Elective-I (Machine Learning)
Mahatma Gandhi Central University, Bihar
B.Tech Computer Science & Engineering
Semester 5 Examination, 2025
Elective-I (Machine Learning) (CSET3180)
Assessment Questions
Elective-I (Machine Learning) (CSET3180) - 2025
Section A (MCQ)
(MULTIPLE CHOICE QUESTIONS – 05 Marks) Attempt all questions. Each question carries one mark.
- 1
A good clustering method will produce high quality clusters with
- 2
What is the entropy at p = 0.5 from the figure given below?
- 3
What is the fundamental objective of reinforcement learning algorithm?
- 4
What is the issue, when a model performs well on training data but poorly perform on test data?
- 5
Which of the following tasks is an example of classification in supervised learning?
Section B (Short Answer)
Write briefly in 250 words (Short Answer Questions – 05 Marks) Attempt ANY TWO questions out of the following. Each question carries 2.5 Marks.
- 1
Explain why computing the proximity between two attributes is often simpler than computing the similarity between two objects.
- 2
Explain the transformations with the help of a suitable table that defines the attribute levels.
- 3
What do you mean by clustering? Explain the K-means clustering by writing a suitable algorithm.
Section C (Long Answer)
Write in 500 Words (Long Answer Questions – 10 Marks) Attempt ALL questions having internal choices. Each question carries 5 marks.
- 1
Answer any ONE of the following
(a)Explain the difference between normalization and standardization.
OR
(b)What do you understand from rule evaluation in rule based classifier? Explain the various rule evaluation schemes with the help of a suitable example.
- 2
Answer any ONE of the following
(a)What do you mean by the measures of selecting the best split in decision tree? Explain the Entropy, Gini, information_gain, Gain_Ratio and Classification error with the help of a suitable example.
OR
(b)Compute the predicted class label of the point S₆ (X₁ = 6, X₂ = 7) using the K-Nearest Neighbors (KNN) algorithm with K = 3, applying Euclidean distance as the similarity measure and majority voting for classification.
The dataset for the same given below, contains two features — Study Hours (X₁) and Sleep Hours (X₂) — and a binary class label representing the Exam Result (Y).
Student Study Hours (X₁) Sleep Hours (X₂) Result (Y) S₁ 2 9 Fail S₂ 4 8 Fail S₃ 7 6 Pass S₄ 9 3 Pass S₅ 10 2 Pass