1)迴歸算法:
- * 最小二乘法(OrdinaryLeast Square)
- * 邏輯迴歸(Logistic Regression)
- * 逐步式迴歸(Stepwise Regression)
- (縮減方法)
- * 多元自適應迴歸樣條(MultivariateAdaptive Regression Splines)
- * 本地散點平滑估計(Locally Estimated Scatterplot Smoothing)
2)基於實例的算法:
- * k-Nearest Neighbor(KNN)
- * 學習矢量量化(Learning Vector Quantization, LVQ)
- * 自組織映射算法(Self-Organizing Map , SOM)
3)基於正則化方法:
- * 嶺迴歸(Ridge Regression)L2
- * 稀疏約束Least Absolute Shrinkage and Selection Operator(LASSO)L1
- * 彈性網絡(Elastic Net)
4)決策樹學習:
- * 分類及迴歸樹(ClassificationAnd Regression Tree, CART)
- * ID3 (Iterative Dichotomiser 3)
- * C4.5
- * Chi-squared Automatic Interaction Detection(CHAID)
- * Decision Stump
- * 隨機森林(Random Forest)
- * 多元自適應迴歸樣條(MARS)
- * 梯度推進機(Gradient Boosting Machine, GBM)
5)基於貝葉斯方法:
- * 樸素貝葉斯算法
- * 平均單依賴估計(AveragedOne-Dependence Estimators, AODE)
- * Bayesian Belief Network(BBN)
-
6)基於核的算法:
- * 支持向量機(SupportVector Machine, SVM)
- * 徑向基函數(Radial Basis Function ,RBF)
- * 線性判別分析(Linear Discriminate Analysis ,LDA)
7)聚類算法:
- * k-Means算法
- * 期望最大化算法(Expectation Maximization, EM)
8)基於關聯規則學習:
- * Apriori算法
- * Eclat算法
9)人工神經網絡:
- * 感知器神經網絡(PerceptronNeural Network)
- * 反向傳遞(Back Propagation)
- * Hopfield網絡
- * 自組織映射(Self-OrganizingMap, SOM)
- * 學習矢量量化(Learning Vector Quantization, LVQ);
10)深度學習:
- * 受限波爾茲曼機(RestrictedBoltzmann Machine, RBN)
- * Deep Belief Networks(DBN)
- * 卷積網絡(Convolutional Network)
- * 堆棧式自動編碼器(Stacked Auto-encoders)
11)降低維度的算法:
- * 主成份分析(PrincipleComponent Analysis, PCA)
- * 偏最小二乘迴歸(Partial Least Square Regression,PLS)
- * Sammon映射
- * 多維尺度(Multi-Dimensional Scaling, MDS)
- * 投影追蹤(ProjectionPursuit)
-
12)集成算法:
- * Boosting
- * Bootstrapped Aggregation(Bagging)
- * AdaBoost
- * 堆疊泛化(Stacked Generalization, Blending)
- * 梯度推進機(GradientBoosting Machine, GBM)
- * 隨機森林(Random Forest)