Classification and Clustering

1. Instead of representing knowledge in a relatively declarative, static way (as a bunch of things that are true), rule-based system represent knowledge in terms of___________ that tell you what you should do or what you could conclude in different situations.
a) Raw Text
b) A bunch of rules
c) Summarized Text
d) Collection of various Texts
Answer: b
Explanation: None.

2. A rule-based system consists of a bunch of IF-THEN rules.
a) True
b) False
Answer: a
Explanation: None.

3. In a backward chaining system you start with the initial facts, and keep using the rules to draw new conclusions (or take certain actions) given those facts.
a) True
b) False
Answer: b
Explanation: Refer the definition of backward chaining.

4. In a backward chaining system, you start with some hypothesis (or goal) you are trying to prove, and keep looking for rules that would allow you to conclude that hypothesis, perhaps setting new sub-goals to prove as you go.
a) True
b) False
Answer: a
Explanation: None.

5. Forward chaining systems are _____________ where as backward chaining systems are ___________
a) Goal-driven, goal-driven
b) Goal-driven, data-driven
c) Data-driven, goal-driven
d) Data-driven, data-driven
Answer: c
Explanation: None.

6. A Horn clause is a clause with _______ positive literal.
a) At least one
b) At most one
c) None
d) All
Answer: b
Explanation: Refer to the definition of Horn Clauses.

7. ___________ trees can be used to infer in Horn clause systems.
a) Min/Max Tree
b) And/Or Trees
c) Minimum Spanning Trees
d) Binary Search Trees
Answer: b
Explanation: Take the analogy using min/max trees in game theory.

8. An expert system is a computer program that contains some of the subject-specific knowledge of one or more human experts.
a) True
b) False
Answer: a
Explanation: None.

9. A knowledge engineer has the job of extracting knowledge from an expert and building the expert system knowledge base.
a) True
b) False
Answer: a
Explanation: None.

10. What is needed to make probabilistic systems feasible in the world?
a) Reliability
b) Crucial robustness
c) Feasibility
d) None of the above
answer : b
Explanation: On a model-based knowledge provides the crucial robustness needed to make probabilistic system feasible in the real world.

11. How many terms are required for building a bayes model?
a) 1
b) 2
c) 3
d) 4
Answer: c
Explanation: The three required terms are a conditional probability and two unconditional probability.

12. What is needed to make probabilistic systems feasible in the world?
a) Reliability
b) Crucial robustness
c) Feasibility
d) None of the mentioned
Answer: b
Explanation: On a model-based knowledge provides the crucial robustness needed to make probabilistic system feasible in the real world.

13. Where does the bayes rule can be used?
a) Solving queries
b) Increasing complexity
c) Decreasing complexity
d) Answering probabilistic query
Answer: d
Explanation: Bayes rule can be used to answer the probabilistic queries conditioned on one piece of evidence.

14. What does the bayesian network provides?
a) Complete description of the domain
b) Partial description of the domain
c) Complete description of the problem
d) None of the mentioned
Answer: a
Explanation: A Bayesian network provides a complete description of the domain.

15. How the entries in the full joint probability distribution can be calculated?
a) Using variables
b) Using information
c) Both Using variables & information
d) None of the mentioned
Answer: b
Explanation: Every entry in the full joint probability distribution can be calculated from the information in the network.

16. How the bayesian network can be used to answer any query?
a) Full distribution
b) Joint distribution
c) Partial distribution
d) All of the mentioned
Answer: b
Explanation: If a bayesian network is a representation of the joint distribution, then it can solve any query, by summing all the relevant joint entries.

17. How the compactness of the bayesian network can be described?
a) Locally structured
b) Fully structured
c) Partial structure
d) All of the mentioned
Answer: a
Explanation: The compactness of the bayesian network is an example of a very general property of a locally structured system.

18. To which does the local structure is associated?
a) Hybrid
b) Dependant
c) Linear
d) None of the mentioned
Answer: c
Explanation: Local structure is usually associated with linear rather than exponential growth in complexity.

19. Which condition is used to influence a variable directly by all the others?
a) Partially connected
b) Fully connected
c) Local connected
d) None of the mentioned
Answer: b
Explanation: None.

