Neural networks and disputed authorship: new challenges

The paper examines a new application of neural networks to a field of statistical linguistics where we believe they have a major role to play, that of determining disputed authorship. Problems involving disputed authorship arise in many practical situations, from the detection of plagiarism to legal issues such as disputed confessions (D. Canter, 1992; B. Niblett and J. Boreham, 1976). There is a strong need to bridge the gap between artificial intelligence and statistics in this area