Beyond rule based reasoning - the meaning and use of cases
Data(s) |
1995
|
---|---|
Resumo |
Commercial legal expert systems are invariably rule based. Such systems are poor at dealing with open texture and the argumentation inherent in law. To overcome these problems we suggest supplementing rule based legal expert systems with case based reasoning or neural networks. Both case based reasoners and neural networks use cases-but in very different ways. We discuss these differences at length. In particular we examine the role of explanation in existing expert systems methodologies. Because neural networks provide poor explanation facilities, we consider the use of Toulmin argument structures to support explanation (S. Toulmin, 1958). We illustrate our ideas with regard to a number of systems built by the authors |
Identificador | |
Publicador |
IEEE |
Relação |
DOI:10.1109/CAIA.1995.378809 Zeleznikow, John, Stranieri, Andrew, & Hunter, Dan (1995) Beyond rule based reasoning - the meaning and use of cases. In Proceedings of 11th Conference on Artificial Intelligence for Applications, IEEE, Los Angeles, Calif., pp. 292-298. |
Direitos |
Copyright 1995 IEEE |
Fonte |
Faculty of Law; School of Law |
Tipo |
Conference Paper |