Developing computational models of discretion to build legal knowledge based systems

  • Authors:

  • Yaakov HaCohen Kerner;Uri Schild;John Zeleznikow

  • Affiliations:

  • Jerusalem College of Technology, 21 Havaad Haleumi St., P.O.B. 16031, 91160 Jerusalem;Department of Mathematics and Computer Science, Bar Ilan University, Ramat Gan, Israel, 52900;Applied Computing Research Institute, La Trobe University, Bundoora, Victoria, Australia, 3083 and Department of Mathematics and Computer Science, Bar Ilan University, Ramat Gan, Israel, 52900

  • Venue:
  • ICAIL '99 Proceedings of the 7th international conference on Artificial intelligence and law
  • Year:
  • 1999

Quantified Score

Hi-index 0.00

Visualization

Abstract

Few legal knowledge based systems have been constructed which provide numerical advice. None have been built in discretionary domains. Our research, directed towards the domains of sentencing and family law property division has lead to the development of three distinct forms of judicial discretion.To model these different discretionary domains we use diverse artificial intelligence tools including case-based reasoning and knowledge discovery from databases. We carry out a detailed comparison of two discretionary legal knowledge based systems. Judge's Apprentice is a case-based reasoner which recommends ranges of sentences for convicted Israeli rapists and robbers. SplitUp uses Knowledge Discovery from Databases to learn what percentage of marital property the partners to a divorce in Australia will receive. The systems are compared with regard to reasoning, explanation, evaluation and coping with conflicting cases.