Comparer des méthodes

	Agrégation par décompte de Borda	Généralisation empilée	Vote Pondéré
Domaine	Apprentissage ensembliste	Apprentissage ensembliste	Prise de décision
Famille	Machine learning	Machine learning	MCDM
Année d'origine	1781	1992	1951
Auteur d'origine	Jean-Charles de Borda	David Wolpert	Arrow, K. J.
Type	rank-based aggregation	meta-learning aggregation	Social choice — weighted positional voting rule
Source fondatrice	Borda, J. C. de (1781). Mémoire sur les élections au scrutin. Histoire de l'Académie Royale des Sciences.	Wolpert, D. H. (1992). Stacked generalization. Neural Networks, 5(2), 241-259.	Arrow, K. J. (1951). Social Choice and Individual Values. Wiley, New York
Alias	weighted voting, rank aggregation	stacking, meta-learning	—
Apparentées	3	3	0
Résumé	Borda count is a preference aggregation method that combines ranked predictions from multiple classifiers by assigning points based on ranking position. Each classifier ranks the possible outcomes, and each class receives points inversely proportional to its rank position. The class with the highest total score is selected. Originally proposed by French mathematician Jean-Charles de Borda in 1781, this method has been adapted for ensemble learning to aggregate soft predictions and rank-ordered outputs.	Stacked generalization, or stacking, is a two-level ensemble method where base-level classifiers are trained on the original data, and a meta-learner is trained on the predictions of the base classifiers. The meta-learner learns how to best combine base predictions rather than using fixed aggregation rules. Introduced by David Wolpert in 1992, stacking achieves state-of-the-art performance by automatically learning the optimal weighting and interaction patterns among base models.	WEIGHTED-VOTING (Weighted Voting — Weighted positional aggregation of multiple rankings) is a ranking multi-criteria decision-making (MCDM) method introduced by Arrow, K. J. in 1951. It turns a decision matrix of alternatives scored on multiple criteria into a structured, reproducible result.