The Anomaly Detection Extension for RapidMiner comprises the most well know unsupervised anomaly detection algorithms, assigning individual anomaly scores to data rows of example sets. It allows you to find data, which is significantly different from the normal, without the need for the data being labeled.
Some of the algorithms are:
- Local Outlier Factor (LOF)
- k-NN Global Anomaly Score
- Connectivity-based Outlier Factor (COF)
- Local Correlation Integral (LOCI)
- Local Outlier Probability (LoOP)
- Cluster-based Local Outlier Factor (CBLOF)
More information and usage examples can be found on the author's homepage
This plugin was developed for the Open Source version of RapidMiner, which is still available here. Development was continued for the commercial version of Rapidminer Studio until version 9.7 by Martin Liebig in this fork. After RapidMiner was bought and renamed to Altair RapidMiner and then later to Altair® AI Studio, I completely lost track and interest of whether it still works with this software.
- In RapidMiner, go to Help->Updates and Extensions (Marketplace) and search for “anomaly detection” and click on “Install”, or
- Copy the jar file to the “lib/plugins” directory of RapidMiner
Copyright 2008-2013 Deutsches Forschungszentrum fuer Kuenstliche Intelligenz
Copyright 2008-2019 Markus Goldstein
This is free software. Licensed under the GNU AGPL, Version 3.
There is NO WARRANTY, to the extent permitted by law.
Markus Goldstein
Mennatallah Amer
Johann Gebhardt
Patrick Kalka
Ahmed Elsawy
This Software is supported by ...
