100 Data mining and data warehousing multiple choice questions with answers Pdf for the preparation of academic and competitive IT exams.
100 Data Mining and Data Warehousing MCQs in Pdf
1. A priori algorithm operates in ___ method
a. Bottom-up search method
b. Breadth-first search method
c. None of above
d. Both a & b
2. A bi-directional search takes advantage of ___ process
a. Bottom-up process
b. Top-down process
c. None
d. Both a & b
3. The pincer-search has an advantage over a priori algorithm when the largest frequent item set is long.
a. True
b. false
4. MCFS stand for
a. Maximum Frequent Candidate Set
b. Minimal Frequent Candidate Set
c. None of above
5. MFCS helps in pruning the candidate set
a. True
b. False
6. DIC algorithm stands for ___
a. Dynamic itemset counting algorithm
b. Dynamic itself counting algorithm
c. Dynamic item set countless algorithms
d. None of above
7. If the item set is in a dashed circle while completing a full pass it moves towards
a. Dashed circle
b. Dashed box
c. Solid Box
d. Solid circle
8. If the item set is in the dashed box then it moves into a solid box after completing a full pass
a. True
b. False
9. The dashed arrow indicates the movement of the item set
a. True
b. False
10. The vertical arrow indicates the movement of the item set after reaching the frequency threshold
a. True
b. False
11. Frequent set properties are:
a. Downward closure property
b. Upward closure property
c. A & B
d. None of these
12. Any subset of a frequent set is a frequent set is
A. Downward closure property
B. Upward closure property
C. A and b
13. Periodic maintenance of a data mart means
a. Loading
b. Refreshing
c. Purging
d. All are true
14. The Fp-tree Growth algorithm was proposed by
a. Srikant
b. Aggrawal
c. Hanetal
d. None of these
15. The main idea of the algorithm is to maintain a frequent pattern tree of the date set. An extended prefix tree structure starting crucial and quantitative information about frequent sets
a. Priori Algorithm
b. Pinchers Algorithm
c. FP- Tree Growth algo.
d. All of these
16. The data warehousing and data mining technologies have extensive potential applications in the govt in various central govt sectors such as :
a. Agriculture
b. Rural Development
c. Health and Energy
d. all of the true
17. ODS Stands for
a. External operational data sources
b. operational data source
c. output data source
d. none of the above
18. Good performance can be achieved in a data mart environment by extensive use of
a. Indexes
b. creating profile records
c. volumes of data
d. all of the above
19. Features of Fp tree are
(i). It is dependent on the support threshold
(ii). It depends on the ordering of the items
(iii). It depends on the different values of trees
(iv). It depends on frequent itemsets with respect to give information
a. (i) & (ii)
b. (iii) & (iv)
c. (i) & (iii)
d. (ii) only
20. For a list T, we denote head_t as its first element and body-t as the remaining part of the list (the portion of the list T often removal of head_t) thus t is
a. {head} {body}
b. {head_t} {body_t}
c. {t_head}{t_body}
d. None of these
21. Partition Algorithm executes in
a. One phase
b. Two-Phase
c. Three phase
d. None of these
22. In the First Phase of the Partition Algorithm
a. Logically divides into a number of non-overlapping partitions
b. Logically divides into a number of overlapping Partitions
c. Not divides into partitions
d. Divides into non-logically and non-overlapping Partitions
23. Functions of the second phase of the partition algorithm are
a. Actual support of item sets are generated
b. Frequent itemsets are identified
c. Both (a) & (b)
d. None of these
24. Partition algorithm is based on the
a. Size of the global Candidate set
b. Size of the local Candidate set
c. Size of frequent itemsets
d. No. Of item sets
25. Pincer search algorithm based on the principle of
a. Bottom-up
b. Top-Down
c. Directional
d. Bi-Directional
26. Pincer-Search Method Algorithm contains
(i) Frequent item set in a bottom-up manner
(ii) Recovery procedure to recover candidates
(iii) List of maximal frequent itemsets
(iv) Generate a number of partitions
a. (i) only
b. (i) & (iii) only
c. (i),(iii) & (iv)
d. (i),(ii)&(iii)
27. Is a full-breadth search, where no background knowledge of frequent itemsets is used for pruning?
a. Level-crises filtering by the single item
b. Level-by-level independent
c. Multi-level mining with uniform support
d. Multi-level mining with reduced support
28. Disadvantage of uniform support is
a. Items at lower levels of abstraction will occur as frequently.