20. What is the consequence between a node and its predecessors while creating bayesian network?
a) Functionally dependent
b) Dependant
c) Conditionally independent
d) Both Conditionally dependant & Dependant
Answer: c
Explanation: The semantics to derive a method for constructing bayesian networks were led to the consequence that a node can be conditionally independent of its predecessors.

21. Which algorithm is used for solving temporal probabilistic reasoning?
a) Hill-climbing search
b) Hidden markov model
c) Depth-first search
d) Breadth-first search
Answer: b
Explanation: Hidden Markov model is used for solving temporal probabilistic reasoning that was independent of transition and sensor model.

22. How does the state of the process is described in HMM?
a) Literal
b) Single random variable
c) Single discrete random variable
d) None of the mentioned
Answer: c
Explanation: An HMM is a temporal probabilistic model in which the state of the process is described by a single discrete random variable.

23. What are the possible values of the variable?
a) Variables
b) Literals
c) Discrete variable
d) Possible states of the world
Answer: d
Explanation: The possible values of the variables are the possible states of the world.

24. Where does the additional variables are added in HMM?
a) Temporal model
b) Reality model
c) Probability model
d) All of the mentioned
Answer: a
Explanation: Additional state variables can be added to a temporal model while staying within the HMM framework.

25. Which allows for a simple and matrix implementation of all the basic algorithm?
a) HMM
b) Restricted structure of HMM
c) Temporary model
d) Reality model
Answer: b
Explanation: Restricted structure of HMM allows for a very simple and elegant matrix implementation of all the basic algorithm.

26. Where does the Hidden Markov Model is used?
a) Speech recognition
b) Understanding of real world
c) Both Speech recognition & Understanding of real world
d) None of the mentioned
Answer: a
Explanation: None.

27. Which variable can give the concrete form to the representation of the transition model?
a) Single variable
b) Discrete state variable
c) Random variable
d) Both Single & Discrete state variable
Answer: d
Explanation: With a single, discrete state variable, we can give concrete form to the representation of the transition model.

28. Which algorithm works by first running the standard forward pass to compute?
a) Smoothing
b) Modified smoothing
c) HMM
d) Depth-first search algorithm
Answer: b
Explanation: The modified smoothing algorithm works by first running the standard forward pass to compute and then running the backward pass.

29. Which reveals an improvement in online smoothing?
a) Matrix formulation
b) Revelation
c) HMM
d) None of the mentioned
Answer: a
Explanation: Matrix formulation reveals an improvement in online smoothing with a fixed lag.

30. Which suggests the existence of an efficient recursive algorithm for online smoothing?
a) Matrix
b) Constant space
c) Constant time
d) None of the mentioned
Answer: b
Explanation: None.

31. The Expectation Maximization algorithm has been used to identify conserved domains in unaligned proteins only.
a) True
b) False
Answer: b
Explanation: This algorithm has been used to identify both conserved domains in unaligned proteins and protein-binding sites in unaligned DNA sequences (Lawrence and Reilly 1990), including sites that may include gaps (Cardon and Stormo 1992). Given are a set of sequences that are expected to have a common sequence pattern and may not be easily recognizable by eye.

32. Which of the following is untrue regarding Expectation Maximization algorithm?
a) An initial guess is made as to the location and size of the site of interest in each of the sequences, and these parts of the sequence are aligned
b) The alignment provides an estimate of the base or amino acid composition of each column in the site
c) The column-by-column composition of the site already available is used to estimate the probability of finding the site at any position in each of the sequences
d) The row-by-column composition of the site already available is used to estimate the probability
Answer: d
Explanation: The EM algorithm then consists of two steps, which are repeated consecutively. In step 1, the expectation step, the column-by-column composition of the site already available is used to estimate the probability of finding the site at any position in each of the sequences. These probabilities are used in turn to provide new information as to the expected base or amino acid distribution for each column in the site.

