Chapter 4: Supervised Learning

🎯 Supervised Learning is a type of machine learning where the model learns from labeled data (i.e., input with known output).

---

✅ What is Supervised Learning?

• The algorithm is "supervised" because it learns from training data that includes both the input and the correct output (label).

• The goal is to predict output for new, unseen inputs.

---

🔁 Supervised Learning Workflow:

markdown
1. Collect labeled data ➝
2. Split into train & test ➝
3. Choose a model ➝
4. Train model ➝
5. Evaluate on test data ➝
6. Predict new data

---

🔹 Types of Supervised Learning

Type	Output	Examples
Regression	Continuous value	Predict price, temperature, salary
Classification	Categorical class	Spam vs. Ham, Yes/No, Cat/Dog

---

🔢 1. Regression Algorithms

a. Linear Regression

• Predicts a continuous output using a straight line:

$$y = mx + c$$

📌 Example: Predict house price based on area.

python
from sklearn.linear_model import LinearRegression
model = LinearRegression()
model.fit(X_train, y_train)

---

b. Polynomial Regression

• Fits a curve (non-linear data).

• Useful when data is not linearly separable.

---

🧭 2. Classification Algorithms

a. Logistic Regression

• For binary classification (yes/no)

• Outputs probabilities between 0 and 1

python
from sklearn.linear_model import LogisticRegression
model = LogisticRegression()
model.fit(X_train, y_train)

---

b. Decision Trees

• Tree-like model for decisions.

• Easy to visualize.

c. K-Nearest Neighbors (KNN)

• Classifies a point based on its K nearest neighbors.

d. Support Vector Machines (SVM)

• Finds the best boundary (hyperplane) that separates classes.

---

📏 Model Evaluation Metrics

Metric	Used For	Description
Accuracy	Classification	% of correct predictions
Precision	Classification	True Positives / (TP + FP)
Recall	Classification	True Positives / (TP + FN)
F1 Score	Classification	Harmonic mean of precision and recall
MSE / RMSE	Regression	Measures prediction error

---

💻 Hands-On Project Ideas

✅ Predict House Prices (Regression)

• Input: Area, Bedrooms, Location

• Output: Price

✅ Email Spam Classifier (Classification)

• Input: Text of email

• Output: Spam or Not Spam

---

Example Code (Classification):

python
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import accuracy_score

# Split data
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)

# Train model
model = LogisticRegression()
model.fit(X_train, y_train)

# Predict
y_pred = model.predict(X_test)

# Evaluate
print("Accuracy:", accuracy_score(y_test, y_pred))

---

📚 Summary of Chapter 4

Concept	Summary
Supervised Learning	Learn from labeled data
Regression	Predict continuous values
Classification	Predict categories or classes
Common Algorithms	Linear, Logistic, KNN, SVM, Decision Trees
Evaluation Metrics	Accuracy, Precision, Recall, RMSE, F1 Score
Hands-on	Predict house price, classify spam

---

✅ Mini Assignment:

1. Build a Linear Regression model to predict house price using area.

2. Build a Spam Classifier using Logistic Regression and check accuracy.

3. Try tuning the value of K in KNN and observe results.