@@ -99,4 +99,32 @@ Advantages:
9999
100100Disadvantages:
101101=================
102- * May not significantly improve performance if base learners are not diverse.
102+ * May not significantly improve performance if base learners are not diverse.
103+ 104+ Boosting
105+ ====================================
106+ * This is an ensemble technique aimed at improving the accuracy and stability of ML models
107+ * It is done by combining weak learners(models that perfrom slightly better than random chance) to create a strong learner
108+ * The strong learner is built in iterations with focus on misclassified instances.
109+ 110+ How Boosting Works:
111+ ===============
112+ * Sequential Learning: Models are trained sequentially, where each new model focuses on correcting errors made by previous models.
113+ * Weight Adjustment: Misclassified instances are given higher weights so that subsequent models pay more attention to them.
114+ * Final Prediction: Combine predictions from all models, typically using weighted voting or averaging
115+ 116+ Popular Boosting Algorithms:
117+ ==========
118+ * AdaBoost
119+ * Gradient Boosting
120+ * XGBoost
121+ 122+ Advantages:
123+ ==========
124+ * Often achieves high accuracy and performs well even with limited data.
125+ * Can handle various types of data and relationships.
126+ 127+ Disadvantages:
128+ =================
129+ * More prone to overfitting than bagging if not carefully tuned.
130+ * Requires careful tuning of parameters.
0 commit comments