Selection of radio pulsar candidates using artificial neural networks
Radio pulsar surveys are producing many more pulsar candidates than can be inspected by human experts in a practical length of time. Here we present a technique to automatically identify credible pulsar candidates from pulsar surveys using an artificial neural network. The technique has been applied to candidates from a recent re-analysis of the Parkes multi-beam pulsar survey resulting in the discovery of a previously unidentified pulsar.