Talk at BILETA 2011

I’m giving a talk tomorrow, April 11 2011, at BILETA, the annual conference of the British & Irish Law, Education and Technology Association at Manchester Metropolitan University School of Law. My collaborators are Wim Peters (University of Sheffield) and Fiona Beveridge (University of Liverpool).
The abstract and slides are below:
Web-based Software Tools to Support Students’ Empirical Study of the Law
Adam Wyner (University of Liverpool, Computer Science), Wim Peters (University of Sheffield, Computer Science), and Fiona Beveridge (University of Liverpool, Law School)
The paper investigates and proposes tools to support students in empirically investigating legal cases using text analytic software. Web-based tools can be used to engage and leverage the collective skills and ambitions of law students to crowd-source the development of legal resource materials. Law school students must develop skills in close textual analysis of legal source material such as legal cases. To use source material such as case decisions to reason about how precedents apply in case-based reasoning, law students must learn to identify a range of elements in legal cases, for example, parties, jurisdiction, material facts, legislative and case citations, cause of action, ratio decideni, and others. Moreover, students should be able to address complex queries to a case or a case base (a corpus of cases) in order to answer questions of particular legal interest; for example, about relationships between a judge, parties, cause of action, and ratio. Currently students either simply rely on their own analytic abilities to read a case or find answers to questions; legal search tools (e.g. Lexis-Nexis) provide search support, but are restricted to a limited number of coarse-grained parameters and cannot search for deep, particular semantic relationships in the text. To enable automated support of queries of the corpus, and so enable deep empirical research on cases, it is essential to have a corpus of legal cases which are annotated with machine readable (XML) tags that signal the semantic properties of passages of text. To create such a corpus requires a tool to annotate the text. Such a tool would reinforce students’ examination of the source document. The paper describes recent developments of tools using Semantic Web technologies, text analysis, and web-based annotation support. With the text analysis software, General Architecture for Text Engineering (GATE), which is customised for legal applications, law students can annotate legal cases for a fine-grained range of legally relevant concepts and linguistic relations; they can also use GATE to write grammars and automatically annotate the text. Using GATE TeamWare, an online text annotation tool that automatically evaluates interannotator agreement, students can collaboratively analyse and agree on a gold standard corpus of legal cases. The corpus can be automatically indexed using Lucene, thereby allowing fast results to complex queries over any string or annotation used.
The slides of the talk are here