Unit 1 - Introduction to Machine Learning (ML)
General & Domain Knowledge:
1. What is Artificial Intelligence (AI)? How does it differ from traditional
programming?
2. Explain the differences between Narrow AI and General AI with examples.
3. List and describe two real-world applications of AI in different industries.
4. How does AI contribute to automation in modern industries?
5. What are the ethical concerns in AI development?
6. What are three real-life applications of Machine Learning?
7. How is Machine Learning different from AI and Deep Learning?
8. Why has Machine Learning become more popular in recent years?
Definition and Types of Machine Learning
1. Define Machine Learning. Why is it important?
2. What are the three types of Machine Learning? Give examples of each.
3. How is Supervised Learning different from Unsupervised Learning?
4. What is Reinforcement Learning? Provide a real-world use case.
5. Explain with examples:
a) Classification vs. Regression
b) Clustering vs. Association Rule Learning
6. What is a labeled dataset? Which type of ML requires it?
7. Differentiate between Parametric and Non-Parametric ML models.
8. Compare Reinforcement Learning and Supervised Learning.
Machine Learning Workflow and Model Development Cycle
1. Describe the Machine Learning Workflow in steps.
2. What are the key phases of a Machine Learning project?
3. Why is feature engineering important in model performance?
4. What are training, validation, and test datasets?
5. Explain the role of hyperparameter tuning in model training.
6. What is cross-validation, and why is it used in ML?
7. What is the difference between bias and variance in ML models?
8. What is overfitting? How can it be prevented?
Data Preprocessing
1. Why is data preprocessing necessary in ML?
2. What are common causes of missing values in a dataset?
3. Explain three techniques for handling missing data.
4. What are outliers? How can they be detected and handled?
5. Why should inconsistent data be removed or corrected?
Encoding Categorical Data
1. Why is encoding categorical data necessary?
2. What is One-Hot Encoding? How does it work?
3. Differentiate between Label Encoding and One-Hot Encoding.
4. When should you use ordinal encoding instead of one-hot encoding?
Data Normalization and Feature Scaling
1. Why do we normalize or standardize numerical data in ML?
2. Explain Min-Max Scaling with an example.
3. What is Z-score normalization?
4. How does StandardScaler differ from MinMaxScaler in Scikit-Learn?
5. When is feature scaling not required in Machine Learning?
Scikit-Learn
● What is Scikit-Learn? Why is it widely used in ML?
● Mention any four functionalities provided by Scikit-Learn.
● How does train_test_split() function work in Scikit-Learn?
● Write a Python snippet to train a Linear Regression model using Scikit-Learn.
Pandas
● What is the role of Pandas in data preprocessing?
● Differentiate between Pandas Series and DataFrame.
● How can you handle missing data using Pandas?
● Write a Python command to drop duplicate rows from a Pandas DataFrame.
NumPy
● Why is NumPy important in ML?
● How is NumPy different from Python lists?
● What is the purpose of np.array() in NumPy?
● Explain the difference between NumPy’s reshape() and ravel() functions.
Dimensionality Reduction (PCA)
● What is dimensionality reduction in ML? Why is it needed?
● Explain the working principle of Principal Component Analysis (PCA).
● How does PCA help in reducing model complexity?
● What is the difference between PCA and LDA (Linear Discriminant Analysis)?
Problems:
Min-Max Scaling Problems
Given values: X=[5,15,25,35,45]
Q2: Given values: X=[50,100,200,400,800]
PCA Problems:
Question samples:
1
For Solutions, refer to:
https://youtu.be/ZtS6sQUAh0c?feature=shared
https://www.youtube.com/watch?v=MLaJbA82nzk
