Machine Learning - Quick Revision Notes
Unit 1: Introduction to Machine Learning
What is Machine Learning?
Machine Learning is a subfield of AI that enables systems to learn from data and improve from experience
without being explicitly programmed.
Types of Machine Learning:
- Supervised Learning: Learn from labeled data (e.g., Spam Detection)
- Unsupervised Learning: Learn from unlabeled data (e.g., Customer Clustering)
- Reinforcement Learning: Learn by reward & punishment (e.g., Game Playing Bots)
Basic Terms:
- Dataset: Collection of data
- Training: Teaching model with data
- Testing: Evaluating model
- Features: Input variables
- Label: Output variable
Unit 2: Supervised Learning Algorithms
Linear Regression:
- Used for predicting numerical values
- Formula: y = mx + c
Logistic Regression:
- Used for classification (Yes/No)
- Uses sigmoid function
K-Nearest Neighbors (KNN):
- Classifies data based on closest data points
Machine Learning - Quick Revision Notes
- Distance formula: Euclidean distance
Decision Tree:
- A tree-like model for decision making
- Nodes = features, Leaves = outcomes
Overfitting vs Underfitting:
- Overfitting: Model fits training data too well, poor on test data
- Underfitting: Model too simple, performs poorly on both
Unit 3: Unsupervised Learning
K-Means Clustering:
- Groups similar data into clusters
- You define number of clusters (K)
PCA (Principal Component Analysis):
- Reduces dimensionality (features) of data
- Helps in simplifying models
Unit 4: Probabilistic Learning
Naive Bayes Classifier:
- Based on Bayes' theorem
- Assumes features are independent
- Used in spam filters, text classification
Unit 5: ML Model Workflow
Machine Learning - Quick Revision Notes
1. Problem Definition
2. Data Collection
3. Data Preprocessing
4. Model Selection
5. Training
6. Testing
7. Evaluation (Accuracy, Precision, Recall)
Expected Questions for Passing
1. Define ML and its types.
2. Difference: Supervised vs Unsupervised.
3. Applications of ML.
4. Explain KNN or Decision Tree with example.
5. Write short note on Naive Bayes.
6. What is overfitting and underfitting?
7. Explain ML workflow steps.
Smart Tips for Writing Exam
- Start answers with definition
- Add example to support
- Use simple diagrams (KNN, Tree, PCA)
- Use headings & bullet points for neatness
- Attempt all questions - even if unsure