33. Out of the two repeated steps in EM algorithm, the step 2 is ________
a) the maximization step
b) the minimization step
c) the optimization step
d) the normalization step
Answer: a
Explanation: In step 2, the maximization step, the new counts of bases or amino acids for each position in the site found in step 1 are substituted for the previous set. Step 1 is then repeated using these new counts. The cycle is repeated until the algorithm converges on a solution and does not change with further cycles. At that time, the best location of the site in each sequence and the best estimate of the residue composition of each column in the site will be available.

34. In EM algorithm, as an example, suppose that there are 10 DNA sequences having very little similarity with each other, each about 100 nucleotides long and thought to contain a binding site near the middle 20 residues, based on biochemical and genetic evidence. the following steps would be used by the EM algorithm to find the most probable location of the binding sites in each of the ______ sequences.
a) 30
b) 10
c) 25
d) 20
Answer: b
Explanation: When examining the EM program MEME, the size and number of binding sites, the location in each sequence, and whether or not the site is present in each sequence do not necessarily have to be known. For the present example, the following steps would be used by the EM algorithm to find the most probable location of the binding sites in each of the 10 sequences.

35. In the initial step of EM algorithm, the 20-residue-long binding motif patterns in each sequence are aligned as an initial guess of the motif.
a) True
b) False
Answer: a
Explanation: The base composition of each column in the aligned patterns is then determined. The composition of the flanking sequence on each side of the site provides the surrounding base or amino acid composition for comparison. Each sequence is assumed to be the same length and to be aligned by the ends.

36. In the intermediate steps of EM algorithm, the number of each base in each column is determined and then converted to fractions.
a) True
b) False
Answer: a
Explanation: For example, that there are four Gs in the first column of the 10 sequences, then the frequency of G in the first column of the site, fSG = 4/10 = 0.4. This procedure is repeated for each base and each column.

37. For the 10-residue DNA sequence example, there are _______ possible starting sites for a 20-residue-long site.
a) 30
b) 21
c) 81
d) 60
Answer: c
Explanation: For the 10-residue DNA sequence example, there are 100 – 20 +1 possible starting sites for a 20-residue-long site. Where the first one is at position 1 in the sequence ending one at 20 and the last beginning at position 81 and ending at 100 (there is not enough sequence for a 20-residue-long site beyond position 81).

38. An alternative method is to produce an odds scoring matrix calculated by dividing each base frequency by the background frequency of that base.
a) True
b) False
Answer: a
Explanation: In this method, the probability of each location is then found by multiplying the odds scores from each column. An even simpler method is to use log odds scores in the matrix. The column scores are then simply added. In this case, the log odds scores must be converted to odds scores before position probabilities are calculated.

39. Which of the following about MEME is untrue?
a) It is a Web resource for performing local MSAs (Multiple Sequence Alignment) by the above expectation maximization method is the program MEME
b) It stands for Multiple EM for Motif Elicitation
c) It was developed at developed at the University of California at San Diego Supercomputing Center
d) The Web page has multiple versions for searching blocks by an EM algorithm
Answer: d
Explanation: The Web page for two versions of MEME, ParaMEME, a Web program that searches for blocks by an EM algorithm (Described below), and a similar program MetaMEME (which searches for profiles using HMMs, described below).The Motif Alignment and Search Tool (MAST) for searching through databases for matches to motifs.

40. Which of the following about the Gibbs sampler is untrue?
a) It is a statistical method for finding motifs in sequences
b) It is dissimilar to the principle of the EM method
c) It searches for the statistically most probable motifs
d) It can find the optimal width and number of given motifs in each sequence
Answer: b
Explanation: It is another statistical method for finding motifs in sequences is the Gibbs sampler. The method is similar in principle to the EM method described above, but the algorithm is different. A combinatorial approach of the Gibbs sampler and MOTIF may be used to make blocks at the BLOCKS Web site.

41. Bayesian Belief Network is also known as ?
A. belief network
B. decision network
C. Bayesian model
D. All of the above
Answer : D
Explanation: Bayesian Belief Network also called a Bayes network, belief network, decision network, or Bayesian model.

42. Bayesian Network consist of ?
A. 2
B. 3
C. 4
D. 5
Answer : A
Explanation: Bayesian Network can be used for building models from data and experts opinions, and it consists of two parts: Directed Acyclic Graph and Table of conditional probabilities.

