方法对比

	孤立森林 (Isolation Forest)	决策树	高斯混合模型	主成分分析
领域	机器学习	机器学习	机器学习	机器学习
方法族	Machine learning	Machine learning	Machine learning	Machine learning
起源年份	2008	1984	1977	2002
提出者	Liu, F.T., Ting, K.M. & Zhou, Z.-H.	Breiman, Friedman, Olshen & Stone	Dempster, Laird & Rubin (EM algorithm)	Jolliffe, I.T. (textbook); Pearson & Hotelling (origins)
类型	Unsupervised ensemble (random partitioning trees)	Recursive partitioning (if-then rules)	Probabilistic (soft) clustering — mixture model	Unsupervised dimensionality reduction
开创性文献	Liu, F.T., Ting, K.M. & Zhou, Z.-H. (2008). Isolation Forest. IEEE ICDM, 413–422.	Breiman, L., Friedman, J.H., Olshen, R.A. & Stone, C.J. (1984). Classification and Regression Trees. Wadsworth.	Dempster, A.P., Laird, N.M. & Rubin, D.B. (1977). Maximum Likelihood from Incomplete Data via the EM Algorithm. Journal of the Royal Statistical Society: Series B, 39(1), 1–22.	Jolliffe, I.T. (2002). Principal Component Analysis (2nd ed.). Springer.
别名	Isolation Forest (Aykırı Değer Tespiti), iForest, isolation forest anomaly detection	Karar Ağacı (Decision Tree), karar ağacı, classification tree, regression tree	Gaussian Karışım Modeli (GMM Kümeleme), GMM, GMM clustering, mixture of Gaussians	Temel Bileşenler Analizi (PCA), PCA, principal components analysis, Karhunen-Loève transform
相关	5	5	4	3
摘要	Isolation Forest is an unsupervised machine-learning method for anomaly and outlier detection, introduced by Liu, Ting and Zhou in 2008, that isolates anomalies through random partitioning of the data. It works without any labelled anomaly data and scales to high-dimensional datasets.	A Decision Tree is an interpretable classification and regression method, formalised by Breiman, Friedman, Olshen and Stone in their 1984 CART framework, that partitions the data with hierarchical if-then rules. Each split sends observations down one branch or another until a prediction is read off the leaf.	A Gaussian Mixture Model is a probabilistic clustering method that models the data as a weighted mixture of several Gaussian distributions, fitted with the Expectation–Maximization algorithm formalized by Dempster, Laird & Rubin in 1977. It is a generalization of K-means in which each cluster can take its own shape, size, and orientation.	Principal Component Analysis (PCA) is an unsupervised dimensionality-reduction method — given its modern textbook treatment by Ian Jolliffe (2002) — that compresses high-dimensional data into fewer dimensions while preserving the maximum possible variance. It re-expresses correlated variables as a small set of uncorrelated principal components ordered by how much of the data's variation each one captures.