Research on prediction model and fuzzy rules extraction in Maersk´s customer management

The advent of data mining has contributed significantly to the field of customer management. The payment center in Maersk finds that it´s difficult to charge customers for the balance due. Thus the research on the application of artificial neural networks in classification of data mining will be conducted in this paper. Since the knowledge captured by neural networks is represented at a sub-symbolic level in terms of connections and weights, the paper aims at not only designing an artificial neural network prediction model for the time of delay payment but also extracting fuzzy production rules, a humanly understandable representation for neural networks. The simple rules with high predictive accuracy rules really help the enterprise to solve customer management to a certain degree.