For each task, we identified a winning team and one or more runners-up teams.
Each task also had a "best student entry" prize, awarded to the best-performing
student-led team.
Task 1: PE Identification
The winners in Task 1 were:
First Place
Robert Bell, Patrick Haffner, and Chris Volinsky (AT&T Research).
First Runner Up
Dmitriy Fradkin (Ask.com).
Second Runner Up
Domonkos Tikk (Budapest University of Technology & Economics), Zsolt
T. Kardkovács (Budapest University of Technology & Economics), Ferenc
P. Szidarovszky (Szidarovszky Ltd. and Budapest University of Technology & Economics),
György Biró (TextMiner Ltd.), and Zoltán Bálint (Budapest
University of Technology & Economics).
Best Student Entry
Karthik Kumara (team leader), Sourangshu Bhattacharya, Mehul Parsana, Shivramkrishnan
K, Rashmin Babaria, Saketha Nath J, and Chiranjib Bhattacharyya (Indian Institute
of Science).
Task 2: Patient Classification
The winners in Task 2 were:
First Place
Domonkos Tikk (Budapest University of Technology & Economics), Zsolt
T. Kardkovács (Budapest University of Technology & Economics), Ferenc
P. Szidarovszky (Szidarovszky Ltd. and Budapest University of Technology & Economics),
György Biró (TextMiner Ltd.), and Zoltán Bálint (Budapest
University of Technology & Economics).
First Runner Up
Ruiping Wang, Yu Su, Ting Liu, Fei Yang, Liangguo Zhang, Dong Zhang, Shiguang
Shan, Weiqiang Wang, Ruixiang Sun, and Wen Gao (Institute of Computing Technology,
Chinese Academy of Sciences).
Second Runner Up
Cas Zhang, Y. Zhou, Q. Wang, and H. Ge (Joint R&D Lab, Chinese Academy
of Sciences).
Third Runner Up
Dmitriy Fradkin (Ask.com).
Best Student Entry
Zhang Cas (IA, PKU).
Task 3: Negative Predictive Value
The winners in Task 3 were:
First Place
William Perrizo and Amal Shehan Perera (DataSURG Group, North Dakota State
University)
Runner Up
Nimisha Gupta and Tarun Agarwal (Strand Life Sciences Pvt. Ltd.)
Best Student Entry
Karthik Kumara (team leader), Sourangshu Bhattacharya, Mehul Parsana, Shivramkrishnan
K, Rashmin Babaria, Saketha Nath J, and Chiranjib Bhattacharyya (Indian Institute
of Science).
KDD Cup Raw Results
Results are listed using the Download ID for privacy.
This year's KDD Cup challenge problem is drawn from the domain of medical
data mining. The tasks are a series of Computer-Aided Detection problems revolving
around the clinical problem of identifying pulmonary embolisms from three-dimensional
computed tomography data. This challenging domain is characterized by:
Multiple instance learning
Non-IID examples
Nonlinear cost functions
Skewed class distributions
Noisy class labels
Sparse data
Many thanks to Siemens Medical Solutions for
providing the CAD data sets, task specifications, and data expertise.
