<h2>How Random Forests Work</h2>
                <p>Random Forest is an ensemble learning method that operates by constructing a multitude of <b>Decision Trees</b> at training time. It corrects for the habit of decision trees overfitting to their training set.</p>

                <h3>Key Concepts:</h3>
                <ul>
                    <li><b>Bagging (Bootstrap Aggregating):</b> Random forests allow each tree to pick only a random sample of the data. This reduces variance.</li>
                    <li><b>Feature Randomness:</b> Each tree can only pick from a random subset of features. This forces trees to be more diverse.</li>
                </ul>
                <div class="alert alert-info"><b>Note:</b> A single decision tree has high variance (it overfits). A random forest has lower variance but slightly higher bias.</div>

                <h2>Scikit-Learn Code Example</h2>
                <pre><code>from sklearn.ensemble import RandomForestClassifier

from sklearn.model_selection import train_test_split

Initialize the model

n_estimators = number of trees

clf = RandomForestClassifier(n_estimators=100, max_depth=2, random_state=0)

Fit to training data

clf.fit(X_train, y_train)

Predict class labels

y_pred = clf.predict(X_test)

Exam Tip: Always check your feature_importances_ attribute to understand which variables are driving your model's decisions. This is crucial for model explainability.