43. The generalized form of Bayesian network that represents and solve decision problems under uncertain knowledge is known as an?
A. Directed Acyclic Graph
B. Table of conditional probabilities
C. Influence diagram
D. None of the above
Answer : C
Explanation: The generalized form of Bayesian network that represents and solve decision problems under uncertain knowledge is known as an Influence diagram

44. How many component does Bayesian network have?
A. 2
B. 3
C. 4
D. 5
Answer : A
Explanation: The Bayesian network has mainly two components: Causal Component and Actual numbers

45. The Bayesian network graph does not contain any cyclic graph. Hence, it is known as a 
A. DCG
B. DAG
C. CAG
D. SAG
Answer : B
Explanation: The Bayesian network graph does not contain any cyclic graph. Hence, it is known as a directed acyclic graph or DAG.

46. In a Bayesian network variable is?
A. continuous
B. discrete
C. Both A and B
D. None of the above
Answer : C
Explanation: Each node corresponds to the random variables, and a variable can be continuous or discrete.

47. If we have variables x1, x2, x3,….., xn, then the probabilities of a different combination of x1, x2, x3.. xn, are known as?
A. Table of conditional probabilities
B. Causal Component
C. Actual numbers
D. Joint probability distribution
Answer : D
Explanation: If we have variables x1, x2, x3,….., xn, then the probabilities of a different combination of x1, x2, x3.. xn, are known as Joint probability distribution.

48. The nodes and links form the structure of the Bayesian network, and we call this the ?
A. structural specification
B. multi-variable nodes
C. Conditional Linear Gaussian distributions
D. None of the above
Answer : A
Explanation: The nodes and links form the structure of the Bayesian network, and we call this the structural specification.

49. Which of the following are used for modeling times series and sequences?
A. Decision graphs
B. Dynamic Bayesian networks
C. Value of information
D. Parameter tuning
Answer : B
Explanation: Dynamic Bayesian networks (DBNs) are used for modeling times series and sequences.

50. How many terms are required for building a bayes model?
A. 1
B. 2
C. 3
D. 4
Answer : C
Explanation: The three required terms are a conditional probability and two unconditional probability.

51) Which of the following algorithms cannot be used for reducing the dimensionality of data?
A. t-SNE
B. PCA
C. LDA False
D. None of these
Answer: (D)

52) [ True or False ] PCA can be used for projecting and visualizing data in lower dimensions.
A. TRUE
B. FALSE
Answer: (A)

53) The most popularly used dimensionality reduction algorithm is Principal Component Analysis (PCA). Which of the following is/are true about PCA?
PCA is an unsupervised method
It searches for the directions that data have the largest variance
Maximum number of principal components <= number of features
All principal components are orthogonal to each other
A. 1 and 2
B. 1 and 3
C. 2 and 3
D. 1, 2 and 3
E. 1,2 and 4
F. All of the above
Answer: (F)

54) Suppose we are using dimensionality reduction as pre-processing technique, i.e, instead of using all the features, we reduce the data to k dimensions with PCA. And then use these PCA projections as our features. Which of the following statement is correct?
A. Higher ‘k’ means more regularization
B. Higher ‘k’ means less regularization
C. Can’t Say
Answer: (B)

55) In which of the following scenarios is t-SNE better to use than PCA for dimensionality reduction while working on a local machine with minimal computational power?
A. Dataset with 1 Million entries and 300 features
B. Dataset with 100000 entries and 310 features
C. Dataset with 10,000 entries and 8 features
D. Dataset with 10,000 entries and 200 features
Answer: (C)

56) Which of the following statement is true for a t-SNE cost function?
A. It is asymmetric in nature.
B. It is symmetric in nature.
C. It is same as the cost function for SNE.
Answer: (B)

57) Imagine you are dealing with text data. To represent the words you are using word embedding (Word2vec). In word embedding, you will end up with 1000 dimensions. Now, you want to reduce the dimensionality of this high dimensional data such that, similar words should have a similar meaning in nearest neighbor space.In such case, which of the following algorithm are you most likely choose?
A. t-SNE
B. PCA
C. LDA
D. None of these
Answer: (A)

