Predicting Customer Behavior using Naive Bayes and Maximum Entropy - Winning the Data-Mining-Cup 2004 -
Arne Mauser, Ilja Bezrukov, Thomas Deselaers, Daniel Keysers · 2004
In this work we describe combinations of classifiers using Naive Bayes, Maximum Entropy, Neural Networks and Logistic Regression for classification of cus- tomer records. Performance of these approaches is confirmed by the 1st, 3rd, and 5th rank in the Data-Mining-Cup 2004.