Operational experience with a neural network in the detection of explosives in checked airline luggage

An artificial neural network (NN) has been implemented in the explosives detection systems fielded at various airports. Tests of the online performance of the NN confirmed its superiority over standard statistical techniques; and the NN was installed as the decision algorithm. Analysis of the mass of data being produced is still underway, but preliminary conclusions are presented. Results are given for a NN with a single hidden layer. Some experiments were also done with a second hidden layer. In general, many more iterations through the test set were required to get these networks to converge to reasonable performance