Overview of PAKDD competition 2007.pdf

International Journal of Data Warehousing Mining, 4(2), 1-8, April-June 2008 1 Copyright ? 2008, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. AbstrAct The PAKDD Competition 2007 involved the problem of predicting customers’ propensity to take up a home loan when a collection of data from credit card users are provided. It is rather difficult to address the problem because 1) the data set is extremely imbalanced; 2) the features are mixture types; and 3) there are many missing values. This article gives an overview on the competition, mainly consisting of three parts: 1) The background of the database and some statistical results of participants are introduced; 2) An analysis from the viewpoint of data preparation, resampling/reweighting and ensemble learning employed by different participants is given; and 3) Finally, some business insights are highlighted. Keywords: business insights; predicting Overview of PAKDD competition 2007 Junping Zhang, Fudan University, China Guo-Zheng Li, Shanghai University, China INtrODUctION The PAKDD Competition 2007 is about find- ing better solutions for a cross-selling busi- ness problem donated by a consumer finance company. In one of their markets, the company currently has a credit card customer base and a housing loan (mortgage) customer base with few customers overlapping between the two. The company would like to make use of this opportunity to cross-sell home loans to its credit card customers. In the problem setting of the competition, a modeling database of 40,700 customers with 40 modeling features plus a binary target feature is provided to the participants. This is a sample of customers who opened a new credit card with the company within a specific 2-year period, while they did not have an existing home loan with the company. In the database, 700 cases that the customer opened a home loan with the company within 12 mo