CS-503-B Pattern Recognition - Nov 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
   aExplain various Design principles of Pattern Recognition systems.
   How to answer

   • Define pattern recognition system and its components
   • Explain design principles: data collection, feature choice, model choice, training, evaluation
   • Discuss trade-offs: complexity, generalization, computational cost
   • Give examples of design cycle
2. 7 Marks
   bDifferentiate between supervised learning and unsupervised learning with suitable example.
   How to answer

   • Define supervised learning with labelled data
   • Define unsupervised learning with unlabelled data
   • Compare on basis of data, feedback, goal, algorithms
   • Give suitable examples for each
3. 7 Marks
   aWrite a detail note on Random Forest.
   How to answer

   • Define Random Forest as ensemble of decision trees
   • Explain bagging and random feature selection
   • Describe training and voting / averaging steps
   • List advantages, limitations and applications
4. 7 Marks
   bDefine the following:

   1. i)Clustering
   2. ii)Metric spaces
   How to answer

   • Define clustering as unsupervised grouping by similarity
   • Mention objective and examples
   • Define metric spaces with metric axioms
   • Give examples of distance metrics
5. 7 Marks
   aHow does K-Nearest Neighbor works? Explain with KNN estimation and KNN rules.
   How to answer

   • Explain KNN principle and distance measures
   • Explain KNN density estimation formula
   • Explain KNN classification / regression rules and choice of K
   • Illustrate with example
6. 7 Marks
   bWhat do you understand by normalization? Explain with example.
   How to answer

   • Define normalization and its need in pattern recognition
   • Explain min-max, z-score methods with formulas
   • Give numerical example
   • Mention effect on learning
7. 7 Marks
   aDescribe about cluster validation.
   How to answer

   • Define cluster validation and its need
   • Describe internal, external and relative criteria
   • Mention validity indices
   • Give example of validation procedure
8. 7 Marks
   bEnlist the clustering Techniques. Explain any one Technique.
   How to answer

   • List clustering techniques: partitional, hierarchical, density, model based
   • Explain any one technique with steps
   • Give example and diagram of clusters
   • Mention advantages and limitations
9. 7 Marks
   aState and explain the different paradigms of pattern recognition.
   How to answer

   • Define paradigm in pattern recognition context
   • Explain statistical, syntactic, neural, template matching paradigms
   • Compare their principles
   • Give applications of each
10. 7 Marks
    bDiscuss the various types of feature extraction.
    How to answer

    • Define feature extraction and its role
    • Discuss types: statistical, transform, edge, texture features
    • Explain methods like PCA, LDA
    • Mention selection criteria
11. 7 Marks
    aIllustrate branch and bound Algorithm with suitable example.
    How to answer

    • Define branch and bound for feature selection / search
    • Explain branching, bounding and pruning steps
    • Illustrate search tree with example
    • Mention optimality and complexity
12. 7 Marks
    bWrite an algorithm for forward selection with suitable example.
    How to answer

    • Define forward selection for feature subset
    • Write stepwise algorithm: start empty, add best feature
    • Illustrate with example evaluation criterion
    • Mention stopping criteria
13. 7 Marks
    aDescribe about the of Neuro-Fuzzy and its Techniques.
    How to answer

    • Define neuro-fuzzy system combining neural and fuzzy logic
    • Explain architecture and learning
    • Describe techniques: ANFIS, fuzzy MLP
    • List applications
14. 7 Marks
    bState and explain the rules of Histogram in detail.
    How to answer

    • Define histogram in pattern recognition
    • State rules: bin width, Sturges rule, normalization
    • Explain construction steps in detail
    • Give example
15. 14 Marks
    Write a short note on any two:

    1. i)Data sets for pattern
    2. ii)Training set
    3. iii)Classifier and variants
    4. iv)FCM
    How to answer

    • Select any two topics from the list
    • Define each with key concepts
    • Explain with examples / diagrams
    • Mention relevance to pattern recognition