BITS WILP Data Mining Quiz-2 2017-H2



      Question 1
Hierarchical agglomerative clustering can be stopped at a certain point
if we want to create n (n < data-points) number of clusters.


Select one:
True
False


        Feedback

The correct answer is 'True'.


      Question 2

k-means algorithm always converges

Select one:
True
False


        Feedback

The correct answer is 'False'.


      Question 3


K-means clustering requires that we give it the number of clusters we
want to be created.

Select one:
True
False


        Feedback

The correct answer is 'True'.


      Question 4


Nested cluster diagram can be used to represent k-means also.

Select one:
True
False


        Feedback

The correct answer is 'False'.


      Question 5


Analysis of patterns for stock market prediction is an example of:

Select one:
evolution analysis
outlier analysis
correlation
characterization and discrimination


        Feedback

The correct answer is: evolution analysis


      Question 6

Given a single conditional FP-tree ending in a given frequent 1-itemset,
we can find all the other frequent itemsets ending in any frequent itemset.

Select one:
True
False


        Feedback

The correct answer is 'False'.


      Question 7


If k is fixed in k-means we always get same clusters of points.

Select one:
True
False

        Feedback

The correct answer is 'False'.


      Question 8

FP-growth is fast due to the hash tree structure that is used to store
candidate itemsets.


Select one:
True
False


        Feedback

The correct answer is 'False'.


      Question 9


A supermarket database has 2000 transactions, out of which 40 include
both items A and B, and 16 out of these 40 also contain item C. In
total, 100 transactions contain C.

The Lift of association rule "If A and B are purchased, then C is
purchased in the same transaction" is:

Select one:
2
8
4
16


        Feedback

The correct answer is: 8


      Question 10

FP-growth has an advantage over apriori algorithm that:

Select one:
a. FP-tree is a compact structure whereas apriori tree is complex
b. FP-growth can find all frequent itemsets
c. FP-growth works by just creating a binary tree structure
d. no candidate itemsets need to be generated


        Feedback

The correct answer is: no candidate itemsets need to be generated


      Question 11

Number of groups is known apriori in:

Select one:
a. classification
b. outlier
c. clustering
d. preprocessing


        Feedback

The correct answer is: classification


      Question 12

In association analysis, support is a symmetric measure of associations.

Select one:
True
False


        Feedback

The correct answer is 'True'.


      Question 13


If in random sub-sampling repetition k=1, it is same as:

Select one:
a. stratified subsampling
b. bootstrap
c. k-fold cross-validation
d. holdout


        Feedback

The correct answer is: holdout


      Question 14


In 10-fold cross validation, sample data is partitioned into 10 equal
subsamples. Of the 10 subsamples, a single subsample is retained as the
validation data for testing the model, and the remaining 9 subsamples
are used as training data. This is done exactly once and the result of
testing is final. True or False?


Select one:
True
False


        Feedback

The correct answer is 'False'.


      Question 15

 

The table below shows marks in math (x) and marks in statistics (y).

What is the value of the slope (m) of simple regression line ?

Select one:
a. 0.744
b. 0.444
c. 0.644
d. 0.544


        Feedback

The correct answer is: 0.644


      Question 16

The following data is about a poll that occurred in 3 states. In state1,
50% of voters support Party1, in state2, 60% of the voters support
Party1, and in state3, 35% of the voters support Party1. Of the total
population of the three states, 40% live in state1, 25% live in state2,
and 35% live in state3. Given that a voter supports Party1, what is the
probability that he lives in state2?


Select one:
0.62
0.32
0.52
0.42


        Feedback

The correct answer is: 0.32


      Question 17

Single link or Complete link in hierarchical clustering is used to
determine the inter-cluster distance. The clustering algorithm always
selects the nearest clusters to merge.

Select one:
True
False


        Feedback

The correct answer is 'True'.


      Question 18
In classification, we evaluate the performance of a classifier on
training data


Select one:
True
False


        Feedback

The correct answer is 'False'.


      Question 19

The following a valid FP-tree: True or False?

 

Select one:
True
False


        Feedback

The correct answer is 'False'.


      Question 20

Once we get all frequent itemsets using a method such as like apriori,
we can use the support counts to generate all strong association rules.
Select one:
True
False


        Feedback

The correct answer is 'False'.



No comments:

Post a Comment