58) [True or False] t-SNE learns non-parametric mapping.
A. TRUE
B. FALSE
Answer: (A)

59) Which of the following statement is correct for t-SNE and PCA?
A. t-SNE is linear whereas PCA is non-linear
B. t-SNE and PCA both are linear
C. t-SNE and PCA both are nonlinear
D. t-SNE is nonlinear whereas PCA is linear
Answer: (D)

60) In t-SNE algorithm, which of the following hyper parameters can be tuned?
A. Number of dimensions
B. Smooth measure of effective number of neighbours
C. Maximum number of iterations
D. All of the above
Answer: (D)

61) The minimum time complexity for training an SVM is O(n2). According to this fact, what sizes of datasets are not best suited for SVM’s?
A) Large datasets
B) Small datasets
C) Medium sized datasets
D) Size does not matter
Answer: A

62) The effectiveness of an SVM depends upon:
A) Selection of Kernel
B) Kernel Parameters
C) Soft Margin Parameter C
D) All of the above
Answer: D

63) Support vectors are the data points that lie closest to the decision surface.
A) TRUE
B) FALSE
Answer: A

64) The SVM’s are less effective when:
A) The data is linearly separable
B) The data is clean and ready to use
C) The data is noisy and contains overlapping points
Answer: C

65) Suppose you are using RBF kernel in SVM with high Gamma value. What does this signify?
A) The model would consider even far away points from hyperplane for modeling
B) The model would consider only the points close to the hyperplane for modeling
C) The model would not be affected by distance of points from hyperplane for modeling
D) None of the above
Answer: B

66) The cost parameter in the SVM means:
A) The number of cross-validations to be made
B) The kernel to be used
C) The tradeoff between misclassification and simplicity of the model
D) None of the above
Answer: C

67)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 it’s hyper parameter. Based upon that give the answer for following question.
What would happen when you use very large value of C(C->infinity)?
Note: For small C was also classifying all data points correctly
A) We can still classify data correctly for given setting of hyper parameter C
B) We can not classify data correctly for given setting of hyper parameter C
C) Can’t Say
D) None of these
Answer: A

68) What would happen when you use very small C (C~0)?
A) Misclassification would happen
B) Data will be correctly classified
C) Can’t say
D) None of these
Answer: A

69) 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?
A) Underfitting
B) Nothing, the model is perfect
C) Overfitting
Answer: C

70) Which of the following are real world applications of the SVM?
A) Text and Hypertext Categorization
B) Image Classification
C) Clustering of News Articles
D) All of the above
Answer: D

Question Context: 71 – 72
Suppose you have trained an SVM with linear decision boundary after training SVM, you correctly infer that your SVM model is under fitting.

71) Which of the following option would you more likely to consider iterating SVM next time?
A) You want to increase your data points
B) You want to decrease your data points
C) You will try to calculate more variables
D) You will try to reduce the features
Answer: C

72) Suppose you gave the correct answer in previous question. What do you think that is actually happening?
1. We are lowering the bias
2. We are lowering the variance
3. We are increasing the bias
4. We are increasing the variance
A) 1 and 2
B) 2 and 3
C) 1 and 4
D) 2 and 4
Answer: C

73) In above question suppose you want to change one of it’s(SVM) hyperparameter so that effect would be same as previous questions i.e model will not under fit?
A) We will increase the parameter C
B) We will decrease the parameter C
C) Changing in C don’t effect
D) None of these
Answer: A

74) 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 variables
3. Feature normalization always helps when we use Gaussian kernel in SVM
A) 1
B) 1 and 2
C) 1 and 3
D) 2 and 3
Answer: B

Question Context: 75
Suppose you are dealing with 4 class classification problem and you want to train a SVM model on the data for that you are using One-vs-all method. Now answer the below questions?

75) How many times we need to train our SVM model in such case?
A) 1
B) 2
C) 3
D) 4
Solution: D

Learning with Regression and Trees

1. In practice, Line of best fit or regression line is found when _____________
a) Sum of residuals (∑(Y – h(X))) is minimum
b) Sum of the absolute value of residuals (∑|Y-h(X)|) is maximum
c) Sum of the square of residuals ( ∑ (Y-h(X))2) is minimum
d) Sum of the square of residuals ( ∑ (Y-h(X))2) is maximum
Answer: c
Explanation: Here we penalize higher error value much more as compared to the smaller one, such that there is a significant difference between making big errors and small errors, which makes it easy to differentiate and select the best fit line.

