Anomaly detection using ensembles

Authors:
Larry Shoemaker;Lawrence O. Hall
Affiliations:
Computer Science and Engineering, University of South Florida, Tampa, FL;Computer Science and Engineering, University of South Florida, Tampa, FL
Venue:
MCS'11 Proceedings of the 10th international conference on Multiple classifier systems
Year:
2011

Citing 16
Cited 1

Random Forests

Machine Learning
Mining distance-based outliers in near linear time with randomization and a simple pruning rule

Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining
Feature bagging for outlier detection

Proceedings of the eleventh ACM SIGKDD international conference on Knowledge discovery in data mining
Introduction to Data Mining, (First Edition)

Introduction to Data Mining, (First Edition)
Estimating the Support of a High-Dimensional Distribution

Neural Computation
Outlier detection by active learning

Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining
An introduction to ROC analysis

Pattern Recognition Letters - Special issue: ROC analysis in pattern recognition
A Comparison of Decision Tree Ensemble Creation Techniques

IEEE Transactions on Pattern Analysis and Machine Intelligence
Intrusion detection in computer networks by a modular ensemble of one-class classifiers

Information Fusion
One-Class Classification by Combining Density and Class Probability Estimation

ECML PKDD '08 Proceedings of the 2008 European Conference on Machine Learning and Knowledge Discovery in Databases - Part I
McPAD: A multiple classifier system for accurate payload-based anomaly detection

Computer Networks: The International Journal of Computer and Telecommunications Networking
Anomaly detection: A survey

ACM Computing Surveys (CSUR)
The use of the area under the ROC curve in the evaluation of machine learning algorithms

Pattern Recognition
Data Editing Techniques to Allow the Application of Distance-Based Outlier Detection to Streams

ICDM '10 Proceedings of the 2010 IEEE International Conference on Data Mining
LIBSVM: A library for support vector machines

ACM Transactions on Intelligent Systems and Technology (TIST)
Ensemble learning with imbalanced data

Ensemble learning with imbalanced data

Automatic network intrusion detection: Current techniques and open issues

Computers and Electrical Engineering

Quantified Score

Hi-index	0.00

Visualization

Abstract

We show that using random forests and distance-based outlier partitioning with ensemble voting methods for supervised learning of anomaly detection provide similar accuracy results when compared to the same methods without partitioning. Further, distance-based outlier and one-class support vector machine partitioning and ensemble methods for semi-supervised learning of anomaly detection also compare favorably to the corresponding nonensemble methods. Partitioning and ensemble methods would be required for very large datasets that need distributed computing approaches. ROC curves often show significant improvement from increased true positives in the low false positive range for ensemble methods used on several datasets.