Book contents
- Frontmatter
- Dedication
- Contents
- List of illustrations
- List of tables
- Acknowledgments
- PART I COMPUTATIONAL MODELS OF LEGAL REASONING
- PART II LEGAL TEXT ANALYTICS
- 6 Representing Legal Concepts in Ontologies and Type Systems
- 7 Making Legal Information Retrieval Smarter
- 8 Machine Learning with Legal Texts
- 9 Extracting Information from Statutory and Regulatory Texts
- 10 Extracting Argument-Related Information from Legal Case Texts
- PART III CONNECTING COMPUTATIONAL REASONING MODELS AND LEGAL TEXTS
- Glossary
- Bibliography
- Index
6 - Representing Legal Concepts in Ontologies and Type Systems
from PART II - LEGAL TEXT ANALYTICS
Published online by Cambridge University Press: 13 July 2017
- Frontmatter
- Dedication
- Contents
- List of illustrations
- List of tables
- Acknowledgments
- PART I COMPUTATIONAL MODELS OF LEGAL REASONING
- PART II LEGAL TEXT ANALYTICS
- 6 Representing Legal Concepts in Ontologies and Type Systems
- 7 Making Legal Information Retrieval Smarter
- 8 Machine Learning with Legal Texts
- 9 Extracting Information from Statutory and Regulatory Texts
- 10 Extracting Argument-Related Information from Legal Case Texts
- PART III CONNECTING COMPUTATIONAL REASONING MODELS AND LEGAL TEXTS
- Glossary
- Bibliography
- Index
Summary
INTRODUCTION
As Part I indicates, knowledge representation has been a key focus of AI & Law research and a key challenge for implementing systems robust enough to serve as real-world legal practice tools.
Ontologies help to meet that challenge. An ontology specifies the fundamental types of things or concepts that exist for purposes of a system and sets out the relations among them.
After introducing some basic information about ontologies, this chapter surveys some historically influential legal ontologies and explains some modern techniques for constructing ontologies semiautomatically. It then turns to ontological supports for statutory reasoning and for legal argumentation. In connection with the latter, an extended example illustrates ontological supports for making arguments with a small collection of cases.
Finally, the chapter introduces a specialized kind of ontology, “type systems,” which are a basic text analytic tool. Type systems support automatically marking-up or annotating legal texts semantically in terms of concepts and their relations. They will play key roles in conceptual legal information retrieval and in cognitive computing.
This chapter addresses the following questions. What is a legal ontology and how are legal ontologies used? What is semantic annotation? What are text annotation pipelines and what role does a type system play? What is a UIMA framework? How are legal ontologies and UIMA type systems constructed? How can developers of legal type systems take advantage of existing legal ontologies and of ontologies already developed for medicine, or for other real world domains that have legal implications?
ONTOLOGY BASICS
Despite its metaphysical connotations, the term “ontology” is not quite so imposing in the context of computational models. An ontology is an “explicit, formal, and general specification of a conceptualization of the properties of and relations between objects in a given domain” (Wyner, 2008). In other words, ontologies make concepts in a domain explicit so that a program can reason with them.
For example, Figure 6.1 shows a simple ontology for the legal concept of contract formation. This kind of an ontology might have been useful for Ann Gardner's first-year contracts problem analyzer (Section 1.4.2) and captures concepts and relations described in Gardner (1987, pp. 121–3) such as “Manifestation of mutual consent” and “Acceptance by verbal promise.”
- Type
- Chapter
- Information
- Artificial Intelligence and Legal AnalyticsNew Tools for Law Practice in the Digital Age, pp. 171 - 209Publisher: Cambridge University PressPrint publication year: 2017