AD-603-A Data Mining and Warchousing - Jun 2025 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
   aDiscuss about the Three-tier data warehouse architecture with a neat diagram.
   How to answer

   • Define three-tier architecture: bottom, middle, top tiers
   • Explain warehouse database server, OLAP server, client tools
   • Describe data flow ETL to OLAP to analysis
   • Neat labelled diagram of three tiers
2. 7 Marks
   bWhat are the steps involved in data preprocessing? Explain about dimensionality reduction technique.
   How to answer

   • List data preprocessing steps: cleaning, integration, reduction, transformation
   • Define dimensionality reduction need
   • Explain techniques: PCA, wavelet, attribute subset selection
   • Benefits and example
3. 7 Marks
   aDiscuss about Metadata in the context of Data Warehousing and its role in data management. Explain how it supports the efficiency of data retrieval, transformation, and analysis?
   How to answer

   • Define metadata in data warehousing
   • Types: operational, extraction, end-user metadata
   • Role in data management, ETL, schema
   • How it supports retrieval, transformation, analysis efficiency
4. 7 Marks
   bExplain OLAP operations with example.
   How to answer

   • Define OLAP and multidimensional model
   • Explain operations: roll-up, drill-down, slice, dice, pivot
   • Give example with sales data cube
   • Usefulness in analysis
5. 14 Marks
   Discuss the key hardware considerations for data warehouse design, particularly focusing on the components necessary for supporting OLAP queries and data mining operations.
   How to answer

   • Need for high performance hardware for OLAP and mining
   • Discuss CPU, memory, storage, I/O, parallelism
   • Disk arrays, scalability, networking considerations
   • Relate to query throughput and mining workload
6. 7 Marks
   aWhy Data Preprocessing is needed and explain Data Preprocessing tasks?
   How to answer

   • Need: noisy, incomplete, inconsistent data
   • List tasks: cleaning, integration, transformation, reduction
   • Brief explanation of each task
   • Impact on mining quality
7. 7 Marks
   bExplain data mining as a step in knowledge discovery process.
   How to answer

   • Define KDD process steps
   • Position data mining as central step
   • Explain selection, preprocessing, transformation, mining, evaluation
   • Diagram of KDD flow
8. 7 Marks
   aElaborate the concept of "fuzzy sets" and discuss how fuzzy logic can be applied to data mining.
   How to answer

   • Define fuzzy set vs crisp set with membership function
   • Explain fuzzy logic concepts: linguistic variables, rules
   • Applications in clustering, classification, association
   • Example of handling uncertainty in mining
9. 7 Marks
   bExplain the rule-based classifier with example.
   How to answer

   • Define rule-based classifier: IF-THEN rules
   • Explain rule induction, coverage, accuracy
   • Example ruleset for classification
   • Rule ordering and pruning
10. 7 Marks
    aWith a neat block diagram, explain the general approach to solve a classification problem.
    How to answer

    • Define classification: learning and testing phases
    • Steps: training, model construction, evaluation
    • Block diagram of general approach
    • Example application
11. 7 Marks
    bExplain the structure, working and real-world applications of Rule-Based Algorithms. What are the strengths and limitations of using a rule-based approach and how might these systems evolve in the future?
    How to answer

    • Structure: rule base, inference engine, working memory
    • Working: matching, conflict resolution, firing
    • Real-world applications and strengths
    • Limitations and future evolution
12. 7 Marks
    aFor a transactional database, we can get a large number of association rules. How can we get useful association rules from such a large set?
    How to answer

    • Problem of large rule sets from transactions
    • Use support, confidence, lift measures
    • Interestingness, pruning, constraints, templates
    • Closed and maximal patterns approach
13. 7 Marks
    bExplain the BIRCH scalable Algorithm.
    How to answer

    • Define BIRCH: Balanced Iterative Reducing and Clustering
    • Explain CF-tree, CF vector structure
    • Phases: building tree, condensing, clustering
    • Scalability for large databases
14. 14 Marks
    Write short notes on any two of the following.

    1. a)DBSCAN clustering Algorithm.
    2. b)Similarity measures.
    3. c)Importance of data marts in data warehouse
    How to answer

    • Select any two topics for short notes
    • Define key concepts, algorithm steps or importance
    • Give examples/diagram where relevant
    • Applications and advantages