04M each+02=10 4 What is the use of Hyperplane and how to calculate distances with example 03+01+03=07 5 Briefly explain Discriminate analysis, list out Two criteria are used by LDA to create a new axis 03+02=05 6 List and explain two criteria are used by Normal Discriminant Analysis or Linear Discriminant Analysis or Discriminant Function Analysis to create a new axis with proper diagram. 02+04+02=08 7 How do you Decide Positive and negative hyperplane the way you draw Maximum margin Hyperplane with example Provide an explanation. 02+02+02+02=0 8 8 Write any two applications for discriminant function analysis. 02M each=04 9 Outline the teams 1) Hidden layer 2) Baias 3) Threshold function 4) activation function 5) forward propagation 6) Back propagation 7) Neurons 02M each 10 What are the pros & cons of K-NN? Why is it called Lazy Learner? 04+02=06 11 Consider the dataset with height & weight variables, with classes – Normal & Overweight, and classify the individual with (Weight – 57 kg, Height -170 cm) as normal or overweight. 07 12 Apply K– NN and classify the cloth to be manufactured by a textile industry, with Durability– 3 & Strength – 7, with reference of given table – 05 13 Determine the fruit type with Sweetness – 7 & Sourness – 6, with the help of given dataset 07
14 Write Short note on K- Nearest Neighbor with algorithm. 05 15 Define Naïve Bayes’ and also Write Naïve Bayes’ Classification Algorithm 02+04=06 16 What are the pros & cons of Naïve Bayes’ Classification Algorithm? 3M each=06 17 State the applications of Naïve Bayes’ Classification Algorithm 5 18 Predict the probability of an individual of being sanctioned loan from a bank based on the classification in to risk class, with the probability A>>B>>C, when, credit rating is A, Employability is Y and Own home is Y. Consider the following dataset – 08 19 Write short notes on neural network and types of neural network 02+03=05 20 Distinguish between Euclidean Distance and Manhattan Distances. value=(5,1) and (4,1) find out both Euclidean and Manhattan calculations. (or) Explain the mechanism of the following 1) Euclidean Distance2) Manhattan Distances. 2M each+ calculation each 1M=06 21 Write short notes on the following: 1) Linear regression 2) Non-linear regression. 10M 22 Find the forecast value of sales for 2019, using simple regression method, with given actual values to be 10000, 20000, 25000, 35000, and 40000 for 2014, 2015, 2016, 2017 & 2018 respectively.
You've reached the end of your free preview.
Want to read all 7 pages?
- Winter '17
- Machine Learning