IT-603-B Data Mining - 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
   aDescribe the steps involved in data mining when viewed as a process of knowledge discovery.
   How to answer

   • Define KDD process
   • List steps: cleaning, integration, selection, transformation, mining, evaluation, presentation
   • Brief purpose of each step
   • Diagram of KDD process flow
2. 7 Marks
   bCompare ROLAP, MOLAP, OLAP. Explain different OLAP operations.
   How to answer

   • Define OLAP, ROLAP, MOLAP
   • Compare architecture, storage, scalability, performance
   • Explain OLAP operations: roll-up, drill-down, slice, dice, pivot
   • Give examples
3. 7 Marks
   aExplain data transformation strategies in detail.
   How to answer

   • Define data transformation
   • Explain smoothing, aggregation, generalization
   • Explain normalization and attribute construction
   • Explain discretization
4. 7 Marks
   bIn real world data, tuples with missing values for some attributes are a common occurrence. Describe various methods for handling this problem.
   How to answer

   • State missing value problem
   • List methods: ignore tuple, fill manually, global constant, mean/median, class mean, most probable value
   • Discuss advantages and drawbacks
   • Recommend context of use
5. 7 Marks
   aDraw a snowflake schema diagram for "University" data warehouse which consist of four dimensions student, course, semester and instructor and two measures count and avg_grade. At the lowest conceptual level, the avg_grade measure stores the actual course grade of student. At higher level avg_grades store average grade for given schema.
   How to answer

   • Identify fact table with count and avg_grade
   • Show four dimensions: student, course, semester, instructor
   • Normalize dimensions into snowflake tables
   • Neat labelled snowflake schema diagram
6. 7 Marks
   bDescribe knowledge discovery process in detail.
   How to answer

   • Define knowledge discovery
   • List KDD steps in order
   • Explain role of each step
   • Show iterative nature
7. 7 Marks
   aGive an example to show that items in a strong association rule actually may be negatively correlated.
   How to answer

   • Define support, confidence and correlation/lift
   • Give contingency table example with high support/confidence but lift <1
   • Show calculation of correlation
   • Conclude strong rule can be negatively correlated
8. 7 Marks
   bApriori algorithm makes use of prior knowledge of subset support properties. Prove that all non empty subsets of a frequent itemset must also be frequent.
   How to answer

   • State Apriori property and support definition
   • Use $support(S) \ge support(I)$ for $S \subseteq I$
   • Argue if itemset frequent then all subsets meet min_sup
   • Conclude all non-empty subsets are frequent
9. 7 Marks
   aWhy is tree pruning useful in decision tree induction? What is a drawback of using a separate set of tuples to evaluate pruning?
   How to answer

   • Define overfitting and tree pruning
   • Explain pre-pruning and post-pruning benefits
   • Explain need for separate pruning set and data loss
   • State drawback of reduced training data
10. 7 Marks
    bGive an example to show why the k-means algorithm may not find the global optimum, that is optimizing the within cluster variation.
    How to answer

    • Define k-means objective: within-cluster variation
    • Give small 1D/2D point set with bad initialization
    • Show convergence to local optimum
    • Explain sensitivity to initial centroids
11. 7 Marks
    aClusters may form a hierarchy, outliers may belong to different granularity level. Explain a clustering based outlier detection method that can find outliers at different levels.
    How to answer

    • Define outliers at different granularity levels
    • Explain clustering-based detection idea
    • Describe CBLOF / hierarchical clustering approach
    • Explain detection at multiple levels
12. 7 Marks
    bIn outlier detection by semi-supervised learning, what is the advantage of using objects without labels in the training dataset?
    How to answer

    • Define semi-supervised outlier detection
    • Explain use of unlabeled data to model normal behavior
    • State advantage: better boundary, less labeling cost
    • Give example
13. 7 Marks
    aWhat is Web mining? Explain its types.
    How to answer

    • Define web mining
    • Explain web content mining
    • Explain web structure and web usage mining
    • Give examples of each
14. 7 Marks
    bDescribe the security issues in data mining.
    How to answer

    • Define privacy/security concern in mining
    • List issues: privacy, data security, misuse
    • Explain countermeasures
    • Give examples
15. 7 Marks
    aDescribe spatial and Temporal mining with suitable example.
    How to answer

    • Define spatial mining with map example
    • Define temporal / time-series mining
    • Give suitable examples
    • Highlight applications
16. 7 Marks
    bWrite k-means algorithm with example.
    How to answer

    • List k-means steps: choose k, initialize, assign, update
    • Explain convergence criterion
    • Illustrate with small numeric example
    • State complexity and limitations