Anomaly Detection

  1. Advanced Anomaly Detection Topics
    1. Streaming and Real-Time Anomaly Detection
      1. Concept Drift
        1. Types of Drift
          1. Sudden Drift
            1. Gradual Drift
              1. Incremental Drift
                1. Recurring Drift
                2. Drift Detection Methods
                  1. Statistical Tests
                    1. Window-Based Methods
                      1. Ensemble-Based Detection
                    2. Online Learning Algorithms
                      1. Incremental Learning
                        1. Online Gradient Descent
                          1. Incremental PCA
                            1. Online Clustering
                            2. Adaptive Algorithms
                              1. Adaptive Windowing
                                1. Forgetting Mechanisms
                                  1. Learning Rate Adaptation
                                2. Windowing Techniques
                                  1. Fixed-Size Windows
                                    1. Sliding Windows
                                      1. Landmark Windows
                                        1. Adaptive Windows
                                        2. Real-Time Processing Constraints
                                          1. Latency Requirements
                                            1. Memory Limitations
                                              1. Computational Complexity
                                            2. High-Dimensional Anomaly Detection
                                              1. Curse of Dimensionality
                                                1. Distance Concentration
                                                  1. Sparsity Problems
                                                    1. Computational Challenges
                                                    2. Subspace Methods
                                                      1. Feature Subset Selection
                                                        1. Subspace Clustering
                                                          1. CLIQUE Algorithm
                                                            1. SUBCLU Algorithm
                                                            2. Projected Clustering
                                                            3. Random Projection Techniques
                                                              1. Johnson-Lindenstrauss Lemma
                                                                1. Sparse Random Projections
                                                                  1. Fast Random Projections
                                                                  2. Manifold Learning
                                                                    1. Locally Linear Embedding
                                                                      1. Isomap
                                                                        1. Laplacian Eigenmaps
                                                                      2. Ensemble Methods for Anomaly Detection
                                                                        1. Diversity in Ensembles
                                                                          1. Feature Diversity
                                                                            1. Model Diversity
                                                                              1. Parameter Diversity
                                                                              2. Combination Strategies
                                                                                1. Voting Methods
                                                                                  1. Majority Voting
                                                                                    1. Weighted Voting
                                                                                    2. Averaging Methods
                                                                                      1. Simple Averaging
                                                                                        1. Weighted Averaging
                                                                                        2. Stacking Methods
                                                                                          1. Meta-Learning
                                                                                            1. Blending
                                                                                          2. Specific Ensemble Algorithms
                                                                                            1. Feature Bagging
                                                                                              1. Isolation Forest Ensembles
                                                                                                1. LODA (Lightweight On-line Detector of Anomalies)
                                                                                                2. Dynamic Ensemble Selection
                                                                                                  1. Competence-Based Selection
                                                                                                    1. Clustering-Based Selection
                                                                                                  2. Graph and Network Anomaly Detection
                                                                                                    1. Graph Representation
                                                                                                      1. Adjacency Matrices
                                                                                                        1. Edge Lists
                                                                                                          1. Graph Properties
                                                                                                            1. Degree Distribution
                                                                                                              1. Clustering Coefficient
                                                                                                                1. Path Lengths
                                                                                                              2. Node-Level Anomalies
                                                                                                                1. Degree-Based Methods
                                                                                                                  1. Centrality-Based Methods
                                                                                                                    1. Betweenness Centrality
                                                                                                                      1. Closeness Centrality
                                                                                                                        1. Eigenvector Centrality
                                                                                                                        2. Local Structure Analysis
                                                                                                                        3. Edge-Level Anomalies
                                                                                                                          1. Edge Weight Analysis
                                                                                                                            1. Temporal Edge Patterns
                                                                                                                            2. Subgraph-Level Anomalies
                                                                                                                              1. Motif Analysis
                                                                                                                                1. Community Detection
                                                                                                                                  1. Modularity-Based Methods
                                                                                                                                    1. Spectral Methods
                                                                                                                                      1. Label Propagation
                                                                                                                                      2. Dense Subgraph Detection
                                                                                                                                      3. Dynamic Graph Anomalies
                                                                                                                                        1. Temporal Networks
                                                                                                                                          1. Evolution Patterns
                                                                                                                                            1. Change Detection in Graphs