Note: Attempt all questions. The qu question carries equal mar that question in the attached OMR sheet. 1. What is Machine learning? 4AT ~The autonomous acquisition of knowledge through the use of computer programs (B) The autonomous acquisition of kmowledge through the use of manual programs (©) The selective acquisition of knowledge through the use of computer programs (D) The selective acquisition of knowledge through the use of manual programs 2. Which of the factors affect the performance of learner system does not include? (A) _ Representation scheme used ® __ Training scenario (©) Type of feedback (D) Good data structures 3, Which of the following statements a “Type-2” errors? V7 Typel is known as false positive and Type2 is known as false negative (i) Type is known as false negative and Type? is known estion paper contain ks. Select the answer and Ga) Oy) ®) ey” ©) 4, How is 70 MCQ type questions. Each ‘fill thie bubble corresponding to ‘typel error oécurs when We reject a null hypothesis when it is true. Only i Only iif j and iii ii and iii do you handle missing oF corrupted data in a dataset? A @) © Drop missing rows or columns Replace missing values with mean/median/mode Assign a unique category to missing values Ley” All of the above 5, Which is of the following option is true about FIND-S Algorithm \ (Ay FIND-S Algorithm starts from @) © the most specific hypothesis and generalize it by considering only positive examples. FIND-S_ algorithm ignores negative examples. FIND-S algorithm finds the most specific hypothesis within H_ that is consistent with the positive training examples. Scanned with CamScannerCO ©) Allofthe above (C) High estimation bias Regarding bias and variance, which (D) None of the above of the following statements are true? Here ‘high’ and ‘low’ are relative to the ideal model.) 9. Adding more basis functions in a linear model... (pick the most probably option) (A) Models which overfit have a AX Decreases model bias high bias, (B) Decreases estimation bias \BY Models which overfit have a (C) Decreases variance (D) Doesn't affect bias and variance low bias, (©) Models which underfit have a high variance, 10. Which of the following will be true () None of these about k in k-NN in terms of Bias? Which of the following sentence is ‘NAT When you increase the k the FALSE regarding regression? bias will be increases (A) _ Itrelates inputs to outputs. (B) When you decrease the k the (B) _ Itis used for prediction. bias will be increases (©) It may be used for © Can’t say interpretation, (D) None of these DY It discovers causal 11. Which of the following distance relationships. measure do we use in case of You observe the following while categorical variables in k-NN? fitting a linear regression to the data: Hamming Distance As you increase the amount of Euclidean Distance training data, the test error decreases, Manhattan Distance and the training error increases. The iy train error is quite low (almost what @) 2 you expect it to), while the test error © 3 is much higher than the train error. (P) 1,2and3 . ‘What do you think is the main reason 12. Imagine, you are working with behind this behavior? Choose the “Analytics Vidhya” and you want to most probable option. develop a machine learning algorithm NAS High variance W25 which predicts the number of views iG High model bias on the articles, Scanned with CamScanner13 14, Your analysis is based on features like author name, number of articles written by the same author on Analytics Vidhya in past and a few Which of the following evaluation metric would other features. you choose in that case? Mean Square Error Accuracy FI Score fA) Only 1 (B) Only2 (©) Only3 (D) Land3 At a certain university, 4% of men are over 6 feet tall and 1% of women are over 6 feet tall. The total student population is divided in the ratio 3:2 in favour of women. If a student is selected at random from among all those over six feet tall, what is the probability that the student is a woman? (A) 25 (B) ¥% US Mi (D) 100 Macromutation operator is also known as (A) Headed Chicken (B) Headless chicken (C) SPX operator (D) BLX operator 15, . Choose the False Statement, Gradient of a continuous and differentiable function CAT” is zero at a minimum UAB) is non-zero at a maximum (YOY is zero ata saddle point {SD} decreases as you get closer to 16. 17. 18. the minimum Computational complexity of Gradient descent is, “ (B) linear in N LCT polynomial in D (D) dependent on the number of linear in D iterations Let’s say, you are using activation function X in hidden layers of neural network. At a neuron for any given input, you get the output as“ 0.0001". Which of the following could X activation function represent? Ay ReLU (B) tanh (©) siGMoID (D) None of these Which of following hyper parameter(s), when increased may the cause random forest to over fit the data? Number of Trees Depth of Tree Learning Rate Scanned with CamScanner(A) Only1 “BY Only 2 © land2 (D) 2and3 19. Which of the following is a disadvantage of decision trees? (A) Factor analysis (B) Decision trees are robust to outliers ACY” Decision trees are prone to be overfit (D) None of the above 20. To find the minimum or the maximum of a function, we set the gradient to zero because: AY The value of the gradient at extrema of a function always zero (B) Depends on the type of problem (©) Both A andB (D) None of the above 21. In Delta Rule for error minimization (A) weights are adjusted w.r.to change in the output Bw difference between desired hts are adjusted w.r.to output and actual output (©) weights are adjusted w.rto difference between input and output (D) none of the above 22. Back propagation is a learning technique that adjusts weights in the neural network by propagating weight changes. (A) _ Forward from source to sink BY Backward from sink to source (©) Forward from source to hidden nodes (D) Backward from sink to hidden nodes 23. Which of the following neural networks uses supervised learning? (A) Multilayer perceptron (B) Self organizing feature map (C) Hopfield network Choose the correct answer: (A) Aonly (B) Bonly (C) Aand B only \_@y~ A and C only 24. Which of the following sentences is incorrect in reference to Information gain? UAT is biased towards. single valued attributes (B) It is biased towards multi- valued attributes (© 1D3 makes use of information gain (D) The approach used by ID3 is greedy 25. What are two steps of tree pruning work? (A) Pessimistic prunine and Scanned with CamScannerOptimistic pruning \{B)Post-pruning and Pre-pruning (C) Cost complexity pruning and time complexity pruning (D) None of the options 26. Which one of these is not a tree- based leamer? (A) CART (B) 1D3 AC)? Bayesian classifier (D) Random Forest 27. What is tree-based classifiers? (A) Classifiers which form a tree with each attribute at one level (B) Classifiers which perform series of condition checking with one attribute at a time eH Bownaandd (D) None of the options 28. Decision Nodes are represented by (A) Disks (B) Squares ey Circles (D) Triangles 29. Previous probabilities in Bayes Theorem that are changed with help of new available information are classified as (A) _ independent probabilities \£By posterior probabilities (©) interior probabilities annandant nrobabilities 30. Which of the following is true about Naive Bayes? (A) Assumes that all the features in a dataset are equally important (B) Assumes that all the features ina dataset are independent Oy Both A and B (D) None of the above options 31. The method in which the previously calculated probabilities are revised with new probabilities is classified as (A) _ updating theorem (B) _ revised theorem (Oy Bayes theorem (D) dependency theorem 32. Which of the following is a widely used and effective machine learning algorithm based on the idea of bagging? — (A) _ Decision Tree \eiBF” Regression © (C) Classification LQ Random Forest 33, Which of the following is a good test dataset characteristic? (A) Large enough (0 yield meaningful results (B) _ Is representative of the dataset as a whole ey Both A and B (p) None of the above +34, What is the arity in case of crossover Scanned with CamScannera 35. 36. 37. ©perator in GA? A) Number of parents used for the operator ®) Number of offspring used for the Operator Both a and b ©) None Which of the following statements about regularization is not correct? (A) Using too large a value of lambda can cause your hypothesis to underfit the data, (B) Using too large a value of lambda can cause your hypothesis to overfit the data, (©) Using a very large value of lambda cannot hurt the performance of your hypothesis. (D) None of the above You are given reviews of movies marked as positive, negative, and neutral. Classifying reviews of a new movie is an example of JAI Supervised Learning (B) Unsupervised Leaning (©) Reinforcement Learning (D) None of these Regarding bias and variance, which of the following statements are true? (Here ‘high’ and ‘low’ are relative to the ideal model.) @ Models which overfit have a high bias. (i) Models which overfit have low bias. Gi) Models which underfit have g high variance. (iv) Models which underfit have q low variance (A) Gand Gi) By Ui) and (iv) (©) Gi) and (iv) (©) None of these 38. What is the purpose of restricting hypothesis space in machine learning? (A) can be easier to search ~(B)_ May avoid overfit since they are usually simpler (e.g, linear w order decision surface) © Both above (D) None of the above 39. 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? (A) You will add more features () You will start introducing higher degree features (You wit features “DY Both a and, 4 40. Consider a simple linear regressio?® remove some model with one independent variable Scanned with CamScanner(X). The output variable ¢ ae ae is x. The slope sind b is the sme the change the input vari ae an unit, by how much eee output variable (Y) will change? (A) Lunit (B) By slope (©) By intercept (D) None 41. You have generated data from a 3- degree polynomial with some noise. What do you expect of the model that was trained on this data using a 5- degree polynomial as function class? (A) _ Low bias, high variance (B) High bias, low variance. (C)__ Low bias, low variance. (D) High bias, low variance. 42, Genetic Algorithm are a part of (A) Evolutionary Computing (B) inspired by Darwin's theory about evolution - “survival of the fittest” (©) are adaptive heuristic algorithm based 0” the ‘as of natural search evolutionary ide selection and genetics CH All ofthe above types of leaming 43. What are the 2 (A) _Improvised and un-improvised supervised and unsupervised < seed and unlayered (D) None of the above 44, Unsupervised leaming is (A) _ learning without computers <(B)Froblem based learning (C) learning from environment (D) learning from teachers 45. In supervised learning (A) classes are not predefined \(B)Uasses are predefined (C) classes are not required (D) classification is not done 46, Mutating a strain is: \_(A)—Changing all the genes in the strain. (B) Removing one gene in the strain. (©) Randomly changing one gene in the strain (D) Removing the strain from the population. 