Lehr- und Forschungseinheit für Datenbanksysteme Ludwig-Maximilians-Universität München
Institut für Informatik
Lehr- und Forschungseinheit für Datenbanksysteme
University of Munich
Institute for Computer Science
Database and Information Systems

elki
Environment for
DeveLoping
KDD-Applications
Supported by Index-Structures

[ Background | Release History | Bug Reports | Agenda | Contact | Contributors | Publications ]


Background

Data mining research leads to many algorithms for similar tasks. A fair and useful comparison of these algorithms is difficult due to several reasons:

On the other hand, efficient data management tools like index-structures can show considerable impact on data mining tasks and are therefore useful for a broad variety of algorithms.

In ELKI, data mining algorithms and data management tasks are separated and allow for an independent evaluation. This separation makes ELKI unique among data mining frameworks like Weka or YALE and frameworks for index structures like GiST. At the same time, ELKI is open to arbitrary data types, distance or similarity measures, or file formats. The fundamental approach is the independence of file parsers or database connections, data types, distances, distance functions, and data mining algorithms. Helper classes, e.g. for algebraic or analytic computations are available for all algorithms on equal terms.

With the development and publication of ELKI, we humbly hope to serve the data mining and database research community beneficially. The framework is free for scientific, non-commercial usage. In case of application of ELKI in scientific publications, it should be cited with the most recent publication below (see publications).

Please note: Algorithms in ELKI are not tuned by implementation for individual efficiency but for fair comparability within ELKI. Runtime comparisons of algorithms using ELKI implementations with algorithms using implementations not based on the ELKI framework are likely to produce misleading results.

Release-History

0.3 (2010, March, 31)
New release with a focus on outlier detection methods and visualization (see publication P3),
along with some general refactoring that improves memory efficiency and performance and
a minimalistic GUI for interactive parameterization of ELKI algorithms. requires Java SE 6
0.2.1 (2009, July, 13)
bug-fix release for 0.2, solving an issue with running PreDeCon, FourC and COPAC. requires Java SE 6
0.2 (2009, July, 6)
publication of an extended version (see publication P2),
including distance measures for time series and a visualization tool for kNN-queries on time series.

Substantial changes in the infrastructure, see documentation.

requires Java SE 6
0.1 (2008, July, 10)
first publication of the framework (see publication P1) requires Java SE 6

Bug-Reports

Although the ELKI-Framework is work of some years, the development is effected along the way and is prone to bugs and errors. We welcome any bug-reports. We also appreciate any comments and suggestions. You can contact us by e-mail: elki (at) dbs ifi lmu de

Agenda

Following issues are intended to be addressed in the near future:

Contact

For questions and suggestions regarding ELKI you can write to elki (at) dbs ifi lmu de

You can also subscribe the mailing list for users of ELKI via https://tools.rz.ifi.lmu.de/mailman/listinfo/elki-user, to exchange questions and ideas among other users or to get announcements (e.g., new releases, major changes) by the ELKI team.

Contributors

Development and Maintenance

The ELKI core team ( elki (at) dbs ifi lmu de ):
 

picture of Elke picture of Erich picture of Arthur
Elke Achtert Erich Schubert Arthur Zimek

Development Contributions

Noemi Andor
picture of Thomas
Thomas Bernecker
picture of Franz
Franz Graf
Ahmed Hettab
picture of Peer
Peer Kröger
Heidi Kolb
Simon Mittermüller
Simon Paradies
Lisa Reichert
picture of Marisa
Marisa Thoma
picture of Steffi
Steffi Wanka
Remigius Wojdanowski
picture of Andreas
Andreas Züfle

Publications

[P3] Elke Achtert, Hans-Peter Kriegel, Lisa Reichert, Erich Schubert, Remigius Wojdanowski, Arthur Zimek:
Visual Evaluation of Outlier Detection Models.
15th International Conference on Database Systems for Advanced Applications (DASFAA 2010), Tsukuba, Japan, 2010.
[EE (springerlink) | poster (pdf)]
[P2] Elke Achtert, Thomas Bernecker, Hans-Peter Kriegel, Erich Schubert, Arthur Zimek:
ELKI in Time: ELKI 0.2 for the Performance Evaluation of Distance Measures for Time Series.
11th International Symposium on Spatial and Temporal Databases (SSTD 2009), Aalborg, Denmark, 2009.
[EE (springerlink) | poster (pdf)]
[P1] Elke Achtert, Hans-Peter Kriegel, Arthur Zimek:
ELKI: A Software System for Evaluation of Subspace Clustering Algorithms.
20th International Conference on Scientific and Statistical Database Management (SSDBM 2008), Hong Kong, China, 2008.
[EE (springerlink) | poster (pdf)]