AdaBoost (Adaptive Boosting) is a machine learning boosting technique used as an ensemble method to adjust the weights of training samples and combine multiple weak classifiers into a single strong classifier.