Engineering B.E. (Computer Engineering) Sem 8 Machine Learning Dec 2019 Question Paper - Mumbai University | munotes

Questions asked in this paper

1. Q2 Answer any three out of remaining five questions
2. Q3 Assume any suitable data wherever required and justify the same
3. Q1 a) Explain the Expectation Maximization Algorithm (EMA) 5 marks
   • b) Explain kernel functions and kernel trick 5
   • c) What are the issues in decision tree learning? 5
   • d) “Entropy is a thermodynamic function used to measure the disorder of a system in Chemistry.” Clarify the concept of entropy in Machine Learning? 5
4. Q2 a) Compare and contrast Linear and Logistic regressions with respect to their mechanisms of prediction 10 marks
   • b) Consider 2-D dataset given in the table below. Construct a SVM classifier model. Given (2, 1), (2, -1) and (4, 0) as support vectors; estimate the parameters of the model and classify (4, 2). Why is SVM called as optimal binary hyper plane 10
5. Q3 a) You are given a data set on cancer detection. You have built a classification model and achieved an accuracy of 96%. Why shouldn’t you be happy with your model performance? What can you do about it? 10 marks
   • b) What is a HMM? What are the issues in Hidden Markov Model (HMM)? 10
6. Q4 You came to know that your model is suffering from low bias and high variance. Which algorithm should you use to tackle it? Why? 10 marks
   • b) Differentiate between simple linkage, average linkage and complete linkage algorithms. Use complete linkage algorithm to find the clusters from the following 10
7. Q5 a) Draw the block diagram of Error Back Propagation Algorithm and explain with flow chart the concept of Back Propagation 10 marks
   • b) The following table consists of training data from an employee database. The data have been generalized. For example, for age represents the age range of 31 to 35. For a given row entry, count represents the number of data tuples having the values for department, status, age, and salary given in that row. Let the status be the class-label attribute 10
   • (i) Design a multilayer feed-forward neural network for the given data. Label the nodes in the input and output layers
   • (ii) Using the multilayer feed-forward neural network obtained in (1), show the weight values after one iteration of the back propagation algorithm, given the Assume initial weight values and biases. Assume learning rate to be 0.9. Use binary input and draw (one input layer, one output layer and one hidden layer) neural network. Solve the problem for one epoch
8. Q6 Write short notes on any two of the following: 20 marks
   • a) Temporal Difference Learning in Reinforcement Learning Over fitting’ in Machine learning

Read from the scan above, so a character or two may differ. The scan is the original.