The Evolving Tree software package

This page hosts the C++ reference implementation for the Evolving Tree algorithm. It is a non-supervised tree-shaped self-organizing neural network that has been especially designed for analyzing and indexing huge data bases.

Copyright information

The software package is © 2004-2006 Jussi Pakkanen, with portions © 2004 Petri Turkulainen. The package is available under the terms of the Gnu GPL, version 2.

Download

The current version of the package is 1.3, which was released May 17. 2006. You can download it here. This version has only minor bugfixes.

Publications

There are currently six publications dealing with the Evolving Tree.

• Jussi Pakkanen, The Evolving Tree, a new kind of self-organizing neural network. Published in the Proceedings of the Workshop on Self-Organizing Maps, 2003.
• Jussi Pakkanen and Jukka Iivarinen, A Novel Self-Organizing Neural Network for Defect Image Classification. Published in the Proceedings of the International Joint Conference on Neural Networks, 2004.
• Jussi Pakkanen, Jukka Iivarinen and Erkki Oja, The Evolving Tree - A Novel Self-Organizing Network for Data Analysis. Published in Neural Processing Letters, Volume 20 Issue 3.
• Jussi Pakkanen and Jukka Iivarinen, Analyzing Large Databases with the Evolving Tree. Published in Proceedings of the Third International Conference on Advances in Pattern Recognition, 2005.
• Jussi Pakkanen, Examining the Behaviour of the Evolving Tree. Published in Proceegins of the 5th Workshop on Self-Organizing Maps, 2005.
• Jussi Pakkanen, Jukka Iivarinen and Erkki Oja, The Evolving Tree, a Hierarchical Tool for Unsupervised Data Analysis. Published in Proceedings of International Joint Conference on Neural Networks, 2005.

A BibTeX file for the above publications is provided for your convenience.