今天整理了一下简单的机器学习算法。打算按照这个提纲学习代码实现,不一定按顺序,但希望自己每学习一个都能了解透彻。打算寒假前就仔细研究这块了。
一、Regression(回归)
- Ordinary Least Squares普通最小二乘法
- Logistic Regression逻辑回归
- Stepwise Regression逐步回归
- Multivariate Adaptive Regression Splines (MARS)多元自适应回归
- Locally Estimated Scatterplot Smoothing (LOESS)本地散点平滑估计
- k-Nearest Neighbour (kNN)
- Learning Vector Quantization (LVQ)学习矢量量化
- Self-Organizing Map (SOM)自组织映射算法
- Ridge Regression岭回归数值计算方法
- Least Absolute Shrinkage and Selection Operator (LASSO)至少绝对的收缩和选择算子
- Elastic Net弹性网络
- Classification and Regression Tree (CART)分类回归树
- Iterative Dichotomiser 3 (ID3)迭代二叉树3代
- C4.5
- Chi-squared Automatic Interaction Detection (CHAID)卡方自动交互检测
- Decision Stump单层决策树
- Random Forest随机森林
- Multivariate Adaptive Regression Splines (MARS)多元自适应回归样条
- Gradient Boosting Machines (GBM)梯度推进机
- Naive Bayes朴素贝叶斯
- Averaged One-Dependence Estimators (AODE)平均单依赖估计
- Bayesian Belief Network (BBN)贝叶斯信念网络
- Support Vector Machines (SVM)支持向量机
- Radial Basis Function (RBF)径向基函数
- Linear Discriminate Analysis (LDA)线性鉴别分析
- k-Means
- Expectation Maximisation (EM)期望最大化算法
八、关联规则学习
- Apriori 算法
- Eclat 算法
- Perceptron感知器神经网络
- Back-Propagation反向传递
- Hopfield Network(Hopfield网络)
- Self-Organizing Map (SOM)自组织映射
- Learning Vector Quantization (LVQ)学习矢量量化
- Restricted Boltzmann Machine (RBM)受限波尔兹曼机
- Deep Belief Networks (DBN)深度信念网络
- Convolutional Network回旋神经网
- Stacked Auto-encoders堆栈式自动编码器
- Principal Component Analysis (PCA)主成分分析
- Partial Least Squares Regression (PLS)偏最小二乘回归
- Sammon Mapping
- Multidimensional Scaling (MDS)多维尺度
- Projection Pursuit投影寻踪
- Boosting
- Bootstrapped Aggregation (Bagging)自展输入引导式聚合
- AdaBoost
- Stacked Generalization (blending)堆栈泛化
- Gradient Boosting Machines (GBM)梯度Boosting机器
- Random Forest随机森林