2. If Linear regression model perfectly first i.e., train error is zero, then _____________________
a) Test error is also always zero
b) Test error is non zero
c) Couldn’t comment on Test error
d) Test error is equal to Train error
Answer: c
Explanation: Test Error depends on the test data. If the Test data is an exact representation of train data then test error is always zero. But this may not be the case.

3. Which of the following metrics can be used for evaluating regression models?
i) R Squared ii) Adjusted R Squared iii) F Statistics iv) RMSE / MSE / MAE
a) ii and iv
b) i and ii
c) ii, iii and iv
d) i, ii, iii and iv
Answer: d
Explanation: These (R Squared, Adjusted R Squared, F Statistics, RMSE / MSE / MAE) are some metrics which you can use to evaluate your regression model.

4. How many coefficients do you need to estimate in a simple linear regression model (One independent variable)?
a) 1
b) 2
c) 3
d) 4
Answer: b
Explanation: In simple linear regression, there is one independent variable so 2 coefficients (Y=a+bx+error).

5. In a simple linear regression model (One independent variable), If we change the input variable by 1 unit. How much output variable will change?
a) by 1
b) no change
c) by intercept
d) by its slope
Answer: d
Explanation: For linear regression Y=a+bx+error. If neglect error then Y=a+bx. If x increases by 1, then Y = a+b(x+1) which implies Y=a+bx+b. So Y increases by its slope.

6. Function used for linear regression in R is __________
a) lm(formula, data)
b) lr(formula, data)
c) lrm(formula, data)
d) regression.linear(formula, data)
Answer: a
Explanation: lm(formula, data) refers to a linear model in which formula is the object of the class “formula”, representing the relation between variables. Now this formula is on applied on the data to create a relationship model.

7. In syntax of linear model lm(formula,data,..), data refers to ______
a) Matrix
b) Vector
c) Array
d) List
Answer: b
Explanation: Formula is just a symbol to show the relationship and is applied on data which is a vector. In General, data.frame are used for data.

8. In the mathematical Equation of Linear Regression Y = β1 + β2X + ϵ, (β1, β2) refers to __________
a) (X-intercept, Slope)
b) (Slope, X-Intercept)
c) (Y-Intercept, Slope)
d) (slope, Y-Intercept)
Answer: c
Explanation: Y-intercept is β1 and X-intercept is – (β1 / β2). Intercepts are defined for axis and formed when the coordinates are on the axis.

9) Looking at above two characteristics, which of the following option is the correct for Pearson correlation between V1 and V2?
If you are given the two variables V1 and V2 and they are following below two characteristics.
1. If V1 increases then V2 also increases
2. If V1 decreases then V2 behavior is unknown
A) Pearson correlation will be close to 1
B) Pearson correlation will be close to -1
C) Pearson correlation will be close to 0
D) None of these
answer: (D)

10) Suppose Pearson correlation between V1 and V2 is zero. In such case, is it right to conclude that V1 and V2 do not have any relation between them?
A) TRUE
B) FALSE
answer: (B)

11) Which of the following offsets, do we use in linear regression’s least square line fit? Suppose horizontal axis is independent variable and vertical axis is dependent variable.

A) Vertical offset
B) Perpendicular offset
C) Both, depending on the situation
D) None of above
answer: (A)

12) True- False: Overfitting is more likely when you have huge amount of data to train?
A) TRUE
B) FALSE
answer: (B)

13) 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?
We don’t have to choose the learning rate
It becomes slow when number of features is very large
Thers is no need to iterate
A) 1 and 2
B) 1 and 3
C) 2 and 3
D) 1,2 and 3
answer: (D)

14) Which of the following statement is true about sum of residuals of A and B?
Below graphs show two fitted regression lines (A & B) on randomly generated data. Now, I want to find the sum of residuals in both cases A and B.
Note:
Scale is same in both graphs for both axis.
X axis is independent variable and Y-axis is dependent variable.

