

22.  Types of machine learning systems

 

UNIT- I

Long Answer Questions

1.     Compare and contrast between supervised and unsupervised learning algorithms with examples.

2.     Define Underfitting the Training Data. How can overcome Underfitting the Training Data? Explain.

3.     Explain how Model -based Learning Algorithms differ from Instance Based Learning with example.

4.      What is Overfitting of Training Data. How can you overcome the Overfitting of Training Data?

5.     Why Use Machine Learning?

6.     Explain Batch and Online Learning with example.

7.      Explain about Instance-Based Versus Model-Based Learning .

8.     List the different types of Machine Learning Systems. Explain them.

9.     Define Machine Learning. Give different applications of Machine Learning.

10.  List the Challenges of Machine Learning. Explain them in detail.

11.   What is Underfitting of Training Data. How can you overcome the underfitting of Training Data?

12.  Discuss the following

a.      Insufficient Quantity of Training Data

b.      Nonrepresentative Training Data

c.      Poor-Quality Data

13.  Discuss the following

a.      Irrelevant Features

b.      Overfitting the Training Data

c.      Underfitting the Training Data