A Brief Explanation of 8 Anomaly Detection Methods with Python

   Anomaly detection can be done by applying several methods in data analysis. I explained my previous tutorials on how to detect anomalies in a dataset by applying methods like Isolation Forest, Local Outlier Factor, Elliptical Envelope, One-Class SVM,  DBSCAN, Gaussian Mixture, K-means, and Kernel Density.

   We applied the classes provided by Scikit-Learn API for these models. The sample dataset is created randomly by using create_blob() function and anomalies are detected by using each methods. Both data and the result are visualized in a plot to confirm visually. The Python source codes are provided for all tutorials.
   I summarised the above mentioned anomaly detection methods in this tutorial. Here, we'll briefly address the following topics.
  1. What is anomaly detection?
  2. Isolation Forest method
  3. Local Outlier Factor method
  4. Elliptical Envelope method
  5. One-Class SVM method
  6. The DBSCAN method
  7. Gaussian Mixture method
  8. K-means method
  9. Kernel Density method
   Let's get started.

What is Anomaly Detection?

   Anomaly (or outlier, noise, novelty) is an element with the properties that differ from the majority of the observation data. Anomalies may define the errors, extremes, or abnormal cases in observation data. Identifying those anomaly samples in a dataset is called anomaly detection in machine learning and data analysis.
   The clustering algorithms are one of the main methods used in this field. Grouping the samples in observation data by their density level can help to extract scarce or rare cases in a dataset. The below plot shows the detected anomalies in a given dataset. The threshold that defines the abnormality level of sample, can be defined according to data content and user's choice in analysis.




Isolation Forest Method

   Isolation forest is a learning algorithm for anomaly detection by isolating the instances in the dataset. The algorithm creates isolation trees (iTrees), holding the path length characteristics of the instance of the dataset, and Isolation Forest (iForest) applies no distance or density measures to detect anomalies.
   The tutorial explains how to detect anomaly in the dataset by using the Isolation Forest method in Python.


Local Outlier Factor Method

   The Local Outlier Factor is an algorithm to detect anomalies in observation data. Measuring the local density score of each sample and weighting their scores are the main concept of the algorithm. By comparing the score of the sample to its neighbors, the algorithm defines the lower density elements as anomalies in data.
   The tutorial explains how to detect anomaly in a dataset by using the Local Outlier Factor method in Python.


Elliptical Envelope Method

   The Elliptical Envelope method detects the outliers in a Gaussian distributed data. Scikit-learn API provides the EllipticEnvelope class to apply this method for anomaly detection.
   The tutorial explains how to detect the anomalies by using the Elliptical Envelope method in Python.


One-Class SVM Method

  A One-class classification method is used to detect the outliers and anomalies in a dataset. Based on Support Vector Machines (SVM) evaluation, the One-class SVM applies a One-class classification method for novelty detection.
   The tutorial briefly explains how to detect anomaly in a dataset by using the One-class SVM method in Python.


The DBSCAN Method

   The DBSCAN (Density-Based Spatial Clustering of Applications with Noise) is a density-based clustering algorithm. The main principle of this algorithm is that it finds core samples in a dense area and groups the samples around those core samples to create clusters. The samples in a low-density area become the outliers. We'll focus on finding out those outliers in this tutorial.
   The tutorial briefly explains how to detect anomaly in a dataset by using the DBSCAN method in Python.


Gaussian Mixture Method

   The Gaussian Mixture is a probabilistic model to represent a mixture of multiple Gaussian distributions on population data. The model is widely used in clustering problems.
   The tutorial explains how to detect anomalies in a dataset by using a Gaussian Mixture method in Python.


The K-means method

   The K-means clustering method is mainly used for clustering purposes. I experimented to apply this model for anomaly detection and it worked for my test scenario. Technically, we can figure out the outliers by using the K-means method.
   The tutorial explains how to detect outliers for regression data by applying the KMeans class of Scikit-learn API in Python.


Kernel Density Method

   The Kernel Density estimation is a method to estimate the probability density function of a random variables. We can apply this model to detect outliers in a dataset.
   The tutorial explains how to detect the outliers of regression data by applying the KernelDensity class of Scikit-learn API in Python.

   Anomaly detection is an interesting and important topic in machine learning nowadays. Fraud detection, sensor data controlling, system health or disturbance monitoring, and other event detection problems can be solved by applying the anomaly detection methods.

   I hope those examples will help you to learn the first steps in an anomaly detection methods with Python and encourage you to investigate further those analysis.


No comments:

Post a Comment