A) A has higher sum of residuals than B
B) A has lower sum of residual than B
C) Both have same sum of residuals
D) None of these
answer: (C)

15) Choose the option which describes bias in best manner.
A) In case of very large x; bias is low
B) In case of very large x; bias is high
C) We can’t say about bias
D) None of these
answer: (B)

16) What will happen when you apply very large penalty?
A) Some of the coefficient will become absolute zero
B) Some of the coefficient will approach zero but not absolute zero
C) Both A and B depending on the situation
D) None of these
answer: (B)

17) What will happen when you apply very large penalty in case of Lasso?
A) Some of the coefficient will become zero
B) Some of the coefficient will be approaching to zero but not absolute zero
C) Both A and B depending on the situation
D) None of these
answer: (A)

18) Which of the following statement is true about outliers in Linear regression?
A) Linear regression is sensitive to outliers
B) Linear regression is not sensitive to outliers
C) Can’t say
D) None of these
answer: (A)

19) Suppose you plotted a scatter plot between the residuals and predicted values in linear regression and you found that there is a relationship between them. Which of the following conclusion do you make about this situation?
A) Since the there is a relationship means our model is not good
B) Since the there is a relationship means our model is good
C) Can’t say
D) None of these
answer: (A)

20) What will happen when you fit degree 4 polynomial in linear regression?
A) There are high chances that degree 4 polynomial will over fit the data
B) There are high chances that degree 4 polynomial will under fit the data
C) Can’t say
D) None of these
answer: (A)

21) What will happen when you fit degree 2 polynomial in linear regression?
A) It is high chances that degree 2 polynomial will over fit the data
B) It is high chances that degree 2 polynomial will under fit the data
C) Can’t say
D) None of these
answer: (B)

22) In terms of bias and variance. Which of the following is true when you fit degree 2 polynomial?
A) Bias will be high, variance will be high
B) Bias will be low, variance will be high
C) Bias will be high, variance will be low
D) Bias will be low, variance will be low
answer: (C)

23) Suppose l1, l2 and l3 are the three learning rates for A,B,C respectively. Which of the following is true about l1,l2 and l3?

A) l2 < l1 < l3
B) l1 > l2 > l3
C) l1 = l2 = l3
D) None of these
answer: (A)

24) 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?
A) Increase
B) Decrease
C) Remain constant
D) Can’t Say
answer: (D)

25) What do you expect will happen with bias and variance as you increase the size of training data?
A) Bias increases and Variance increases
B) Bias decreases and Variance increases
C) Bias decreases and Variance decreases
D) Bias increases and Variance decreases
E) Can’t Say False
answer: (D)

26) What would be the root mean square training error for this data if you run a Linear Regression model of the form (Y = A0+A1X)?

A) Less than 0
B) Greater than zero
C) Equal to 0
D) None of these
answer: (C)

Question Context 27-28:
Suppose you have been given the following scenario for training and validation error for Linear Regression.

Scenario	Learning Rate	Number of iterations	Training Error	Validation Error
1	0.1	1000	100	110
2	0.2	600	90	105
3	0.3	400	110	110
4	0.4	300	120	130
5	0.4	250	130	150

27) Which of the following scenario would give you the right hyper parameter?
A) 1
B) 2
C) 3
D) 4
answer: (B)

28) Suppose you got the tuned hyper parameters from the previous question. Now, Imagine you want to add a variable in variable space such that this added feature is important. Which of the following thing would you observe in such case?
A) Training Error will decrease and Validation error will increase
B) Training Error will increase and Validation error will increase
C) Training Error will increase and Validation error will decrease
D) Training Error will decrease and Validation error will decrease
E) None of the above
answer: (D)

29) In such situation which of the following options would you consider?
Add more variables
Start introducing polynomial degree variables
Remove some variables
A) 1 and 2
B) 2 and 3
C) 1 and 3
D) 1, 2 and 3
answer: (A)

30) Now situation is same as written in previous question(under fitting).Which of following regularization algorithm would you prefer?
A) L1
B) L2
C) Any
D) None of these
answer: (D)