b. If the minimum support threshold is set too high, I could miss several meaningful associations
c. Both (a) & (b)
d. None of these
29.Warehouse administrator responsible for
a. Administrator
b. maintenance
c. both a and b
d. none of the above
30. The pincer-search has an advantage over a priori algorithm when the largest frequent itemset is long
a. True
b. false
31. What are the common approaches to tree pruning?
a. Prepruning and Postpruning approach.
b. Prepruning.
c. Postpruning.
d. None of the above.
32. Tree pruning methods address this problem of ___?
a. Overfitting the branches
b. Overfitting the data
c. a and b both
d. None of the above
33. What is the Full Form of MDL.
a. Maximum Description Length
b. Minimum Description Length
c. Mean Described Length
d. Minimum Described Length
34. State that the Statements are True / False:
a. Post pruning approach Removes Branches from a ‘Fully Grown’ Tree.
a. True
b. False
b. The “Best Pruned Tree is the one that maximizes the number of encoding bits.
a. True
b. False
35. Upon halting, the node becomes a ___
A. Heap
B. Subset
C. Leaf
D. Superset
36. demographic and neural clustering are methods of clustering based on
a. data types
b. methodology of calculation
c. Inter record distance
d. all of the above
37. POS stands for
a. Peer of sale
b. Point of sale
c. part of the sale
d. none of the above
38. Classification and Prediction are two forms of
a. Data analysis
b. Decision Tree
c. A and B
d. None of these
39. Classification predicts
a. Categorical labels
b. Prediction models continued valued function
c. A and B
d. None of these
40. True / False
a. Each Tuple is assumed to belong to a predefined class as determined by one of the attributes, called the class label attribute.
b. The individual tuples making up the training set are referred to as the training data set.
c. Classification and Regression are the two major type of data analysis.
Ans. A-True, B-True, C-False
41.True / False
a. Classification and Regression are the two major type of data analysis.
b. Classification is used to predict discrete or nominal values.
c. Regression is used to predict continuous or ordered values.
d. All are true
42. Classification and Prediction have numerous applications:
a. Credit approval
b. Medical diagnosis
c. Performance prediction & selective marketing
d. All of these
43. Class label of each training sample is provided with this step is known as
a. Unsupervised learning
b. Supervised learning
c. Training samples
d. Clustering
44. Decision tree is based on
a. Bottom-down technique
b. Top-down technique
c. Divide-and-conquer manner
d. Top-down recursive divide-and-conquer manner
45. Recursive Partitioning stops in Decision Tree when
a. All samples for a given node belong to the same class.
b. There are no remaining attributes on which samples may be further partitioned.
c. There are no samples for the branch test.
d. All the above.
46. To select the test attribute of each node in a decision tree we use
a. Entity Selection Measure
b. Data Selection Measure
c. Information Gain Measure
d. None of these
47.Test attribute for the current node in the decision tree is chosen on the basis of
a. Lowest entity gain
b. Highest data gain
c. Highest Information Gain
d. Lowest Attribute Gain
48. Advantage of the Information-theoretic approach of the decision tree is
a. Minimizes the expected number of tests needed
b. Minimizes the number of Nodes
c. Maximizes the number of nodes
d. Maximizes the number of tests
49. Let us be the no. of samples of S in class Ci then expected information to classify a given sample is given by
a. L(s1,s2,……..sm)=_log2(pi)
b. L(s1,s2,……..sm)=-_pilog2(pi)
c. L(s1,s2,……..sm)=_pilog2x
d. L(s1,s2,……..sm)=_pilog2(pi)
