gjIHDE$%^&*(*UYHBN BT%$ERDXSW#$%^&*()OPLKNBVFT^%ESDCVB>:POIH545

Naïve Bayes algorithm is based on ______.

• Bayes Theorem

• Candidate elimination algorithm

• EM algorithm

• Nove of these

How do we perform bayesian classification when some features are missing ?

• We assume the missing values as the mean of all values
• We ignore the missing features
• We integrate the posteriors probalities over the missing features
• Drop the features completely

Which is suitable metric for categorical variable to find similarity ?

• Euclidiean distance

• Manhattan distance

• Minkowski distance

• Hamming distance

Which is a decision support tool that uses a tree-like graph Or model of decision and their possible consequences including chance event outcomes, resources cost, and utility ?

• Decision Tree

• Graphs

• Trees

• Neural Networks

What happens to the entropy of a set/subset when we traverse from root towards the child node in a decision tree ?

• Increases towards child node
• Decreases towards child node
• Can not be decided
• depends on dataset

How do you choose the suitable node for selection of root of the tree/sub-tree while constructing a decision tree ?

• An attribute having high entropy

• An attribute having high entropy and information gain

• An attribute having lowest information gain

• An attribute having highest information gain

Which of the following techniques could perform better for reducing dimensions of the dataset ?

• Removing columns which have too many missing values

• Removing columns having high variance in data

• Removing columns having similar data trends

• None of these

What will happen when eigen values are roughly equal ?

• PCA will perform outstandingly

• PCA will perform badly

• Can not be decided

• For large dataset PCA will perform better

Which clustering algorithm is not an aglomverative approach of clustering?

• Complete Linkage clustering

• nearest neighbour clustering

• Single Linkage clustering algorithm

Which is not a good idea to select features in a dataset?

• Selecting features with low variance.

• Not selecting features having many missing values.

• applying principal component analysis and selecting main components.

• Selecting features with low co-relation filter.