Lecture Six: Introduction to Machine Learning
Instructions:
Visit and review the following resources:
[Link]
[Link]
The PDF document "Machine Learning" for the course DECAP737 from
Lovely Professional University. url: chrome-
extension://efaidnbmnnnibpcajpcglclefindmkaj/[Link]
Master%20of%20Computer%20Applications/Sem_4/DECAP737_MACHINE%20
[Link]
Using your own understanding and notes from those, answer all of the
following questions.
More instructions are on VLE platform.
Part A: Theory – Machine Learning Fundamentals
1. Define Machine Learning in your own words and explain how it differs
from classical programming.
2. Distinguish between AI, ML and Deep Learning, using examples.
3. Describe the main types of Machine Learning:
o Supervised Learning
o Unsupervised Learning
o Reinforcement Learning
o Semi-supervised Learning
4. What is a dataset? Explain the difference between features (input
variables) and targets/outcomes (labels).
5. Explain the difference between a classification problem and a
regression problem, giving examples.
6. What are training, validation, and testing datasets, and why is each
used?
7. What does overfitting mean? What does underfitting mean? Why is the
balance important?
8. What is the “curse of dimensionality”? How can it impact a machine
learning model?
9. Define feature engineering and explain why it matters in machine
learning.
10. List three real-world applications of machine learning and explain how
ML is used in each case.
Part B: Algorithms & Concepts
11. Describe any four machine learning algorithms (e.g., Linear Regression,
Logistic Regression, K-Means, Decision Trees) and indicate whether they
are supervised or unsupervised.
12. Explain the difference between parametric and non-parametric models
with examples.
13. What is a cost/loss function in machine learning? Provide an example.
14. What is the purpose of gradient descent in training models?
15. Name and explain at least three evaluation metrics (for example:
accuracy, precision, recall, F1-score, MSE).
16. Explain why data preprocessing is critical, and mention at least three
preprocessing tasks (e.g., handling missing data, scaling, encoding
categorical variables).
17. What does it mean for a model to generalise well?
18. Mention two advantages and two limitations of machine learning, based
on your reading of the resources.
19. What is the algorithm K-Means clustering used for, and how does it work
in simple terms?
20. Describe the bias-variance tradeoff in machine learning and why
finding the right balance is important.
Part C: Python for Machine Learning
21. Why is Python so widely used in machine learning?
22. What is the difference between a Python list and a tuple? Provide a short
example in code.
23. List four Python libraries used in machine learning (e.g., NumPy, Pandas,
Matplotlib, scikit-learn) and briefly state what each is used for.
24. Write Python code to create a list of numbers from 1 to 10 and print each
number in a loop.
25. Write Python code to convert that list into a NumPy array.
26. Write Python code using pandas to read a CSV file named [Link].
27. In the context of scikit-learn (or similar library), what do the methods fit()
and predict() do?
28. Write Python code to perform Min-Max scaling on a list of numbers (you
may assume you use numpy).
29. Provide a short Python snippet using Matplotlib to plot a simple line
graph of x vs y.
30. Write Python code to split a dataset into training and testing sets using
scikit-learn (you may assume from sklearn.model_selection import
train_test_split).
Part D: Simple Machine Learning Model (Practical)
Problem: Predict a student’s exam score based on hours studied using linear
regression.
Outline the full steps you would follow to build this model (from problem
definition to deployment).
Write full Python code using scikit-learn to implement this. Use the
sample data:
Hours = [2, 4, 6, 8, 10]
Scores = [50, 58, 65, 78, 85]
Use your model to predict the score for a student who studies for 7
hours.
32. Use the clustering algorithm K-Means to group 6 students by their
marks:
Marks = [45, 50, 52, 80, 85, 88]
Explain your steps.
Write Python code to apply K-Means clustering (you may assume from
[Link] import KMeans).
Display or explain how you would interpret the resulting clusters.