50. Steps applied to the data in order to improve the accuracy, efficiency, and scalability are:-
a. Data cleaning
b. Relevance analysis
c. Data transformation
d. All of the above
51. The process used to remove or reduce noise and the treatment of missing values
a. Data cleaning
b. Relevance analysis
c. Data transformation
d. None of above
52. Relevance analysis may be performed on the data by removing any irrelevant attribute from the process.
a. True
b. False
53. Classification and prediction method can be affected by:-
a. Accuracy & Speed
b. Robustness & Scalability
c. Interpretability
d. All of the above
54. In a decision tree internal node denotes a test on an attribute and Leaf nodes represent classes or class distributions
a. True
b. false
55. ___ attempts to identify and remove branches, with Improving accuracy
a. decision tree
b. tree pruning
c. both of them
d. none of above
56. To deal with larger data sets, a sampling method, called ___
a. Clara
b. Dara
c. Pam
d. None
57. What is the Full Form of CLARA.
a. Clustering Large Applicant
b. Close Large Applicant
c. Clustering Large Applications
d. None of the above
58. What is the Full Form of CLARANS.
a. Clustering Large Applications Based Upon Randomized Search
b. Close Large Applicant Based Upon Role Search
c. Clustering Large Applicant Based Upon Randomized Search
d. None of the above
59. Which Algorithm was proposed that combines the Sapling Technique with PAM.
a. CLARA
b. CLARANS
c. Both a and b
d. None of these.
60. Which are the two type of Hierarchical Clustering?
a. Agglomerative Hierarchical Clustering and Density Hierarchical Clustering
b. Agglomerative Hierarchical Clustering and Divisive Hierarchical Clustering
c. Divisive Hierarchical Clustering and Density Hierarchical Clustering
d. None of the above
61. Cluster is a :
a. The process of grouping a set of physical or abstract objects into classes of similar objects is called clustering.
b. A cluster of data objects can be treated collectively as one group in many applications
c. Cluster analysis is an important human activity.
d. All of the above
62. Cluster analysis tools based on
a. K-means
b. K-medosis
c. A and B
d. None of these
63. S-Plus, SPSS, SAS software packages use for
a. Data Mining
b. Classification
c. Clustering
d. Prediction
64. Unsupervised learning is an example of
a. Classification and prediction
b. Classification and Regression
c. clustering
d. Data Mining
65. Requirement of Clustering in Data Mining
a. Scalability
b. Ability to deal with different types of attributes
c. Ability to deal with noisy data
d. Discovery of clusters with arbitrary shape
e. Minimal requirement for domain knowledge to determine input parameters
f. Insensitivity to the order of input records
g. High dimensionality
h. Constraint-based clustering
(a). a, c, d, f
(b). g, h
(c). All of these
(d.) None of these
66. Clustering method can be classified
a. Partitioning Methods
b. Hierarchical methods
c. Density-based methods
d. All of these
67. Hierarchical methods can be classified
a. Agglomerative Approach
b. Divisive Approach
c. A and B
d. None of these
68. Agglomerative approach is called as
a. Bottom-up Approach
b. Top-Down Approach
c. A and B
d. None of these
69. Top-Down Approach is
a. Agglomerative Approach
b. Divisive Approach
70. Drawback of Hierarchical Methods
a. Suffer from the fact that once a step is done, it can never be undone.
b. A technique is that they cannot correct erroneous decision.
c. Both a & b
d. None of these
71. Two approaches to improving the quality of hierarchical clustering:
a. Perform careful analysis of object “linkages” at each hierarchical partitioning, such as in CURE and Chameleon
b. Integrate Hierarchical agglomeration and iterative relocation by first using a hierarchical agglomerative algorithm and refining the result using an iterative relocation
c. Both a & b
d. None of these
72. Classical Portioning methods are
a. k-means and k-median
b. k-means and k-medoids
c. k-modes only
d. none of these
73. K-means technique is based on
a. Centroid Object
b. Reference object
c. Representative object
d. Partition Object
74. K-medoids technique is based on
a. Centroid Object
b. Representative object
c. Partition Object
d. None of these
75. The k-means and the k-modes methods can be integrated to cluster data with mixed numeric and categorical values, resulting in
a. k-median method
b. k-partition method
c. k-prototypes method
d. k-medoids method
76. The squared-error criterion is used in a k-means method defined as
a. E=_I=1tok _pεci [p-mi]
b. E=_I=1tok _pεci [mi]2
c. E=_I=1tok _pεci [p]2
d. E=_I=1tok _pεci [p-mi]2
77. The Computational Complexity of the k-means method algorithm is
a. O(log x)
b. Θ(nkt)
c. O(nkt)
d. Θ(log x)
78. Which Method is more Robust-k-means or k-medoids?
a. The k means is more robust in the presence of noise
b. The k-medoids method is more robust in the presence of noise and outliers
c. The k-medoids method is more robust due to no. of partitions
d. The k means is more robust due to its less complexity
79. First k-medoids algorithm introduced is
a. Prototype Above Medoids
b. Partition Below Medoids
c. Prototype Around Medoids
d. Partitioning Around Medoids
80. PAM stands for
a. Prototype Above Medoids
b. Prototype Around Means
c. Partitioning Around Medoids
d. Partitioning Above Means
81.Which statements are true fork-means
(i). It can apply only when the mean of the cluster is defined.
