Hauptnavigation

Go to the list of all software

RapidMiner Data Stream Plugin (formerly: YALE Concept Drift Plugin) - Machine Learning of Time-Varying and Drifting Concepts from Data Streams

Description:

The data stream plugin (formerly: concept drift plugin) for RapidMiner (formerly: YALE (Yet Another Learning Environment)), a freely available open-source environment for machine learning, data mining, and knowledge discovery, extends RapidMiner by operators for handling real and simulated concept drift in time-varying data streams.

The data stream mining and concept drift handling operators provided in this plugin can be combined with all other RapidMiner operators. For example, the audio and text preprocessing of the RapidMiner package can be used to detect and handle concept changes in audio and text data streams and all machine learning methods for classification available in RapidMiner (and WEKA) can be combined with the concept drift handling frameworks. However, some of these frameworks require the learners to be able to estimate their classification performance.

The concept drift handling frameworks provided in this plugin include:

  • baseline strategies for comparison: learning with complete (full) or no memory
  • time windows of fixed or adaptive size
  • batch or example selection
  • global age-based or local performance-based example weighting strategies
  • an ensemble method using knowledge-based sampling (KBS-stream)

The data stream plugin and its source code can be obtained from the RapidMiner download page. For installing and using the plugin, simply copy the file rapidminer-datastream-4.0beta.jar into the lib/plugins/ subdirectory of your RapidMiner installation. On Microsoft Windows systems, you can alternatively use the Windows auto-installer (rapidminer-datastream-4.0beta-installer.exe).

In case of questions, please contact Ralf Klinkenberg.

Link:

http://yale.sf.net/

Software File:

Authors:

Klinkenberg, Ralf

Projects:

SFB 475 subproject A4

Publications:

Mierswa/etal/2003a Mierswa, Ingo and Klinkenberg, Ralf and Fischer, Simon and Ritthoff, Oliver. A Flexible Platform for Knowledge Discovery Experiments: YALE -- Yet Another Learning Environment. In LLWA 03 - Tagungsband der GI-Workshop-Woche Lernen - Lehren - Wissen - Adaptivitat, 2003.


Klinkenberg/Rueping/2003a Klinkenberg, Ralf and RĂ¼ping, Stefan. Concept Drift and the Importance of Examples. In Franke, Jurgen and Nakhaeizadeh, Gholamreza and Renz, Ingrid (editors), Text Mining -- Theoretical Aspects and Applications, pages 55--77, Berlin, Germany, Physica-Verlag, 2003.  


Klinkenberg/Renz/98b Klinkenberg, Ralf and Renz, Ingrid. Adaptive Information Filtering: Learning Drifting Concepts. In Wysotzki, F. and Geibel, P. and Schadler, K. (editors), Beitrage zum Treffen der GI-Fachgruppe 1.1.3 Maschinelles Lernen (FGML-98), No. 98/11, pages 98--105, Germany, Fachbereich Informatik, TU Berlin, 1998.


Klinkenberg/Renz/98a Klinkenberg, Ralf and Renz, Ingrid. Adaptive Information Filtering: Learning in the Presence of Concept Drifts. In Sahami, Mehran and Craven, Mark and Joachims, Thorsten and McCallum, Andrew (editors), Workshop Notes of the ICML/AAAI-98 Workshop \em Learning for Text Categorization, pages 33--40, Menlo Park, CA, USA, AAAI Press, 1998.


Klinkenberg/Joachims/2000a Klinkenberg, Ralf and Joachims, Thorsten. Detecting Concept Drift with Support Vector Machines. In Langley, Pat (editors), Proceedings of the Seventeenth International Conference on Machine Learning (ICML), pages 487--494, San Francisco, CA, USA, Morgan Kaufmann, 2000.


Klinkenberg/98a Klinkenberg, Ralf. Maschinelle Lernverfahren zum adaptiven Informationsfiltern bei sich verandernden Konzepten. Fachbereich Informatik, Universitat Dortmund, Germany, 1998.


Klinkenberg/2004a Klinkenberg, Ralf. Learning Drifting Concepts: Example Selection vs. Example Weighting. In Intelligent Data Analysis (IDA), Special Issue on Incremental Learning Systems Capable of Dealing with Concept Drift, Vol. 8, No. 3, pages 281--300, 2004.  


Klinkenberg/2003a Klinkenberg, Ralf. Predicting Phases in Business Cycles Under Concept Drift. In Hotho, Andreas and Stumme, Gerd (editors), LLWA 2003 -- Tagungsband der GI-Workshop-Woche \em Lehren -- Lernen -- Wissen -- Adaptivitat, Proceedings of the Workshop Week \em Teaching -- Learning -- Knowledge -- Adaptivity of the National German Computer Science Society (GI) / Annual Workshop on Machine Learning, pages 3--10, Karlsruhe, Germany, 2003.


Fischer/etal/2002a Fischer, Simon and Klinkenberg, Ralf and Mierswa, Ingo and Ritthoff, Oliver. \sc Yale: Yet Another Learning Environment -- Tutorial. No. CI-136/02, Collaborative Research Center 531, University of Dortmund, Dortmund, Germany, 2002.


Scholz/Klinkenberg/2005a Scholz, Martin and Klinkenberg, Ralf. An Ensemble Classifier for Drifting Concepts. In Gama, J. and Aguilar-Ruiz, J. S. (editors), Proceedings of the Second International Workshop on Knowledge Discovery in Data Streams, pages 53--64, Porto, Portugal, 2005.


Scholz/Klinkenberg/2006a Scholz, Martin and Klinkenberg, Ralf. Boosting Classifiers for Drifting Concepts. No. 6/06, Collaborative Research Center on the Reduction of Complexity for Multivariate Data Structures (SFB 475), University of Dortmund, Dortmund, Germany, 2006.


Scholz/Klinkenberg/2006b Scholz, Martin and Klinkenberg, Ralf. Boosting Classifiers for Drifting Concepts. In Intelligent Data Analysis (IDA), Special Issue on Knowledge Discovery from Data Streams, Vol. 11, No. 1, pages 3--28, 2007.  


Mierswa/etal/2006a Mierswa, Ingo and Wurst, Michael and Klinkenberg, Ralf and Scholz, Martin and Euler, Timm. YALE: Rapid Prototyping for Complex Data Mining Tasks. In Tina Eliassi-Rad and Lyle H. Ungar and Mark Craven and Dimitrios Gunopulos (editors), Proceedings of the 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD 2006), pages 935--940, ACM, New York, USA, ACM Press, 2006.