31) Which of the following evaluation metrics can not be applied in case of logistic regression output to compare with target?
A) AUC-ROC
B) Accuracy
C) Logloss
D) Mean-Squared-Error
answer: D

32) One of the very good methods to analyze the performance of Logistic Regression is AIC, which is similar to R-Squared in Linear Regression. Which of the following is true about AIC?
A) We prefer a model with minimum AIC value
B) We prefer a model with maximum AIC value
C) Both but depend on the situation
D) None of these
answer: A

33) [True-False] Standardisation of features is required before training a Logistic Regression.
A) TRUE
B) FALSE
answer: B

34) Which of the following algorithms do we use for Variable Selection?
A) LASSO
B) Ridge
C) Both
D) None of these
answer: A

Context: 35-36
Consider a following model for logistic regression: P (y =1|x, w)= g(w0 + w1x)
where g(z) is the logistic function.
In the above equation the P (y =1|x; w) , viewed as a function of x, that we can get by changing the parameters w.

35) What would be the range of p in such case?
A) (0, inf)
B) (-inf, 0 )
C) (0, 1)
D) (-inf, inf)
answer: C

36) In above question what do you think which function would make p between (0,1)?
A) logistic function
B) Log likelihood function
C) Mixture of both
D) None of them
answer: A

37) Suppose you have been given a fair coin and you want to find out the odds of getting heads. Which of the following option is true for such a case?
A) odds will be 0
B) odds will be 0.5
C) odds will be 1
D) None of these
answer: C

38) The logit function(given as l(x)) is the log of odds function. What could be the range of logit function in the domain x=[0,1]?
A) (– ∞ , ∞)
B) (0,1)
C) (0, ∞)
D) (- ∞, 0)
answer: A

39) Which of the following option is true?
A) Linear Regression errors values has to be normally distributed but in case of Logistic Regression it is not the case
B) Logistic Regression errors values has to be normally distributed but in case of Linear Regression it is not the case
C) Both Linear Regression and Logistic Regression error values have to be normally distributed
D) Both Linear Regression and Logistic Regression error values have not to be normally distributed
answer: A

40) Which of the following is true regarding the logistic function for any value “x”?
Note:
Logistic(x): is a logistic function of any number “x”
Logit(x): is a logit function of any number “x”
Logit_inv(x): is a inverse logit function of any number “x”
A) Logistic(x) = Logit(x)
B) Logistic(x) = Logit_inv(x)
C) Logit_inv(x) = Logit(x)
D) None of these
answer: B

41. A _________ is a decision support tool that uses a tree-like graph or model of decisions and their possible consequences, including chance event outcomes, resource costs, and utility.
a) Decision tree
b) Graphs
c) Trees
d) Neural Networks
Answer: a
Explanation: Refer the definition of Decision tree.

42. Decision Tree is a display of an algorithm.
a) True
b) False
Answer: a
Explanation: None.

43. What is Decision Tree?
a) Flow-Chart
b) Structure in which internal node represents test on an attribute, each branch represents outcome of test and each leaf node represents class label
c) Flow-Chart & Structure in which internal node represents test on an attribute, each branch represents outcome of test and each leaf node represents class label
d) None of the mentioned
Answer: c
Explanation: Refer the definition of Decision tree.

44. Decision Trees can be used for Classification Tasks.
a) True
b) False
Answer: a
Explanation: None.

45. Choose from the following that are Decision Tree nodes?
a) Decision Nodes
b) End Nodes
c) Chance Nodes
d) All of the mentioned
Answer: d
Explanation: None.

46. Decision Nodes are represented by ____________
a) Disks
b) Squares
c) Circles
d) Triangles
Answer: b
Explanation: None.

47. Chance Nodes are represented by __________
a) Disks
b) Squares
c) Circles
d) Triangles
Answer: c
Explanation: None.

48. End Nodes are represented by __________
a) Disks
b) Squares
c) Circles
d) Triangles
Answer: d
Explanation: None.

49. Which of the following are the advantage/s of Decision Trees?
a) Possible Scenarios can be added
b) Use a white box model, If given result is provided by a model
c) Worst, best and expected values can be determined for different scenarios
d) All of the mentioned
Answer: d
Explanation: None.