Event Extraction for Legal Case Building and Reasoning

Abstract

We are interested in developing tools to support the activities of lawyers in corporate litigation. In current applications, information such as characters that have played a significant role in a case, events in which they have participated, people they have been in contact, etc., have to be manually identified. There is little in the way of support to help them identify the relevant information in the first place. In this paper, we describe an approach to semiautomatically extracting such information from the collection of documents the lawyers are searching. Our approach is based on Natural Language Processing techniques and it enables the use of entity related information corresponding to the relations among the key players of a case, extracted in the form of events.

A Knowledge-Based System to Support Legal Case Construction

“…In order to describe the data to be extracted for supporting case building activities we have defined a Knowledge Model with three different layers (Lagos et al, 2010). Figure 5 shows a fragment of the KM' hierarchy.…”

“…using coreference). The various components of the NLP subsystem have been described in more detail in (Lagos et al, 2010).…”

“…The XIP Parser (Ait-Mokhtar et al, 2002) has been used for this task, as described in (Lagos et al, 2010). The NLP system outputs a collection of RDF files containing the semantic information extracted from the documents.…”

“…The primary goal of the searching and browsing facilities offered in current litigation tools is to find relevant documents, often using keyword/boolean based search techniques. Although this has proved to be relatively useful in the e-discovery phase, during case construction the emphasis shifts from finding documents to finding entities and actionable information derived from these entities (Noel and Azemard, 2008;Sheth et al, 2002;Lagos et al, 2010). This kind of search is an important part of the lawyers' work and tools currently on the market allow users to store information on relevant characters and events.…”

“…Previous work on some components of the system has already been described in Lagos et al, 2010). In this paper we illustrate the overall design of the system and a first prototype for it based on semantic technologies that implement parts of the design.…”