Knowledge Discovery and Data Mining Professional Interest Community (KDD PIC) - KDD External Recognition


Parallel Machine Learning Package

The toolbox enables the application of machine learning tools to large data sets by distributing the required computations to computing nodes in a parallel fashion. The toolbox can work on various types of architecture, from multi-core machines to Blue Gene.

http://www.alphaworks.ibm.com/tech/pml

Winning Record of Data Mining Competitions

Competition

Domain

Description

Number of Submissions

Our Finish

KDD Cup 2007 Task 1

Movie Reviews

(Netflix Data)

Predict who will rate a specific movie in 2006, using historical data from 1998 - 2005

39

Runner-up

KDD Cup 2007 Task 2

Movie Reviews

(Netflix Data)

Estimate the number of reviews a specific movie will receive in 2006

34

First

KDD Cup 2008 Task 1

Breast Cancer Identification

Rank patients in terms of the likelihood of being cancerous

110

First

KDD Cup 2008 Task 2

Breast Cancer Identification

Produce a maximal list of non-cancerous patients

110

First

INFORMS 2008 Task 1

Pneumonia

Identification

Identify which patients are likely to contract nosocomial pneumonia during a hospital stay

Approx 10

First

KDD Cup 2009

Customer Marketing

Identify telecom customers likely to attrite (churn), purchase new services (propensity), and purchase upgrades (upsell)

79

First