Comparer des méthodes

	Inertie	Indice de Davies-Bouldin	Méthode du coude
Domaine	Évaluation de modèles	Évaluation de modèles	Évaluation de modèles
Famille	MCDM	MCDM	MCDM
Année d'origine	1967	1979	1953
Auteur d'origine	Stuart Lloyd, James MacQueen	David L. Davies, Donald W. Bouldin	Robert Thorndike
Type	Clustering quality metric	Cluster quality metric	Heuristic optimization criterion
Source fondatrice	Lloyd, S. P. (1982). Least squares quantization in PCM. IEEE Transactions on Information Theory, 28(2), 129-137.	Davies, D. L., & Bouldin, D. W. (1979). A cluster separation measure. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1(2), 224-227.	Hastie, T., Tibshirani, R., & Friedman, J. (2009). The Elements of Statistical Learning: Data Mining, Inference, and Prediction. Springer Series in Statistics.
Alias	WCSS, within-cluster sum of squares, cluster cohesion	DBI, Davies Bouldin index	elbow analysis, knee detection
Apparentées	5	5	5
Résumé	Inertia, also called Within-Cluster Sum of Squares (WCSS), is a measure of cluster cohesion that quantifies how tightly points are grouped around their cluster centroids. Lower values indicate more compact, cohesive clusters. Inertia is the primary objective function for k-means clustering and has been a fundamental metric since the method's introduction.	The Davies-Bouldin Index, introduced by Davies and Bouldin in 1979, is a metric for evaluating clustering quality based on the average similarity between each cluster and its most similar neighboring cluster. Lower values indicate better clustering, with a minimum of 0 representing perfectly separated, non-overlapping clusters.	The Elbow Method is a heuristic for selecting the optimal number of clusters in partitional clustering. Introduced by Robert Thorndike in 1953, it involves fitting clustering models for increasing numbers of clusters and plotting the within-cluster sum of squares (WCSS) against the number of clusters. The 'elbow' occurs where the rate of WCSS decrease sharply changes, suggesting an optimal cluster count.