(ii). It is not suitable for discovering clusters with non-convex shapes
(iii). This method is relatively efficient in processing only small data.
a. (i) only
b. (i) & (ii) only
c. (iii) only
d. All the above
82 DBSCAN stands for:
a. Divisive Based Clustering Method
b. Density-Based Clustering Method
c. Both a & b
d. None of above
83: DBSCAN defines a cluster as a maximal set of density –
Connected points
a. True
b. False
84: For a non-negative value ε,Ne(Oi)={ Oj ∈D I d(Oi,Oj)≤ ε}
a. True
b. false
85. The ___ client is a desktop that relies on the server to which it is connected for the majority of its computing power.
a. thin
b. none
c. thick
d. web server
86. An object is said to be the Core Object if
a._ Ne(O)_ ≥ MinPts
b. _ N (O)_ ≥ MaxPts
c. none of above
d. both a & b
87. The density-reachability relation is transitive but not symmetric
a. True
b. False
88. Non-core objects are:-
a. border object
b. noise object
c. non-object
d. both a & b
89. DBSCAN algorithm can classify into:
a. classified
b. unclassified
c. noise
d. all of above
90. Unsupervised learning is an example of
a. Classification and prediction
b. Classification and Regression
c. clustering
d. Data Mining
91. Data can be classified as
a. reference data
b. transaction data and derived data
c. derived data
d. all of the above
92. Reference and transaction data originates from
a. operational system
b. Unnormalized data
c. data marts
d. all are true
93. Derived data is derived from
a. reference data
b. transaction data
c. reference and transaction data
d. none of the above
94. Unnormalized data, which is the basis for online analytical processing tools are prepared periodically but is directly based on detailed ___.
a. reference data
b. transaction data
c. reference and transaction data
d. none of the above
95. The data mart is loaded with data from a data warehouse by means of a ___
a. load program
b. process
c. project
d. all is valid
96. The chief considerations for a Load program are:
a. frequency and schedule
b. total or partial refreshment
c. customization and re-sequencing
d. all are true
97. Periodic maintenance of a data mart means
a. all are true
b. loading
c. refreshing
d. purging
98. Detailed level data, summary level, preprocessed and Adhoc data are data in
a. data warehouse
b. data mart
c. both
d. none of the above
99. Data sources in the data warehouse are referred to as
a. External data source
b. Operational data source
c. External operational data source
d. none of the above
100. ___ Table help and enable the end-users of the data mart to relate the data to its expanded version.
a. data
b. reference
c. both a and b
d. none of the above
Best MCQs for Advanced Computer Network and Security
Thanks
. most useful mcq
Hey, I think your website might be having browser
compatibility issues. When I look at your blog in Chrome, it looks fine but when opening in Internet Explorer, it has some overlapping.
I just wanted to give you a quick heads up! Other then that, superb blog!
It’s not my first time to visit this web page, i am browsing this
site dailly and take pleasant information from here all the
time.
Thankfulness to my father who informed me concerning this
blog, this weblog is truly remarkable.
There is certainly a lot to know about this subject.
I love all the points you’ve made.
Good post. I learn something totally new and challenging on websites
I stumbleupon everyday. It will always be useful to read articles from other authors and use something from their websites.
WOW just what I was searching for. Came here by searching for website
Very quickly this web page will be famous among all
blog people, due to it’s nice articles
You ought to be a part of a contest for one of the greatest blogs online.
I will highly recommend this web site!
Awesome! Its really remarkable post, I have got much clear idea on the
topic of from this article.
keep doing a great job
Love this dude
Outstanding article Michele! You’re such an awesome teacher and we
all appreciate and love you.
Thanks for sharing.