Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!van-bc!ubc-cs!alberta!herald.usask.ca!ccu.umanitoba.ca!bison!sys6626!inqmind!system From: system@inqmind.bison.mb.ca ( SYSOP ENTITY ) Newsgroups: comp.ai.neural-nets Subject: Re: Neural Network applications to Law Message-ID: Date: Sat, 13 Apr 91 18:13:44 CDT References: <8754@skye.cs.ed.ac.uk> Organization: The Inquiring Mind 1 204 488-1607 I am interested in the application of Neural Networks to Law also. I am a Forensic Scientist and am currently exploring (albeit tentatively) potential applications of Neural Networks to decision making in Forensic Science. Naturally many decisions in Forensic Science are extendable to the larger context of decision making in law. I understand that at this time there is a trend in the UK to employ Bayes' Theorem in decision making as it applies to law. In the course of my own attempts to understand decision making in Forensic Science I have considered Bayes' Theorem but I do not find it to be good enough. Neural Networks can be configured to make decisions that are Bayesian or that employ fuzzy logic, but these are only limiting cases for specific designed Neural Nets. In fact Neural Networks in general are capable of much more complex non-linear correlations than can be achieved by conventional statistical approaches. In this sense they have the capability to assist us in modelling decision making and also to teach u s more about how we in fact reach decisions. I am extremely interested in this area so any dialogue either in this newsgroup or privately is welcome. Tod Christianson ______________________________________________________________________________ Tod Christianson "Sysop Entity" at: 478 Beaverbrook St. Wpg., Mb., Canada THE INQUIRING MIND R3N-1N3 204-488-1607 USENET:system%inqmind.bison.mb.ca@niven.cc.umanitoba.ca ------------------------------------------------------------------------------