Towards a Legal Reasoning System based on Description Logics: A Position Paper

Legal reasoning is concerned with various kinds of legal concepts that relate to another concepts used in our daily life. Although those legal concepts are abstract in the sense that there always exist possibilities to interpret them in various ways, lawyers seem to understand them as real things at least in their "legal reality". For instance, Japanese Civil Codes include a notion of falsity that means a false declaration of intention. In spite of its abstractness, well-trained lawyers can judge that one legal act involves the falsity and that the others do not. From this point of view, it seems natural to consider that each legal act has its legal identity. Since the standard first order logic (FOL, for short) is founded on the assumption that every object treated within the logic has its identity, it seems natural to develop a way of representing the legal acts as first-order terms denoting individual objects. However, they are highly related each other at their conceptual level. For instance, a person fills the role of agent in a contract. Thus the two concepts "person" and "contract" are related by the role "agent". Recently in Japan, some researchers [12; 17] try to develop a legal ontology consisting legal concepts and conceptual relationships between them. The latter ones can be roles in description logics. So if we like to build a large legal knowledge base, we would have a description logic containing many legal conceptual terms linked by roles. Based on this, an efficient reasoner (RKB interpreter) would be designed as an extension of FOL prover, for both DL and FOL makes the same assumption that each object has its identity and every concept is a set of objects.

[1]  William W. Cohen,et al.  The learnability of description logics with equality constraints , 1994, Machine Learning.

[2]  Makoto Haraguchi,et al.  A Theory of Knowledge Revision using Multiple Generalization , 1996 .

[3]  Josh Tenenberg.  Abstracting First-Order Theories , 1990 .

[4]  Bernhard Nebel,et al.  Reasoning and Revision in Hybrid Representation Systems , 1990, Lecture Notes in Computer Science.

[5]  Dedre Gentner,et al.  Are Scientific Analogies Metaphors , 1981 .

[6]  D. Benjamin.  Change of Representation and Inductive Bias , 1989 .

[7]  Michael Kifer,et al.  Logical foundations of object-oriented and frame-based languages , 1995, JACM.

[8]  Patrick Henry Winston,et al.  Learning and reasoning by analogy , 1980, CACM.

[9]  Makoto Haraguchi,et al.  Towards a legal analogical reasoning system: knowledge representation and reasoning methods , 1993, ICAIL '93.

[10]  Makoto Haraguchi.  A Reasoning System for Legal Analogy , 1993, Machine Intelligence 14.

[11]  S. Wrobel.  Concept Formation and Knowledge Revision , 1994, Springer US.

[12]  Andreas Podelski,et al.  Towards a Meaning of LIFE , 1991, PLILP.

[13]  Makoto Haraguchi,et al.  Constructing Predicate Mappings for Goal-Dependent Abstraction , 1994, AII/ALT.

[14]  Andreas Podelski,et al.  Towards a Meaning of LIFE , 1991, J. Log. Program..

[15]  Hiroki Arimura,et al.  A Polynomial Time Algorithm for Finding Finite Unions of Tree Pattern Languages , 1991, Nonmonotonic and Inductive Logic.

[16]  Peter F. Patel-Schneider,et al.  A Semantics and Complete Algorithm for Subsumption in the CLASSIC Description Logic , 1993, J. Artif. Intell. Res..