DEV Community

Cover image for MAKAUT B.Tech Data Warehousing and Data Mining MCQs
ANIRUDDHA ADAK
ANIRUDDHA ADAK

Posted on

MAKAUT B.Tech Data Warehousing and Data Mining MCQs

#discuss #database #datascience #tutorial

Multiple Choice Questions (Ordered by Frequency)

1. Question: A data warehouse is said to contain a 'time-varying' collection of data because

  • Options:
    • a) Its content vary automatically with time
    • b) Its life-span is very limited
    • c) Every key structure of data warehouse contains either implicitly or explicitly an element of time
    • d) Its content has explicit time-stamp
  • WBUT Years: 2010, 2013, 2014, 2015, 2016, 2017
  • Answer: (c)
  • Explanation: > "Time-varying implies that data in the warehouse includes historical context, often through explicit or implicit time elements in its structure."

2. Question: Which of the following techniques are appropriate for data warehousing?

  • Options:
    • a) Hashing on primary keys
    • b) Indexing on foreign keys of the fact table
    • c) Bit-map indexing
    • d) Join indexing
  • WBUT Years: 2009, 2013, 2018
  • Answer: (c)
  • Explanation: > "Bit-map indexing is efficient for queries on low-cardinality columns, which are common in dimensional data warehouses."

3. Question: A drill-down operation is concerned with

  • Options:
    • a) which merges cells of two dimension
    • b) which merges cells of any one dimension based on the characteristics of the dimension
    • c) which splits cells of two dimensions
    • d) which splits cells of any one dimension based on the characteristics of the dimension
  • WBUT Years: 2009, 2016, 2018
  • Answer: (d)
  • Explanation: > "Drill-down allows users to navigate from summary data to more detailed data by adding new dimensions or stepping down a hierarchy."

4. Question: K-means is based on

  • Options:
    • a) Euclidian distance
    • b) Hamming distance
    • c) RMS
    • d) None of these
  • WBUT Years: 2011, 2014, 2015
  • Answer: (a)
  • Explanation: > "K-means clustering uses Euclidean distance to measure the similarity between data points and cluster centroids."

5. Question: Data Warehousing is used for

  • Options:
    • a) Decision Support System
    • b) OLTP applications
    • c) Database applications
    • d) Data Manipulation applications
  • WBUT Years: 2010, 2012, 2016
  • Answer: (a)
  • Explanation: > "Data Warehousing is primarily designed to support decision-making and analytical processing."

6. Question: A data warehouse is an integrated collection of data

  • Options:
    • a) It is a collection of data of different types
    • b) It is a collection of data derived from multiple sources
    • c) It is a relational database
    • d) It contains summarized data
  • WBUT Years: 2009, 2015
  • Answer: (b)
  • Explanation: > "Data warehouses integrate data from various, often disparate, operational sources into a unified system."

7. Question: A data warehouse is said to contain a 'subject oriented' collection of data because

  • Options:
    • a) Its contents have a common theme
    • b) It is built for a specific application
    • c) It cannot support multiple subjects
    • d) It is a generalization of 'object-oriented'
  • WBUT Years: 2009, 2013
  • Answer: (a)
  • Explanation: > "Subject-oriented means data is organized around core business subjects rather than specific applications."

8. Question: Which of the following is TRUE?

  • Options:
    • a) Data warehouse can be used for analytical processing only
    • b) Data warehouse can be used for information processing (query, report) and analytical processing
    • c) Data warehouse can be used for data mining only
    • d) Data warehouse can be used for information processing (query, report), analytical processing and data mining
  • WBUT Years: 2010, 2012
  • Answer: (d)
  • Explanation: > "A data warehouse supports a wide range of analytical activities including querying, reporting, OLAP, and data mining."

9. Question: A data warehouse is built as a separate repository of data, different from the operational data of an enterprise because

  • Options:
    • a) It is necessary to keep the operational data free of any warehouse operations
    • b) A data warehouse cannot afford to allow corrupted data within it
    • c) A data warehouse contains summarized data whereas the operational database contains transactional data
    • d) None of these
  • WBUT Years: 2012, 2013
  • Answer: (c)
  • Explanation: > "Data warehouses store summarized and historical data for analysis, unlike operational databases which focus on transactional processing."

10. Question: Dimension data within a warehouse exhibits which one of the following properties?

  • Options:
    • a) Dimension data consists of the minor part of the warehouse
    • b) The aggregated information is actually dimension data
    • c) It contains historical data
    • d) Dimension data is the information that is used to analyze the elemental transaction
  • WBUT Years: 2012, 2015
  • Answer: (b)
  • Explanation: > "Dimension data provides the descriptive context and hierarchies through which aggregated facts are analyzed."

