<?xml version="1.0" standalone="yes"?>
<Paper uid="W03-0505">
  <Title>Summarising Legal Texts: Sentential Tense and Argumentative Roles</Title>
  <Section position="2" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> Law reports form the most important part of a lawyer's or law student's reading matter. These reports are records of the proceedings of a court and their importance derives from the role that precedents play in English law. They are used as evidence for or against a particular line of legal reasoning. In order to make judgments accessible and to enable rapid scrutiny of their relevance, they are usually summarised by legal experts. These summaries vary according to target audience (e.g. students, solicitors).</Paragraph>
    <Paragraph position="1"> Manual summarisation can be considered as a form of information selection using an unconstrained vocabulary with no artificial linguistic limitations. Automatic summarisation, on the other hand, has postponed the goal of text generation de novo and currently focuses largely on the retrieval of relevant sections of the original text. The retrieved sections can then be used as the basis of summaries with the aid of suitable smoothing phrases.</Paragraph>
    <Paragraph position="2"> In the SUM project we are investigating methods for generating flexible summaries of documents in the legal domain. Our methodology builds and extends the Teufel and Moens (Teufel and Moens, 2002) approach to automatic summarisation. The work we report on in this paper deals with judgments from the judicial branch of the House of Lords. We have completed a preliminary study using a small sample of judgment documents. We have hand-annotated the sentences in these documents and performed automatic linguistic processing in order to study the link between the argumentative role and linguistic features of a sentence. Our primary focus is on correlations between sentence type and verb group properties (e.g. tense, aspect). To this end, we have used state-of-the-art NLP techniques to distinguish main and subordinate clauses and to find the tense and aspect features of the main verb in each sentence. In this paper we report on our NLP techniques and on the findings of our study.</Paragraph>
    <Paragraph position="3"> We discuss the implications for the summarisation system that we are in the process of developing.</Paragraph>
    <Paragraph position="4"> Section 2 provides a brief background to our work including an overview of the Teufel and Moens approach and a description of the annotation scheme we have developed for the House of Lords judgments. Section 3 provides an overview of the tools and techniques we have used in the automatic linguistic processing of the judgments. Our processing paradigm is XML-based and we use specialist XML-aware tools to perform tasks such as tokenisation, part-of-speech tagging and chunking-these are described in Section 3.1. Our primary interest is tense information about individual sentences and to compute this we need to distinguish main from subordinate clauses in order to identify the main verb group.</Paragraph>
    <Paragraph position="5"> We report on our statistically-based approach to this task in Section 3.2. In Section 3.3 we present the results of our preliminary evaluations based on the small corpus of hand-annotated judgments. Finally, in Section 4 we draw some conclusions and outline future work.</Paragraph>
  </Section>
class="xml-element"></Paper>