A data mining approach to neural network training Available to Purchase

Neural networks are widely used in model building. Applications include predicting bankruptcy, and assessing the relationship quality between a customer and a salesperson. This paper utilizes data from an Internet users survey to investigate Internet usage for personal and not for work or school related needs, and the impact of pattern presentation order and frequency on the learning accuracy of the network. The data were sorted using the association‐rule mining approach. The software used is BrainMaker from California Scientific Software.