Data Mining - MCQS 2

Question
This clustering approach initially assumes that each data instance represents a single cluster.

Select one:
a. expectation maximization
b. K-Means clustering
c. agglomerative clustering
d. conceptual clustering

The correct answer is:agglomerative clustering

Question
The correlation coefficient for two real-valued attributes is –0.85. What does this value tell you?

Select one:
a. The attributes are not linearly related.
b. As the value of one attribute decreases the value of the second attribute increases.
c. As the value of one attribute increases the value of the second attribute also increases.
d. The attributes show a linear relationship

The correct answer is: As the value of one attribute decreases the value of the second attribute increases.

Question
Time Complexity of k-means is given by

Select one:
a. O(mn)
b. O(tkn)
c. O(kn)
d. O(t2kn)

The correct answer is: O(tkn)

Question
Given a rule of the form IF X THEN Y, rule confidence is defined as the conditional probability that

Select one:
a. Y is false when X is known to be false.
b. Y is true when X is known to be true.
c. X is true when Y is known to be true
d. X is false when Y is known to be false.

The correct answer is: Y is true when X is known to be true.



Question
Chameleon is

Select one:
a. Density based clustering algorithm
b. Partitioning based algorithm
c. Model based algorithm
d. Hierarchical clustering algorithm

The correct answer is: Hierarchical clustering algorithm

Question
In _________ clusterings, points may belong to multiple clusters

Select one:
a. Non exclusivce
b. Partial
c. Fuzzy
d. Exclusive

The correct answer is: Fuzzy

Question
Find odd man out

Select one:
a. DBSCAN
b. K mean
c. PAM
d. K medoid

The correct answer is: DBSCAN

Question
Which statement is true about the K-Means algorithm?

Select one:
a. The output attribute must be cateogrical.
b. All attribute values must be categorical.
c. All attributes must be numeric
d. Attribute values may be either categorical or numeric

The correct answer is: All attributes must be numeric

Question
This data transformation technique works well when minimum and maximum values for a real-valued attribute are known.

Select one:
a. z-score normalization
b. min-max normalization
c. logarithmic normalization
d. decimal scaling

The correct answer is: min-max normalization

Question
The number of iterations in apriori ___________

Select one:
a. increases with the size of the data
b. decreases with the increase in size of the data
c. increases with the size of the maximum frequent set
d. decreases with increase in size of the maximum frequent set

The correct answer is: increases with the size of the maximum frequent set

Question
Which of the following are interestingness measures for association rules?

Select one:
a. recall
b. lift
c. accuracy
d. compactness

The correct answer is: lift

Question
Which one of the following is not a major strength of the neural network approach?

Select one:
a. Neural network learning algorithms are guaranteed to converge to an optimal solution
b. Neural networks work well with datasets containing noisy data.
c. Neural networks can be used for both supervised learning and unsupervised clustering
d. Neural networks can be used for applications that require a time element to be included in the data

The correct answer is: Neural network learning algorithms are guaranteed to converge to an optimal solution

Question
Given a frequent itemset L, If |L| = k, then there are

Select one:
a. 2k   - 1 candidate association rules
b. 2k   candidate association rules
c. 2k   - 2 candidate association rules
d. 2k -2 candidate association rules

The correct answer is: 2k -2 candidate association rules

Question 
. _________ is an example for case based-learning

Select one:
a. Decision trees
b. Neural networks
c. Genetic algorithm
d. K-nearest neighbor

The correct answer is: K-nearest neighbor

Question 
The average positive difference between computed and desired outcome values.

Select one:
a. mean positive error
b. mean squared error
c. mean absolute error
d. root mean squared error

The correct answer is: mean absolute error

Question 
Frequent item sets is

Select one:
a. Superset of only closed frequent item sets
b. Superset of only maximal frequent item sets
c. Subset of maximal frequent item sets
d. Superset of both closed frequent item sets and maximal frequent item sets

The correct answer is: Superset of both closed frequent item sets and maximal frequent item sets

Question 
1. Assume that we have a dataset containing information about 200 individuals.  A supervised data mining session has discovered the following rule:

                IF  age < 30 & credit card insurance = yes   THEN life insurance = yes
                        Rule Accuracy:    70%   and  Rule Coverage:   63%

How many individuals in the class life insurance= no have credit card insurance and are less than 30 years old?
Select one:
a. 63
b. 30
c. 38
d. 70

The correct answer is: 38

Question
Use the three-class confusion matrix below to answer percent of the instances were correctly classified?

