CY-604-C Data Warehousing and Mining - May 2023 Question Paper

Under each question, How to answer lists the points an answer should cover, outlined by CampusPrep from the syllabus. Check the details against your textbook.

1. 7 Marks
   aWhat is Vertical Partitioning and how it is done? Explain in detail.
   How to answer

   • Define vertical partitioning in data warehouse / distributed database context
   • Explain how it is done - splitting by columns/attributes, need for primary key / TID in each fragment
   • Steps / criteria for partitioning and reconstruction via join
   • Advantages, disadvantages and example
2. 7 Marks
   bDraw the data warehouse architecture and discuss its various components in detail.
   How to answer

   • Draw 3-tier data warehouse architecture diagram
   • Explain data sources, ETL / staging area, warehouse, data marts, OLAP / analysis tools
   • Explain metadata repository and management
   • Role of each component in information flow
3. 7 Marks
   aWhat is the need of MOLAP server? Explain.
   How to answer

   • Define MOLAP server and multidimensional storage
   • Need: fast OLAP query, precomputed cubes, multidimensional analysis
   • Explain advantages / limitations vs ROLAP
   • Example use case
4. 7 Marks
   bExplain about OLAP and OLTP with suitable examples.
   How to answer

   • Define OLAP and OLTP with characteristics
   • Compare purpose, users, query type, data model
   • Give suitable examples for each
   • Summary of differences
5. 7 Marks
   aExplain data mining task primitives in detail.
   How to answer

   • Define data mining task primitives
   • Explain task, knowledge, background knowledge, interestingness measures
   • Explain presentation / visualization primitives
   • Role in defining a mining query
6. 7 Marks
   bDiscuss about the role of data mining engine in KDD process.
   How to answer

   • Outline KDD process steps
   • Position of data mining engine in KDD
   • Functions: characterization, association, classification, clustering etc.
   • Interaction with knowledge base and user
7. 7 Marks
   aDifferentiate between top-down and bottom-up Hierarchical algorithms.
   How to answer

   • Define hierarchical clustering
   • Explain top-down (divisive) approach steps
   • Explain bottom-up (agglomerative) approach steps
   • Compare on basis of strategy, complexity, examples like DIANA vs AGNES
8. 7 Marks
   bExplain the Apriori algorithm to find the frequent item sets with suitable example.
   How to answer

   • Define frequent itemsets, support and confidence
   • Explain Apriori join and prune steps iteratively
   • Illustrate with sample transaction database
   • State advantages and limitations
9. 7 Marks
   aDefine Classifier. Explain the process of verifying the accuracy of classifier.
   How to answer

   • Define classifier in classification context
   • Explain training vs test split, confusion matrix
   • Explain accuracy estimation methods like holdout, cross-validation
   • Interpretation of accuracy
10. 7 Marks
    bExplain about Rule based algorithms to classify the data.
    How to answer

    • Define rule-based classification
    • Explain IF-THEN rules, rule coverage and accuracy
    • Explain rule induction / sequential covering and pruning
    • Example rule set for given data
11. 7 Marks
    aWrite about Galaxy Schema for Multidimensional databases in detail.
    How to answer

    • Define Galaxy / Fact constellation schema
    • Explain multiple fact tables sharing dimension tables
    • Diagram with example
    • Comparison with star and snowflake schema
12. 7 Marks
    bExplain the Data Cubes Computation process in detail.
    How to answer

    • Define data cube and cuboids
    • Explain full vs partial materialization
    • Explain computation methods like BUC, sorting / hashing
    • Optimization and indexing issues
13. 7 Marks
    aWhat are Probabilistic Classifiers? Explain.
    How to answer

    • Define probabilistic classifiers
    • Explain Bayes theorem basis and probability estimation
    • Examples like Naive Bayes
    • Advantages for uncertain data
14. 7 Marks
    bExplain FP-growth algorithm using suitable example.
    How to answer

    • Define FP-tree concept without candidate generation
    • Explain steps: frequency ordering, tree construction, mining conditional pattern bases
    • Illustrate with suitable transaction example
    • Compare with Apriori efficiency
15. 14 Marks
    Write short notes on any two:

    1. a)Multidimensional Databases
    2. b)Time Series Mining Association Rules
    3. c)Bayesian Classification
    4. d)Spatial Mining
    How to answer

    • Select any two topics from list
    • Give definition, key concepts and techniques for each
    • Give example / application for each
    • Brief advantages / uses