Mining multi-dimensional data for decision support

Personal bankruptcy is an increasingly common yet little understood phenome non. Attempts to predict bankruptcy have involved the application of data m ining techniques to credit card data. This is a difficult problem, since cr edit card data is multi-dimensional, consisting of monthly account records and daily transaction records. In this paper, we describe a two-stage appro ach that combines decision trees and neural networks to predict personal ba nkruptcy using credit card data. (C) 1999 Elsevier Science B.V. All rights reserved.