11. Question: The important aspect of the data warehouse environment is that data found within the data warehouse is

  • Options:
    • a) subject-oriented
    • b) time-variant
    • c) integrated
    • d) all of these
  • WBUT Years: 2016, 2018
  • Answer: (a)
  • Explanation: > "Subject-orientation is a core characteristic of a data warehouse, organizing data around business subjects."

12. Question: ...... is an example of predictive type of data mining whereas ...... is an example . of descriptive type of data mining.

  • Options:
    • a) Association Rule, Clustering
    • b) Association Rule, Classification
    • c) Classification, Clustering
    • d) Clustering, Classification
  • WBUT Years: 2010, 2012
  • Answer: (c)
  • Explanation: > "Classification predicts a target variable, making it predictive, while clustering discovers patterns without a target, making it descriptive."

13. Question: The 'Dice' operation is concerned with

  • Options:
    • a) Multiple runs of slice
    • b) slice on more than one dimension
    • c) selecting certain cells of more than one dimension
    • d) two consecutive slice operations in two different dimensions
  • WBUT Years: 2009, 2014
  • Answer: (d)
  • Explanation: > "The dice operation filters data on multiple dimensions, effectively creating a subcube."

14. Question: The major drawback of CLARANS algorithms is

  • Options:
    • a) it cannot handle very large volumes of data
    • b) it assumes that all objects fit into the main memory, and the result is very sensitive to input order
    • c) it cannot find the best clustering if any sampled medoit is not among the best k methods
    • d) None of these
  • WBUT Years: 2009, 2011
  • Answer: (b)
  • Explanation: > "A limitation of CLARANS is its sensitivity to the data input order and its memory-intensive nature for very large datasets."

15. Question: Parameters used for association Rule Mining are

  • Options:
    • a) Confidence and Support
    • b) Confidence and Itemcount
    • c) Support and Itemcount
    • d) Support, Confidence and Itemcount
  • WBUT Years: 2010, 2018
  • Answer: (a)
  • Explanation: > "Support and Confidence are the primary metrics used to evaluate the strength and interestingness of association rules."

16. Question: Two main types of clustering techniques in data mining are

  • Options:
    • a) Serial clustering and parallel clustering
    • b) Hierarchical clustering and partitioning clustering
    • c) Homogeneous clustering and heterogeneous clustering
    • d) k-medoids clustering and K-means clustering
  • WBUT Years: 2010, 2018
  • Answer: (b)
  • Explanation: > "Clustering methods are broadly categorized into hierarchical (building a tree of clusters) and partitioning (dividing into non-overlapping groups) approaches."

17. Question: Which one is not a data mining task?

  • Options:
    • a) indexing
    • b) classification
    • c) clustering
    • d) regression
  • WBUT Years: 2014, 2015
  • Answer: (a)
  • Explanation: > "Indexing is a database optimization technique, not a fundamental data mining task like classification, clustering, or regression."

18. Question: An example of hierarchical clustering algorithm is

  • Options:
    • a) clarans
    • b) C4.5
    • c) average linkage
    • d) rock
  • WBUT Years: 2014, 2018
  • Answer: (d)
  • Explanation: > "ROCK (Robust Clustering using links) is a hierarchical clustering algorithm for categorical data."

19. Question: The mining activity which mines web log records to discover user access patterns of web pages is

  • Options:
    • a) web content mining
    • b) web usage mining
    • c) web structure mining
    • d) web search mining
  • WBUT Years: 2011, 2014
  • Answer: (b)
  • Explanation: > "Web usage mining analyzes user behavior patterns from web server logs and clickstreams."

20. Question: Data warehouse architecture is just an over guideline. It is not a blueprint for the data warehouse

  • Options:
    • a) True
    • b) False
  • WBUT Years: 2011
  • Answer: (b)
  • Explanation: > "A data warehouse architecture provides a structured blueprint and a clear framework for its design and implementation."

21. Question: The most distinguishing characteristic of DSS data is

  • Options:
    • a) Granularity
    • b) Timespan
    • c) Dimensionality
    • d) Data currency
  • WBUT Years: 2011
  • Answer: (c)
  • Explanation: > "Dimensionality is crucial for DSS data, enabling multi-perspective analysis of business performance."

