Comparar métodos

	Boosting Semi-supervisionado	Gradient Boosting
Área	Aprendizado de máquina	Aprendizado de máquina
Família	Machine learning	Machine learning
Ano de origem	1999–2009	2001
Autor original	Mallapragada, P. K.; Bennett, K. P.; and others	Friedman, J. H.
Tipo	Semi-supervised ensemble method	Ensemble (sequential boosting of decision trees)
Fonte seminal	Mallapragada, P. K., Jin, R., Jain, A. K., & Liu, Y. (2009). SemiBoost: Boosting for Semi-supervised Learning. IEEE Transactions on Pattern Analysis and Machine Intelligence, 31(11), 2000–2014.	Friedman, J. H. (2001). Greedy Function Approximation: A Gradient Boosting Machine. Annals of Statistics, 29(5), 1189–1232.
Outros nomes	SemiBoost, SSL boosting, boosting with unlabeled data, semi-supervised ensemble boosting	Gradient Boosting (GBM), GBM, gradient boosted trees, gradient boosting machine
Relacionados	5	5
Resumo	Semi-supervised Boosting is an ensemble learning paradigm that extends classical boosting algorithms — such as AdaBoost — to exploit both labeled and unlabeled data. By propagating label information through a similarity structure over unlabeled instances, it trains stronger classifiers than supervised boosting alone when labeled data are scarce.	Gradient Boosting is an ensemble learning method, formalised by Jerome H. Friedman in 2001, that combines a sequence of weak learners — typically shallow decision trees — so that each new tree is fitted to minimise the residual errors of the trees before it. It is the core algorithm behind popular implementations such as XGBoost, LightGBM and CatBoost.