Data Mining and Machine Learning
CSE · Final · Summer 25
Batch 63
Read the questions as text
Read off the PDF by AI, so it may have mistakes. The PDF is the original.
Daffodil International University
Faculty of Science & Information Technology
Department of Computer Science & Engineering
Final Examination, Summer 2025
Course Code: CSE325, Course Title: Data Mining and Machine Learning
Level: 3 Term: 3 Batch: 63
Time: 02:00 Hrs
Marks: 40
Answer ALL Questions
[The figures in the right margin indicate the full marks and corresponding course outcomes. All portions of each question must be answered sequentially.]
In a quiet neighborhood, there's a small family-owned shop called The Corner Mart. It's known for its essentials: fresh milk, warm bread, farm-fresh eggs, and creamy butter. Each morning, regular customers stop by:
- Mrs. Green always buys milk, bread, apples, and eggs for breakfast.
- Mr. Patel, a bachelor, usually grabs milk, eggs, meat, and bread for his sandwiches.
- Sophie, a college student, prefers a protein-heavy combo of milk and eggs.
- The Thompson family likes milk, bread, eggs, and sometimes butter and meat on weekends.
- Elderly Mr. Clark comes by often just for coffee, milk, and butter.
- Young Jake, who lives nearby, likes to help his mom by buying milk, bread, coffee, and butter.
The shop records the transactions of 6 regular customers. The owner, curious about what products are frequently bought together, wants to analyze these transactions to make smarter inventory decisions and offer some discounts.
Now, based on the story above, answer the following questions 1 and 2.
1 | Organize the table based on the given data. Which algorithm is suitable for analyzing this dataset, and why? Explain your reasoning with Examples. | 5 | CO1
2 | Find all possible 3-itemset association rules (meeting the same minimum thresholds of support and confidence) that can be generated from these transactions. Additionally, Analyze and interpret the rule {Bread, Eggs} => {Milk} in comparison with the other strong rules to provide meaningful business insights for inventory planning and promotional offers.
Min Support = 20% and Min confidence = 60% | 10 | CO3
Dr. Karim, a public health researcher, is developing a patient monitoring system for early detection of cardiovascular risks. She gathers basic health metrics from patients during routine checkups, specifically:
- Body Mass Index (BMI) (min 18.5, max 35.0)
- Systolic Blood Pressure (mmHg) (min 60, max 160)
ID | BMI | BP
p1 | 22 | 118
p2 | 28.5 | 145
p3 | 31.2 | 160
p4 | 25.5 | 200
p5 | 20.0 | 110
During her data collection, she found irrelevant data.
3 | Analyze the suitability of K-Means clustering for the given patient dataset. Based on its limitations, identify and justify one alternative unsupervised learning method that could better group patients by health risk. | 5 | CO3
4 | Apply the K-Means clustering technique to cluster the patients based on BMI and blood pressure. Use the results to assign each patient to a group. (K=2) | 10 | CO2
You are given a simple Artificial Neural Network (ANN) with the following data. The last inputs in the hidden units are "Bias" and Sigmoid is the activation function in the hidden layer.
X = [[0.5, 0.9], [0.8, 0.1]] Y = [[0.60], [0.15]]
W1 | 0.1 | 0.4
W1 | 0.2 | 0.6
W1 | 0.3 | 0.3
W2 | 0.3 | 0.8
W2 | 0.5 | 0.1
W2 | -0.2 | 0.2
W3 | 0.7
W3 | 0.9
W3 | -0.4
5 | Draw the Neural Network, and Apply forward propagation on the given data using the provided weights. Calculate the final outputs and find the Mean Squared Error. Show each step of the calculations clearly. | 10 | CO2