22. Question: ......... is a subject-oriented, integrated, time-variant, non-volatile collection of data

  • Options:
    • a) Data Mining
    • b) Data Warehousing
    • c) Document Mining
    • d) Text Mining
  • WBUT Years: 2017
  • Answer: (b)
  • Explanation: > "This is the standard definition of a data warehouse, highlighting its four key characteristics."

23. Question: What is Metadata?

  • Options:
    • a) Summarized data
    • b) Operational data
    • c) Data about data
    • d) None of these
  • WBUT Years: 2017
  • Answer: (c)
  • Explanation: > "Metadata provides descriptive information about other data, defining its structure, meaning, and context."

24. Question: The full form of OLAP is

  • Options:
    • a) Online Analytical Processing
    • b) Online Advanced Processing
    • c) Online Advanced preparation
    • d) Online Analytical Performance
  • WBUT Years: 2017
  • Answer: (a)
  • Explanation: > "OLAP stands for Online Analytical Processing, which enables fast, interactive analysis of multidimensional data."

25. Question: The apriori algorithm is a

  • Options:
    • a) top - down search
    • b) breadth first search
    • c) depth first search
    • d) bottom-up search
  • WBUT Years: 2017
  • Answer: (d)
  • Explanation: > "The Apriori algorithm uses a bottom-up approach, building frequent itemsets from smaller ones."

26. Question: Classification rules are extracted from

  • Options:
    • a) Root node
    • b) Decision tree
    • c) Siblings
    • d) Branches
  • WBUT Years: 2017
  • Answer: (b)
  • Explanation: > "Decision trees provide clear, interpretable rules for classification by mapping decision paths."

27. Question: Which of the following is a predictive model?

  • Options:
    • a) Clustering
    • b) Regression
    • c) Summarization
    • d) Association rules
  • WBUT Years: 2017
  • Answer: (b)
  • Explanation: > "Regression is a predictive modeling technique used to forecast continuous numerical values."

28. Question: All set of items whose support is greater than the user-specified minimum support are called as

  • Options:
    • a) Border set
    • b) Frequent set
    • c) Maximal frequent set
    • d) Lattice
  • WBUT Years: 2017
  • Answer: (b)
  • Explanation: > "A frequent set (or frequent itemset) is a collection of items that appears together in transactions above a specified support threshold."

29. Question: The algorithm which uses the concept of a train running over data to find associations of items in data mining known as

  • Options:
    • a) Apriority Algorithm
    • b) Partition Algorithm
    • c) Dynamic Item-set Counting Algorithm
    • d) FP-Tree growth Algorithm
  • WBUT Years: 2011
  • Answer: (c)
  • Explanation: > "Dynamic Item-set Counting uses a 'train' metaphor to incrementally count itemsets as data transactions are processed."

30. Question: If we know exactly what information we need then...would suffice, but if we vaguely know the possible patterns then.......are useful.

  • Options:
    • a) Data Warehouse, Data Mining techniques
    • b) DBMS Query, Data Mining techniques
    • c) DBMS Query, Data Warehouse applications
    • d) Data Warehouse applications, Data Mining techniques
  • WBUT Years: 2012
  • Answer: (b)
  • Explanation: > "DBMS queries are for retrieving specific, known information, while data mining techniques are used to discover hidden or vague patterns."

31. Question: Association analysis is used for

  • Options:
    • a) transaction data analysis
    • b) olap
    • c) molap
    • d) none of these
  • WBUT Years: 2014
  • Answer: (a)
  • Explanation: > "Association analysis is predominantly applied to transactional databases to find relationships between items."

32. Question: Which frequent pattern mining technique mines without candidate generation?

  • Options:
    • a) Partitioning
    • b) Apriori
    • c) FP-growth
    • d) Dynamic intensive counting
  • WBUT Years: 2018
  • Answer: (c)
  • Explanation: > "FP-Growth avoids the costly candidate generation step by using a compact FP-tree structure for frequent pattern discovery."

33. Question: Choose correct alternatives from the following options:

  • Options:
    • a) Both (i) and (ii) is true
    • b) Both (ii) and (iii) is true
    • c) (i) is true and (iv) is false
    • d) (i) is true and (iii) is false
    • (i) The attribute with the highest information gain is chosen as the splitting attribute
    • (ii) The attribute with the lowest information gain is chosen as the splitting attribute
    • (iii) The attribute with the Highest Gini index is chosen as the splitting attribute
    • (iv) The attribute with the Lowest Gini index is chosen as the splitting attribute
  • WBUT Years: 2018
  • Answer: (d)
  • Explanation: > "In decision tree construction, the goal is to reduce impurity, which means selecting attributes with the highest information gain or the lowest Gini index for splitting."

Top comments (0)