Computed Decision
	Class 1	Class 2	Class 3
Class 1	10	5	3
Class 2	5	15	3
Class 3	2	2	5

Select one:

a. 60

b. 40

c. 50

d. 30

The correct answer is: 60

Question 
Which of the following is cluster analysis?

Select one:
a. Simple segmentation
b. Grouping similar objects
c. Labeled classification
d. Query results grouping

The correct answer is: Grouping similar objects

Question 
A good clustering method will produce high quality clusters with

Select one:
a. high inter class similarity
b. low intra class similarity
c. high intra class similarity
d. no inter class similarity

The correct answer is: high intra class similarity

Question 
Which two parameters are needed for DBSCAN

Select one:
a. Min threshold
b. Min points and eps
c. Min sup and min confidence
d. Number of centroids

The correct answer is: Min points and eps

Question 
Which statement is true about neural network and linear regression models?

Select one:
a. Both techniques build models whose output  is determined by a  linear sum of weighted input attribute values.
b. The output of both models is a categorical attribute value.
c. Both models require numeric attributes to range between 0 and 1.
d. Both models require input attributes to be numeric.

The correct answer is: Both models require input attributes to be numeric.

Question 
In Apriori algorithm, if 1 item-sets are 100, then the number of candidate 2 item-sets are

Select one:
a. 100
b. 4950
c. 200
d. 5000

The correct answer is: 4950

Question 
Significant Bottleneck in the Apriori algorithm is

Select one:
a. Finding frequent itemsets
b. Pruning
c. Candidate generation
d. Number of iterations

The correct answer is: Candidate generation

Question 
The concept of core, border and noise points fall into this category?

Select one:
a. DENCLUE
b. Subspace clustering
c. Grid based
d. DBSCAN

The correct answer is: DBSCAN

Question 
The correlation coefficient for two real-valued attributes is –0.85. What does this value tell you?

Select one:
a. The attributes show a linear relationship
b. The attributes are not linearly related.
c. As the value of one attribute increases the value of the second attribute also increases.
d. As the value of one attribute decreases the value of the second attribute increases.

The correct answer is: As the value of one attribute decreases the value of the second attribute increases.

Question 
Machine learning techniques differ from statistical techniques in that machine learning methods

Select one:
a. are better able to deal with missing and noisy data
b. typically assume an underlying distribution for the data
c. have trouble with large-sized datasets
d. are not able to explain their behavior.

The correct answer is: are better able to deal with missing and noisy data

Question 
The probability of a hypothesis before the presentation of evidence.

Select one:
a. a priori
b. posterior
c. conditional
d. subjective

The correct answer is: a priori

Question 
KDD represents extraction of

Select one:
a. data
b. knowledge
c. rules
d. model

The correct answer is: knowledge

Question 
Which statement about outliers is true?

Select one:
a. Outliers should be part of the training dataset but should not be present in the test data.
b. Outliers should be identified and removed from a dataset.
c. The nature of the problem determines how outliers are used
d. Outliers should be part of the test dataset but should not be present in the training data.

The correct answer is: The nature of the problem determines how outliers are used

Question 
The most general form of distance is

Select one:
a. Manhattan
b. Eucledian
c. Mean
d. Minkowski

The correct answer is: Minkowski

Question 
Arbitrary shaped clusters can be found by using

Select one:
a. Density methods
b. Partitional methods
c. Hierarchical methods
d. Agglomerative

The correct answer is: Density methods

Question 
Which Association Rule would you prefer

Select one:
a. High support and medium confidence
b. High support and low confidence
c. Low support and high confidence
d. Low support and low confidence

The correct answer is: Low support and high confidence

Question 
With Bayes theorem the probability of hypothesis H¾ specified by P(H) ¾ is referred to as

Select one:
a. a conditional probability
b. an a priori probability
c. a bidirectional probability
d. a posterior probability

The correct answer is: an a priori probability

Question 
In a Rule based classifier, If there is a rule for each combination of attribute values, what do you called that rule set R

Select one:
a. Exhaustive
b. Inclusive
c. Comprehensive
d. Mutually exclusive

The correct answer is: Exhaustive

Question
The apriori property means

Select one:
a. If a set cannot pass a test, its supersets will also fail the same test
b. To decrease the efficiency, do level-wise generation of frequent item sets
c. To improve the efficiency, do level-wise generation of frequent item sets
d. If a set can pass a test, its supersets will fail the same test

The correct answer is: If a set cannot pass a test, its supersets will also fail the same test

Question 
Clustering is ___________ and is example of ____________learning

Select one:
a. Predictive and supervised
b. Predictive and unsupervised
c. Descriptive and supervised
d. Descriptive and unsupervised

The correct answer is: Descriptive and unsupervised

Question 
The probability that a person owns a sports car given that they subscribe to automotive magazine is 40%.  We also know that 3% of the adult population subscribes to automotive magazine. The probability of a person owning a sports car given that they don’t subscribe to automotive magazine is 30%.  Use this information to compute the probability that a person subscribes to automotive magazine given that they own a sports car

Select one:
a. 0.0368
b. 0.0396
c. 0.0389
d. 0.0398

The correct answer is: 0.0396

Question 
Simple regression assumes a __________ relationship between the input  attribute and output attribute.

