|
| ML domain | ML subdomains | Algorithms | Tools |
|
| Machine learning | Unsupervised learning | Clustering | (i) K-means
(ii) K-median
(iii) Hierarchical clustering (HC) | (i) Spatial cluster (SC)
(ii) Gaussian mixture model (GMM) | (i) Local outlier factor (LOF)
(ii) Neighbour-based clustering (NBC)
(iii) Parzen windows (PW) |
| Dimensionality reduction | (i) Principal component analysis (PCA)
(ii) Linear discriminant analysis (LDA)
(iii) Kernel density estimator (KDE) | (i) Kernel principal component analysis (K-PCA)
(ii) Singular value decomposition (SVD) | (i) t-distributed stochastic neighbour embedding (t-SNE)
(ii) Uniform manifold approx. and projection (UMAP)
(iii) Self-organizing maps (SOM) |
| Supervised learning | Regression | (i) Neural networks (NN)
(ii) NN, back propagation (BP)
(iii) NN, convolutional neural network (CNN)
(iv) NN, extreme learning machine (ELM)
(v) NN, long-short term memory (LSTM)
(vi) NN, deep learning (DL)
(vii) NN, adaptive neuro-fuzzy inference system (ANFIS) | (i) NN, multilayer perception (MLP)
(ii) NN, radial basis function (RBF)
(iii) NN, recurrent neural network (RNN)
(iv) Linear regression (LR)
(v) Polynomial regression (PR)
(vi) Fuzzy regression (FR)
(vii) Bayesian regression (BR)
(viii) Lasso regression (LASSO) | (i) Locally weighted regression (LWR)
(ii) Support vector machine (SVM)- regressor
(iii) Gradient boosting (GBoost)
(iv) Random forest (RF)- regressor
(v) K-nearest neighbor (KNN)-regressor
(vi) Gaussian process regression (GPR) |
| Classification | (i) Decision tree (DT)
(ii) Gradient boosting (GBoost)
(iii) Naive bayes (NB)
(iv) Bayesian network (BN)
(v) Kernel method (KM)
(vi) Support vector machine (SVM) | (i) Adaptive support vector machine (ASVM)
(ii) Learning vector quantization (LVQ)
(iii) Linear discriminant analysis (LDA)
(iv) Stochastic gradient descent (SGD) | (i) K-nearest neighbor (KNN)
(ii) Quadratic discriminant analysis (QDA)
(iii) Random forest (RF)
(iv) Logistic regression (LogR)
(v) Pattern recognition (PattR)
(vi) Multi-layer perception (MLP) |
| Reinforcement learning | | (i) Adaptive heuristic critic (AHC)
(ii) Deep deterministic policy gradient (DDPG)
(iii) Q-learning (QL) | (i) Approximate dynamic programming (ADP)
(ii) Proximal policy optimization (PPO)
(iii) Deep Q- learning (DQL) | (i) State-action-reward-state-action (SARSA)
(ii) Temporal difference learning (TD)
(iii) Trust region policy optimization (TRPO) |
|
