V. P. Pethe, C. P. Rippey, and L. V. Kale. A specialized expert system for judicial decision support. In Proceedings of the Second International Conference on Artificial Intelligence and Law, pages 190--194, Vancouver, B.C., June 13-16 1989.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....justifications to documents is necessary. One approach to creating documents from justification structures is described in [Bra91a] 4 Related Research A prototype issue oriented judicial document assembly program, JEDA, was developed by Vishwas Pethe, Judge Charles P. Rippey, and L. V. Kale [PRK89]. JEDA uses PROLOG s depth first, satisficing search strategy and performs incremental mapping from justifications to text. JEDA uses mixed initiative data entry: uncontested facts (e.g. names the parties, docket number) and allegations concerning a case (e.g. medical reports, test results) are ....

....predicates occurring as subgoals. Judge Rippey, an Administrative Law Judge at the U.S. Department of Labor who has used JEDA since 1987, reports that JEDA reduces the time necessary to write decisions on claims for benefits under the federal Black Lung Benefits Act by a factor of at least two [PRK89]. Unfortunately, JEDA has not been accepted by other ALJ s on Judge Rippey s division [Rip91] A key limitation of JEDA is its inability to accommodate changes in the law or to apply to domains other than Black Lung Benefits Act cases. JEDA s knowledge of statutory rules, precedents, and ....

V. P. Pethe, C. P. Rippey, and L. V. Kale. A specialized expert system for judicial decision support. In Proceedings of the Second International Conference on Artificial Intelligence and Law, pages 190--194, Vancouver, B.C., June 13-16 1989.

.... argumentative texts in terms of the concepts of warrant, ground, conclusion, backing, and qualification [Tou58] Toulmin s model has been used for explanation generation [ZS95] document drafting in Plaid [BCS95] and analysis [FF96] Goal trees were also the basis of the document models of Jeda [PRK89] and LawClerk [Bra93] Performative documents, such as orders, judgments, wills, and contracts, and argumentative documents, such as briefs, generally consist in part of text segments clearly connected to specific legal predicates. Thus, the discourse structure needed for NLG techniques is ....

V. P. Pethe, C. P. Rippey, and L. V. Kale. A specialized expert system for judicial decision support. In Proceedings of the Second International Conference on Artificial Intelligence and Law, pages 190--194, Vancouver, B.C., June 13-16 1989.

....according to their role in the argument (claim, rebuttal, support, qualification, etc. This structure is then pruned to exclude premises that should be implicit in the final presentation and organized into a structure, including linking text, based on a high level rhetorical template. JEDA (Pethe et al. 1989) and LawClerk (Branting 1993) discussed in more detail below, used a declarative representation of legal rules to organize text templates around a goal tree generated during subgoaling. Argument theory provides a discourse model for documents whose purpose is to persuade, justify, establish, or ....

....corresponding to variations in the ways that administrative rules can apply to possible sets of facts. Thus, an argument based discourse model of administrative decisions is best suited to such cases. JEDA and LawClerk are argument based systems for drafting routine administrative decisions. JEDA (Pethe et al. 1989) is a system to assist administrative law judges in drafting decisions on claims for benefits under the federal Black Lung Benefits Act. JEDA uses mixed initiative data entry: uncontested facts (e.g. the names of the parties, the docket number) and allegations concerning a case (e.g. medical ....

V. P. Pethe, C. P. Rippey, and L. V. Kale, A Specialized Expert System for Judicial Decision Support' (1989) Proceedings of the Second International Conference on Artificial Intelligence and Law (ICAIL-89) Vancouver, B.C., pages 190-194.

....There are several possible sources of the relevant case facts. One approach is a conventional backward chaining rule based system, which would chain through the applicable jurisdictional rules querying the user when necessary. Interactions under this approach would be system initiated. JEDA [Pethe et al. 1989] and Law Clerk [Branting, 1993a] illustrate system initiated acquisition of case facts for document drafting. An alternative, user initiated, approach would use an electronic form with entries corresponding to possible case values. A third alternative is a mixed initiative approach under which ....

....is essential for the development of tools that embody the expertise of legal drafting experts [Gordon, 1989, Lauritsen, 1993] Several recent systems have used declarative representations of legal rules but only partially declarative representations of rhetorical structure. For example, JEDA [Pethe et al. 1989] used a declarative representation of legal rules, but mediated document construction entirely through procedural rules. Similarly, Law Clerk [Branting, 1993a] used an explicit representation of legal rules and a simple record structure for administrative law decisions. Law Clerk instantiated and ....

Pethe, V. P., Rippey, C. P., and Kale, L. V. (1989). A specialized expert system for judicial decision support. In Proceedings of the Second International Conference on Artificial Intelligence and Law, pages 190--194, Vancouver, B.C.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC