Machine Learning MCQ



Question 1 : Why SVM’s are more accurate than logistic regression?

  1. SVM gives more weightage to wrongly classified data points.
  2. SVM gives more weightages to data points which are correctively classified .
  3. SVM uses all the data points assuming a probabilistic model.
  4. SVM uses concept of large margin seperator and for non linearity it uses kernel functions
  

Question 2 : What is true about the discount factor in reinforcement learning?

  1. discount factor should be greater than 1
  2. discount factor should always be negative
  3. discount factor should be in range of 0 and 1
  4. discount factor can be any real number
  

Question 3 : Which algorithm is used for performing probabilistic reasoning on temporal data?

  1. Hill-climbing search
  2. Hidden markov model
  3. Naïve Method
  4. Support Vector Machine
  

Question 4 : Choose correct applications of reinforcement learning?

  1. Aircraft Control
  2. Sentimental analysis
  3. House price prediction
  4. Spam Email Filtering
  

Question 5 : Support Vector Machine(SVM) can be used for both classification or regression challenges.Which kind of learning technique SVM uses?

  1. supervised
  2. unsupervised
  3. reinforced
  4. clustered
  

Question 6 : What are two steps of pruning in decision tree ?

  1. Pessimistic pruning and Optimistic pruning
  2. Postpruning and Prepruning
  3. Cost complexity pruning and time complexity pruning
  4. High pruning and low pruning
  

Question 7 : Various ____ methods and techniques are used for calculation of the outliers.

  1. distance calculation
  2. prediction
  3. optimization
  4. integration
  

Question 8 : Which of the following is a clustering algorithm in machine learning?

  1. Expectation Maximization
  2. CART
  3. Gaussian Naïve Bayes
  4. Apriori
  

Question 9 : What is the approach of basic algorithm for decision tree induction?

  1. Greedy
  2. Top Down
  3. Procedural
  4. Step by Step
  

Question 10 : Support Vector Machine(SVM) performs well in _____ dimension spaces.

  1. high
  2. low
  3. wide
  4. single
  

Question 11 : You ran gardient descent for 20 iterations with learning rate=0.2 and compute cost for each iteration.You observe that cost decreases after each iteration .Based on this which conclusion is more suitable.

  1. Try smaller values of learning rate like 0.01
  2. 0.2 is effective choice of learning rate.
  3. Try larger values of learning rate like 1.
  4. Try any negetive value for learning rate
  

Question 12 : What are support vectors?

  1. These are the datapoints which help the SVM to generate optimal hyperplane.
  2. It is an intermediate vector generated during calculation of optimal hyperplane
  3. In SVM all the data points are called support vectors.
  4. This are predefined vectors used in calculating hyperplane
  

Question 13 : Which of the following problems can be solved by supervised learning too? Assume appropriate dataset is available.

  1. From a large collection of spam emails, discover if there are sub types of spam emails.
  2. Given data on how 1000 medical patients respond to an experimental medicine , discover whether there are different categories of patients in terms of how they respond to , and if so what are these categories
  3. Given a large dataset of medical records from patients suffering from heart disease, try to learn whether there might be different groups of patients for which customised treatment is required
  4. Given genetic (DNA) data from a person, predict the odds of the person developing diabetes over the next 10 years
  

Question 14 : Principal component analysis(PCA) is used for___.

  1. Dimensionality Enhancement
  2. LU Decomposition
  3. QR Decomposition
  4. Dimensionality Reduction
  

Question 15 : ___________ phenomenon refers that a model is neither trained on training data nor generalized properly on new data.

  1. good fitting
  2. overfitting
  3. moderate fitting
  4. underfitting
  

Question 16 : In Logistic regression predicted value of Y lies within _____ range.

  1. 0 to 1
  2. 0 to -∞
  3. -∞ to +∞
  4. -1 to 1
  

Question 17 : Choose the correct tree based learner.

  1. Rule based
  2. Hidden markov model
  3. Bayesian classifier
  4. CART
  

Question 18 : In principal component analysis ,if eigenvalues are equal.What does it mean?

  1. PCA will perform outstandingly
  2. PCA will perform badly
  3. No effect
  4. Model will be unstable
  

Question 19 : For a trained logistic classifer given a sample x,it gives prediction as 0.8.This means that___.

  1. P(Y=0|x)=0.8
  2. P(Y=1|x)=0.8
  3. P(Y=0|x)=0.2
  4. P(Y=1|x)=0.2
  

Question 20 : K-fold cross-validation is____.

  1. linear in K
  2. quadratic in K
  3. cubic in K
  4. exponential in K
  

Question 21 : Consider a point that is correctly classified and distant from the decision boundary. Which of the following methods will be unaffected by this point?

  1. Nearest neighbor
  2. Support Vector Machine
  3. Logistic regression
  4. Linear regression
  

Question 22 : You are training an RBF SVM with the following parameters: C (slack penalty) and γ = 1/2σ 2 (where σ 2 is the variance of the RBF kernel). How should you tweak the parameters to reduce overfitting?

  1. Increase C and/or reduce γ
  2. Reduce C and/or increase γ
  3. Reduce C and/or reduce γ
  4. Reduce C only (γ has no predictable effect on overfitting)
  

Question 23 : Machine Learning comes under which of the following domain?

  1. Artificial Intelligence
  2. Network Security
  3. Engineeering sciences
  4. System programming
  

Question 24 : Choose the reason for pruning a Decision Tree?

  1. To save computing time during testing
  2. To avoid overfitting the training set
  3. To save space for storing the Decision Tree
  4. To make the training set error smaller
  

Question 25 : While comparing reinforcement learning and supervised learning, which of the following statement is true?

  1. Both in reinforcement and supervised learning decisions are taken sequentially
  2. Supervised learning is best suited where human interaction is prevalant wheareas reinforcement learning is best suited for sofware systems.
  3. Reinforcement learning works by interacting with environment wheareas supervised learning works on sample data
  4. Both in reinforcement and supervised learning decisions taken at one time step is independent with respect to previous timestep.
  

Question 26 : Neural networks:

  1. Optimize a convex objective function
  2. Can use a mix of different activation functions
  3. are not suitable for learning.
  4. Can only be trained with stochastic gradient descent
  

Question 27 : Which of the following is not a clustering algorithm?

  1. EM-Algorithm
  2. K-means clustering
  3. Radial Basis Function
  4. Decision Tree
  

Question 28 : Principal component analysis is a technique for performing

  1. Dimensionality reduction
  2. Pruning
  3. Aggregation
  4. Sampling
  

Question 29 : The process of obtaining best result under given constraints is called as

  1. Optimization
  2. Generalization
  3. Summation
  4. Regularization
  

Question 30 : Below are the 8 actual values of the target variable in the train file.[0,0,0,1,1,1,1,1]What is the entropy of the target variable?

  1. -(5/8 log(5/8) + 3/8 log(3/8))
  2. 5/8 log(5/8) + 3/8 log(3/8)
  3. 3/8 log(5/8) + 5/8 log(3/8)
  4. 5/8 log(3/8) – 3/8 log(5/8)
  
  • chevron_left
  • 1
  • 2
  • chevron_right