KDD Cup -- Competition in Data Mining and Knowledge Discovery

KDnuggets : Datasets : KDD Cup

KDD Cup

KDD Cup is the annual Data Mining and Knowledge Discovery competition organized by ACM Special Interest Group on Knowledge Discovery and Data Mining, the leading professional organization of data miners. Here are the results for

KDD Cup 2004, features tasks in particle physics and bioinformatics evaluated on a variety of different measures.
KDD Cup 2003, focuses on problems motivated by network mining and the analysis of usage logs.
KDD Cup 2002, focus: bioinformatics and text mining
KDD Cup 2001, focus: bioinformatics and drug discovery
KDD Cup 2000, focus: web mining tasks
KDD Cup 1999, focus: intrusion detection and report
KDD Cup 1998, focus: direct marketing, list with best donation value; best report
KDD Cup 1997, focus: predicting most likely donors for a charity

1999

There were two competitions: a classifier learning contest, and a knowledge discovery "report" contest.

In Classifier Learning contest, the goal was to build a predictive model for identifying network intrusions. 24 entries were submitted.
Winners were:

Dr. Bernhard Pfahringer of the Austrian Research Institute for AI.
Itzhak Levin from LLSoft, Inc (Israel), using Kernel Miner.
Team from MP13 in Moscow, Russia.

Here are the full results

In Knowledge Discovery contest, the goal was to apply a range of knowledge discovery techniques to the same data used in the 1998 competition, and a produce a report summarizing the findings.
Winners were:

Two teams were co-winners: J. Georges and A.Milley (SAS) and S. Rosset and A. Inger (Amdocs, Israel).
An honorable mention was awarded to: Paola Sebastiani, Marco Ramoni, and Alexander Crea of Bayesware Ltd.

1998

This competition also used data from a charity; the goal was to select the best list to mail a solicitation. The winners were chosed based on the net revenue generated on the hold-out or validation sample. 21 teams completed the challenge and submitted results (although many more downloaded the data).

Winners were:

1st place: Urban Science Applications, Inc. with their software GainSmarts.
2nd place: SAS Institute, Inc. with their software Enterprise Miner.
3rd place: Quadstone Limited with their software Decisionhouse.

Here are full KDD Cup 1998 results.

1997

The goal was to predict who is most likely to donate to a charity. Contestants were evaluated on the accuracy on the validation data set. 45 teams participated.

Winners were:

Shared 1-2 place: Charles Elkan (UCSD) with BNB software; and Urban Science Applications, Inc. with their software Gain, Direct Marketing Selection System.
3rd Place: SGI with Mineset software.

Here are full KDD Cup 1997 results.

KDnuggets : Publications : KDD Cup