Results
KDD Cup 2009: Customer relationship prediction
Winners of KDD Cup 2009: Fast Track
- First Place: IBM Research
Ensemble Selection for the KDD Cup Orange Challenge - First Runner Up: ID Analytics, Inc
KDD Cup Fast Scoring on a Large Database - Second Runner Up: Old dogs with new tricks (David Slate, Peter W. Frey)
Winners of KDD Cup 2009: Slow Track
- First Place: University of Melbourne
University of Melbourne entry - First Runner Up: Financial Engineering Group, Inc. Japan
Stochastic Gradient Boosting - Second Runner Up: National Taiwan University, Computer Science and Information Engineering
Fast Scoring on a Large Database using regularized maximum entropy model, categorical/numerical balanced AdaBoost and selective Naive Bayes
Full Results: Fast Track
| Rank | Team Name | Method | AUC | |||
| Churn | Appetency | Upselling | Score | |||
| 1 | IBM Research | Final Submission | 0.7611 | 0.8830 | 0.9038 | 0.8493 |
| 2 | ID Analytics, Inc | DT | 0.7565 | 0.8724 | 0.9056 | 0.8448 |
| 3 | Old dogs with new tricks | Our own method | 0.7541 | 0.8740 | 0.9050 | 0.8443 |
| 4 | Crusaders | Joint Score Technique | 0.7569 | 0.8688 | 0.9034 | 0.8430 |
| 5 | Financial Engineering Group, Inc. Japan | boosting | 0.7498 | 0.8732 | 0.9057 | 0.8429 |
| 6 | LatentView Analytics | Boosting | 0.7579 | 0.8670 | 0.9034 | 0.8428 |
| 7 | Data Mining | Logistic | 0.7580 | 0.8659 | 0.9034 | 0.8424 |
| 8 | StatConsulting (K.Ciesielski, M.Sapinski, M.Tafil) | AdvancedMiner | 0.7544 | 0.8723 | 0.8997 | 0.8421 |
| 9 | Sigma | Decision Tree Algo | 0.7568 | 0.8644 | 0.9034 | 0.8415 |
| 10 | Analytics | CART | 0.7564 | 0.8644 | 0.9034 | 0.8414 |
| 11 | Ming Li & Yuwei Zhang | me | 0.7507 | 0.8683 | 0.9050 | 0.8413 |
| 12 | Hungarian Academy of Sciences | fri4 | 0.7496 | 0.8683 | 0.9042 | 0.8407 |
| 13 | Oldham Athletic Reserves | tiberius10 | 0.7492 | 0.8699 | 0.9026 | 0.8406 |
| 14 | Swetha | Logistic | 0.7550 | 0.8659 | 0.8996 | 0.8401 |
| 15 | VladN | vnf8c | 0.7415 | 0.8692 | 0.9012 | 0.8373 |
| 16 | VADIS | Bagging | 0.7474 | 0.8631 | 0.8994 | 0.8366 |
| 17 | brendano | random forests (res11) | 0.7468 | 0.8627 | 0.9003 | 0.8366 |
| 18 | commendo | 1 before noon | 0.7381 | 0.8693 | 0.8988 | 0.8354 |
| 19 | FEG CTeam | Boosting | 0.7389 | 0.8616 | 0.9011 | 0.8338 |
| 20 | Vadis Team 2 | Best final | 0.7442 | 0.8568 | 0.8996 | 0.8335 |
| 21 | National Taiwan University, Computer Science and Information Engineering | all2 | 0.7428 | 0.8679 | 0.8890 | 0.8332 |
| 22 | Kranf | TIM | 0.7463 | 0.8478 | 0.8980 | 0.8307 |
| 23 | Neo Metrics | final2 | 0.7454 | 0.8449 | 0.8994 | 0.8299 |
| 24 | ooo | 10-3 | 0.7427 | 0.8520 | 0.8920 | 0.8289 |
| 25 | TonyM | mymethod5 | 0.7397 | 0.8481 | 0.8988 | 0.8289 |
| 26 | AIIALAB | ensemble | 0.7413 | 0.8458 | 0.8969 | 0.8280 |
| 27 | Uni Melb | hfinal | 0.7087 | 0.8669 | 0.8996 | 0.8251 |
| 28 | Christian Colot | My GoldMiner | 0.7183 | 0.8577 | 0.8958 | 0.8240 |
| 29 | Céline Theeuws | final | 0.7346 | 0.8476 | 0.8835 | 0.8219 |
| 30 | m&m | final test | 0.7218 | 0.8423 | 0.8924 | 0.8189 |
| 31 | Predictive Analytics | Logistic | 0.7131 | 0.8336 | 0.8917 | 0.8128 |
| 32 | DKW | NN / Logistic Regression on Laptop | 0.6980 | 0.8449 | 0.8928 | 0.8119 |
| 33 | NICAL | Dys | 0.7108 | 0.8461 | 0.8707 | 0.8092 |
| 34 | UW | eq+uneq | 0.6804 | 0.8531 | 0.8815 | 0.8050 |
| 35 | Prem Swaroop | thmdkd4 | 0.6972 | 0.8384 | 0.8794 | 0.8050 |
| 36 | Dr. Bunsen Honeydew | submission #004 | 0.7048 | 0.8235 | 0.8760 | 0.8015 |
| 37 | dodio | L2 | 0.7179 | 0.8474 | 0.8356 | 0.8003 |
| 38 | FEG D TEAM | mix2 | 0.6997 | 0.8139 | 0.8824 | 0.7987 |
| 39 | minos | rdf | 0.6828 | 0.8233 | 0.8698 | 0.7920 |
| 40 | M | Release1 | 0.7289 | |||