IBaI Institute |
Past events |
2nd Industrial Conference on Data Mining ICDM´2002
13. 06. - 15. 06. 2002 Leipzig/Germany
Mining Customer Profiles with Decision Trees
Dr. P. Perner, J. Perner, B. Müller, Institute of Computer Vision and Applied Computer Sciences, Leipzig/Germany
The diversity of products and services and the increasingly sharp competition make it necessary to identify the potential customers as exactly as possible, to tune the products to the customer needs and to do the advertising precisely aimed at this lientele, whereby scattering losses and costs should be kept as low as possible. This is not possible without detailed knowledge on the potential clientele.
In each enterprise a great number of data accumulates from which with intelligent evaluation processes knowledge on customer groups can be gained and their purchase behavior predicted.
In the project MININT (sponsored by the Germany Ministry of Economy BMWI) we are devoloping new methods for data mining on marketing data. In the talk we will show by means of a real example how with different decision tree methods customer profiles can be obtained from data and be used for a marketing action aimed at a special clientele.
We will explain the criteria that have to be considered when applying the decision tree methods and show which effects different influences have on the result and which results can be obtained with binary and n-ary decision trees.