手法を比較

	Self-supervised DBSCAN	HDBSCAN	K-means クラスタリング
分野	機械学習	機械学習	機械学習
系統	Machine learning	Machine learning	Machine learning
提唱年	2018–2021	2013	1967 (formalized 1982)
提唱者	Ester et al. (DBSCAN base); pipeline pattern established in multiple works c. 2018–2021	Campello, R. J. G. B.; Moulavi, D.; Sander, J.	MacQueen, J. B.; Lloyd, S. P.
種類	Two-stage pipeline (self-supervised pre-training + density-based clustering)	Hierarchical density-based clustering	Partitional clustering
原典	Ester, M., Kriegel, H.-P., Sander, J., & Xu, X. (1996). A density-based algorithm for discovering clusters in large spatial databases with noise. In Proceedings of the 2nd International Conference on Knowledge Discovery and Data Mining (KDD-96), pp. 226–231. AAAI Press.	Campello, R. J. G. B., Moulavi, D., & Sander, J. (2013). Density-Based Clustering Based on Hierarchical Density Estimates. In J. Pei et al. (Eds.), Advances in Knowledge Discovery and Data Mining. PAKDD 2013. Lecture Notes in Computer Science, vol. 7819 (pp. 160–172). Springer, Berlin, Heidelberg.	Lloyd, S. P. (1982). Least squares quantization in PCM. IEEE Transactions on Information Theory, 28(2), 129–137.
別名	SSL-DBSCAN, self-supervised density clustering, contrastive DBSCAN, representation-based DBSCAN	HDBSCAN, Hierarchical DBSCAN, hierarchical density-based clustering, HDBSCAN*	k-means clustering, Lloyd's algorithm, k-means partitioning, hard k-means
関連	5	3	4
概要	Self-supervised DBSCAN is a two-stage unsupervised pipeline that first trains a neural encoder on a pretext task — such as contrastive learning or masked reconstruction — to produce compact, semantically meaningful embeddings from unlabeled data, and then applies DBSCAN in the resulting embedding space to discover arbitrarily shaped clusters without requiring any class labels.	HDBSCAN (Hierarchical Density-Based Spatial Clustering of Applications with Noise) is a density-based clustering algorithm introduced by Campello, Moulavi, and Sander in 2013. It extends DBSCAN by building a full hierarchy of density-based clusters across all density scales and then extracting a stable flat partition, making it robust to datasets where cluster densities vary substantially across regions.	K-means is a classic unsupervised partitional clustering algorithm that divides a dataset into K non-overlapping groups by iteratively assigning each observation to its nearest centroid and updating centroids as the mean of their assigned points. It is one of the most widely used exploratory tools in machine learning and data analysis.