Select one:
a. quadratic
b. inverse
c. linear
d. reciprocal

The correct answer is: linear

Question 
Which of the following algorithm comes under the classification

Select one:
a. Apriori
b. Brute force
c. DBSCAN
d. K-nearest neighbor

The correct answer is: K-nearest neighbor

Question 
Hierarchical agglomerative clustering is typically visualized as?

Select one:
a. Dendrogram
b. Binary trees
c. Block diagram
d. Graph

The correct answer is: Dendrogram

Question 
The _______ step eliminates the extensions of (k-1)-itemsets which are not found to be frequent,from being considered for counting support

Select one:
a. Partitioning
b. Candidate generation
c. Itemset eliminations
d. Pruning

The correct answer is: Pruning

Question 
To determine association rules from frequent item sets

Select one:
a. Only minimum confidence needed
b. Neither support not confidence needed
c. Both minimum support and confidence are needed
d. Minimum support is needed

The correct answer is: Only minimum confidence needed

Question 
What is the final resultant cluster size in Divisive algorithm, which is one of the hierarchical clustering approaches?

Select one:
a. Zero
b. Three
c. singleton
d. Two

The correct answer is: singleton

Question
If {A,B,C,D} is a frequent itemset, candidate rules which is not possible is

Select one:
a. C --> A
b. D -->ABCD
c. A --> BC
d. B --> ADC

The correct answer is: D -->ABCD

Question
Which Association Rule would you prefer

Select one:
a. High support and low confidence
b. Low support and high confidence
c. Low support and low confidence
d. High support and medium confidence

The correct answer is: Low support and high confidence

Question
The probability that a person owns a sports car given that they subscribe to automotive magazine is 40%.  We also know that 3% of the adult population subscribes to automotive magazine. The probability of a person owning a sports car given that they don’t subscribe to automotive magazine is 30%.  Use this information to compute the probability that a person subscribes to automotive magazine given that they own a sports car

Select one:

a. 0.0398

b. 0.0389

c. 0.0368

d. 0.0396

The correct answer is: 0.0396

Question
This  clustering algorithm terminates when mean values computed for the current iteration of the algorithm are identical to the computed mean values for the previous iteration

Select one:
a. conceptual clustering
b. K-Means clustering
c. expectation maximization
d. agglomerative clustering

The correct answer is: K-Means clustering

Question
The distance between two points calculated using Pythagoras theorem is

Select one:
a. Supremum distance
b. Eucledian distance
c. Linear distance
d. Manhattan Distance

The correct answer is: Eucledian distance

Question
Classification rules are extracted from _____________

Select one:
a. decision tree
b. root node
c. branches
d. siblings

The correct answer is: decision tree

Question
What does K refers in the K-Means algorithm which is a non-hierarchical clustering approach?

Select one:
a.Fixed value
b.No of iterations
c.Complexity

d.number of clusters

The correct answer is: number of clusters

Question
Which is not part of the categories of clustering methods?

Select one:
a.Hierarchical methods
b.Density based methods
c.Partitioning methods
d.Rule-based methods

The correct answer is: Rule-based methods

Question
Given desired class C and population P, lift is defined as

Select one:
a.the probability of class C given population P divided by the probability of C given a sample taken from the population
b.the probability of class C given a sample taken from population P.
c.the probability of class C given a sample taken from population P divided by the probability of C within the entire population P.
d.the probability of  population P given a sample taken from P

The correct answer is: the probability of class C given a sample taken from population P divided by the probability of C within the entire population P

Question

If  an item set ‘XYZ’ is a frequent item set, then all subsets of that frequent item set are

Select one:
a. Frequent
b. Not frequent
c. Can not say
d. Undefined

The correct answer is: Frequent

Question
With Bayes theorem the probability of hypothesis H¾ specified by P(H) ¾ is referred to as

Select one:
a. a posterior probability
b. a bidirectional probability
c. a conditional probability
d. an a priori probability

The correct answer is: an a priori probability