Module–III: Machine Learning (AI Notes)
Comprehensive 2 Marks & 10 Marks Questions with Answers
2 MARKS QUESTIONS WITH SHORT ANSWERS
1. Define Machine Learning.
Machine Learning is a branch of AI that enables systems to learn from data and improve
automatically without being explicitly programmed.
2. What are the main types of Machine Learning?
Supervised Learning, Unsupervised Learning, and Reinforcement Learning.
3. What is the difference between AI and ML?
AI is the broader concept of creating intelligent systems, while ML focuses on systems that
learn from data.
4. Define regression in Machine Learning.
Regression is a technique used to predict continuous values based on input variables.
5. What is classification?
Classification assigns input data into predefined categories or classes.
6. Define clustering.
Clustering groups similar data points based on patterns or similarities without predefined
labels.
7. What is the purpose of a Machine Learning model?
It learns relationships from data to make predictions or decisions.
8. Write any two applications of ML.
Image recognition and spam email filtering.
9. What is Naïve Bayes classifier?
It is a probabilistic classifier based on Bayes’ theorem assuming independence among
predictors.
10. What is Support Vector Machine (SVM)?
SVM is a supervised learning model that separates data into classes using the best
possible hyperplane.
10 MARKS QUESTIONS WITH DETAILED ANSWERS
1. Explain the various techniques in Artificial Intelligence used for Machine
Learning.
The major AI techniques used in Machine Learning include:
1. **Neural Networks** – Mimic the human brain structure to identify patterns.
2. **Support Vector Machines (SVM)** – Separate data using optimal hyperplanes.
3. **Decision Trees** – Split data into branches based on attribute values.
4. **Bayesian Networks** – Represent probabilistic relationships among variables.
5. **Fuzzy Logic** – Handles uncertain and imprecise data using fuzzy rules.
2. Describe the Machine Learning model with its components and working.
A Machine Learning model consists of:
1. **Input Data** – Training data with features and labels.
2. **Feature Extraction** – Selecting relevant attributes.
3. **Learning Algorithm** – Finds patterns from the data.
4. **Model Evaluation** – Tests accuracy using metrics.
5. **Prediction** – Produces output for new inputs.
The model is trained on data and evaluated for performance before deployment.
3. Explain Regression Analysis in Machine Learning.
Regression is a supervised learning method used for predicting continuous outcomes.
Types include:
- **Linear Regression:** Establishes a straight-line relationship.
- **Polynomial Regression:** Uses higher-degree equations.
- **Logistic Regression:** Used for binary classification problems.
4. Discuss Classification Techniques in detail.
Classification assigns data into categories. Popular methods:
- **Decision Trees**
- **Naïve Bayes Classifier**
- **K-Nearest Neighbors (KNN)**
- **Support Vector Machines (SVM)**
Applications include disease diagnosis, image classification, and email filtering.
5. Explain Clustering Techniques.
Clustering is an unsupervised learning process grouping similar data points.
Common techniques:
- **K-Means Clustering** – Divides data into K clusters.
- **Hierarchical Clustering** – Builds a hierarchy of clusters.
Used in market segmentation, pattern recognition, and data compression.
6. Explain Naïve Bayes Classification with example.
Naïve Bayes uses probability based on Bayes’ theorem.
Formula: P(A|B) = [P(B|A) * P(A)] / P(B)
Example: Predicting if an email is spam or not based on word occurrence.
7. Describe the structure and working of a Neural Network.
A Neural Network consists of layers:
- **Input Layer** – Takes features.
- **Hidden Layers** – Process data through weighted neurons.
- **Output Layer** – Produces final results.
It learns using backpropagation to minimize error.
8. Explain Support Vector Machine (SVM) with diagram.
SVM finds the optimal hyperplane that separates classes with maximum margin.
It can be used for linear and non-linear classification using kernel functions.
9. Compare Regression, Classification, and Clustering.
Regression predicts continuous values, classification predicts discrete labels, and
clustering groups unlabeled data.
10. Write applications and challenges of Machine Learning.
Applications: Healthcare, finance, autonomous vehicles, and recommendation systems.
Challenges: Data quality, model interpretability, overfitting, and ethical issues.