47. Genetic Algorithms are considered pseudo-random because they: (A) Search the solution space in a random fashion. (B) Search the solution space using the previous generation as a starting point. (©) Have no knowledge of what strains are contained in the next generation. Ugpy Use random numbers. 4g, The three gene operators We have discussed can be thought of as: Scanned with CamScanner(A) Crossover: Receiving the best genes from both parents. (B) Mutation: Changing one gene so that the child is almost like the parent. (C) Mirror: Changing a string of genes in the child so itis like a cousinto the parent. Ney and B only 49. If a population contains only one strain, you can introduce new strains by: Using the Crossover operator. (A) _ Injecting random strains into the population. (B) Using the Mutation operator. Oy-Bonly (D) Band C only 50. The efficiency of a Genetic Algorithm (how quickly it arrives at the best solution) is dependent upon: (A) The initial conditions. (B) The size of the population. (©) The types of operators employed. \@ All of the above 51, Which of the following methods do we use, to find the best fit line for data in Linear Regression? Least Square Error : (B) Maximum Likelihood (©) Logarithmic Loss (D) Both A and B 52. Among the following, which one is not "hyperparameter"? (A) Bearning rate a (B) number of layers L in the neural network (©) activation values afl] (D) size of the hidden layers nfl] 53. (The deeper layers of a neural network —are_—_typically computing more complex features of the input than the earlier layers. (ii) The earlier layers of a neural network are _typically computing more complex features of the input than the deeper layers. Which of the following option is correct? (A) (i) is correct and Gi) is incorrect (B) Gi) is incorrect while (ii) is correct (©) _ both are correct (D) _ both are incorrect 54. There are certain functions with the following properties: () To compute the function using a shallow network circuit, you will need a large network (where we measure size by the number of logic gates in the network) (i) To compute it using a deep network circuit, you need only an exnonentially —aneattar si Scanned with CamScanner55. network, Whi , ; hich of the following option is correct? wn (A) @_ is correct and (ii) is incorrect (B) (i) is incorrect while (ii) is correct (C)__ both are comect (D) _ both are incorrect Factor Analysis involves: (A) dimensionality reduction technique _-(BY finding correlation among variables (©) capturing maximum variance in the data with minimum number of variables \ DyAN the above 56, Which of the following is way (© 57. reduce the skewness of a variable? ar Taking log of the skewed variable (B) _ Dividing each value of skewed variable by its standard deviation. (c)_ Normalizing the skewed variable () Standardizing the skewed variable. what causes overfitting? ; (A) Large number of features in the data (B) Noise in the dat “OC BothAandB (D) None of the above 58. Given an image of a person, (i) predicting the height of that person (ii) finding whether the person is in happy, angry or sad mood. type of ML problem is (A) (iis classification while (i) is regression problem (8) Ti) is classification while (i) is regression problem (©) bothare classification problem (D) _ both are regression problem 59, what does fitness function represent to describe optimization problem? AT Objective function (B) Scaling function (©) Chromosome decoding function (D) Allofthe above 60. which of the following algorithms is called Lazy Learner? “Ay KNN (B) SVM (©) Naive Bayes (D) Decision Tree 61. What are the main driving operators of GA? (A) Selection (B) Crossover Yer Both a and b (D) None of these Scanned with CamScanner62, Which of the following is true about bagging and boosting? (A) Both are ensemble learning, techniques ©) Both combine the output of Weak leamers to make Consistent predictions (C) Both can be used to solve classification as well as regression problems EDTA of the above 63. what causes underfitting? (A) Less number of features in the data (B) Less number of observations in the data \t@ Both a and b (D) None of the above 64. The performance of GA is influenced by (A) Population size (B) Crossover rate (©) Mutation rate \ (DAM of the above 65. which of the following are main 66. components of ~_ evolutionary computation? (A) _ Initial population (B) Fitness function ET Crossover, selection (D) Allof the above which of the following statement(s) mutation and tobi 67. 68. 69. 70. is/are true’? (A) Genetic algorithm — mimic process from natural selection (B) Chromosomes play vital roles inGA Ley” Botha and b (D) Chromosomes can’t be encoded characteristics of individual is represented by A) Thromosomes (B) Gray Code (©) _ Initial population (D) None of the above what is the main concept of Evolutionary computation? \-AY Survival of the fittest (B) Survival of the weakest (C) Phenotype (D) None of these selective pressure is also known as (A) Takeover Time (B) candidate solution (CY Proportionate time () . Which None of the above strategy is high selection low selection susceptible to a pressure and population diversity? EAT Roulette-wheel selection (B) (© (Dd) Rank based selection Tournament selection All of the above Scanned with CamScanner