Anomaly Detection
Introduction to Anomaly Detection
Anomaly Detection
Anomaly/Outlier/Novelty Detection
This course includes
6 articles
14 mins reading time
6 videos • 1 hr 27 mins
Course content
6 sections • 47 topics • Read: 14 mins • Video: 1 hr 27 mins
Anomaly Detection
What is Anomaly?
Anomaly Detection
What to do with Outliers ?
Anomaly Detection Methods
Known Methods
Unsupervised Methods
Anomaly/Outlier/Novelty Detection Methods | Explained with Example
Elliptic Envelope
Use Case
Intuition
Issue with Euclidean Distance
Mahalanobis Distance (Solution)
Problem
Solution
Goal
Minimum Covariance Determinant (MCD) Algorithm
Limitations
Elliptic Envelope | Anomaly Detection | Minimum Covariance Determinant (MCD)| Explained with Example
One Class SVM
Use Case (Novelty Detection)
Intuition
Problem
Solution
Goal
One Class SVM
Explanation of Terms
Working
One Class Support Vector Machine (OC-SVM) | Anomaly Detection | Explained with Examples
Local Outlier Factor
Use Case
Intuition
Problem
Solution
Goal
Local Outlier Factor (LOF)
LOF Score Calculation
LOF Score Interpretation
Local Outlier Factor (LOF) | Anomaly Detection | Explained with Example
Isolation Forest
Use Case
Intuition
Problem
Solution
Goal
Isolation Forest (iForest)
Algorithm
Scoring Function
Anomaly Score
Drawbacks
Isolation Forest (iForest) | Anomaly Detection | Explained with Example
RANSAC
Use Case
Intuition
Problem
Solution
RANSAC Algorithm
How Many Iterations ‘k’ ?
Random Sampling Consensus (RANSAC) | Anomaly Detection | Explained with Example
Elliptic Envelope
Elliptic Envelope & Minimum Covariance Determinant (MCD)
One Class SVM
One Class SVM
Local Outlier Factor
Local Outlier Factor
Isolation Forest
Isolation Forest
RANSAC
RANdom SAmple Consensus